File: Samba3-HOWTO.xml

package info (click to toggle)
samba 2%3A3.2.5-4lenny15
links: PTS
area: main
in suites: lenny
size: 124,704 kB
ctags: 69,181
sloc: ansic: 564,153; xml: 219,271; sh: 11,039; perl: 4,458; makefile: 4,301; python: 1,975; cpp: 1,224; exp: 1,147; yacc: 881; awk: 557; csh: 58; sed: 45
file content (46455 lines) | stat: -rw-r--r-- 2,277,315 bytes
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE book PUBLIC "-//OASIS//DTD DocBook XML V4.2//EN" "http://www.oasis-open.org/docbook/xml/4.2/docbookx.dtd">
<book xmlns:xi="http://www.w3.org/2003/XInclude" id="Samba-HOWTO-Collection">
<title>The Official Samba 3.2.x HOWTO and Reference Guide</title>

<bookinfo>
	<authorgroup>
	<editor>
<firstname>Jelmer</firstname><othername>R.</othername><surname>Vernooij</surname><othername>R.</othername>
<affiliation>
	<orgname>The Samba Team</orgname>
	<address format="linespecific"><email>jelmer@samba.org</email></address>
</affiliation></editor>
	<editor>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></editor>
	<editor>
<firstname>Gerald</firstname><surname>Carter</surname><othername>(Jerry)</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jerry@samba.org</email></address>
</affiliation></editor>
	</authorgroup>
	<pubdate><?latex \today ?></pubdate>
</bookinfo>

	<?latex \setcounter{page}{5} ?>

	<preface lang="en-US">
<title>About the Cover Artwork</title>

	<para>
	The cover artwork of this book continues the freedom theme of the first edition of <quote>The Official Samba-3
	HOWTO and Reference Guide</quote>.  We may look back upon the past to question the motives of those who have
	gone before us. Seldom do we realise that the past owes us no answer, and despite what we may think of the
	actions of those who have travelled lifes' road before us, we must feel a sense of pride and gratitude for
	those who, in the past, have protected our liberties.
	</para>

	<para>
	Developments in information technology continue to move at an alarming pace. Human nature causes us
	to adopt and embrace new developments that appear to answer the needs of the moment, but that can entrap
	us at a future date. There are many examples in the short history of information technology. MS-DOS was
	seen as a tool that liberated users from the tyrany of large computer system operating costs, and that
	made possible the rapid progres we are beneficiaries of today. Yet today we are inclined to look back with
	disdain on MS-DOS as an obsolete and constraining technology that belongs are an era that is best
	forgotten.
	</para>

	<para>
	The embrace of Windows networking, Windows NT4, and MS Active Directory in more recent times, may seem
	modern and progressive today, but sooner or later something better will replace them. The current
	preoccupation with extended identity management solutions and with directories is not unexpected.
	The day will come that these too will be evaluated, and what may seem refreshing and powerful may
	be better recogized as the chilly winds of the night. To argue against progress is unthinkable,
	no matter what may lie ahead.
	</para>

	<para>
	The development of Samba is moving forwards. The changes since Samba 3.0.0 are amazing, yet many
	users would like to see more and faster progress. The benefits of recent developments can be realized
	quickly, but documentation is necessary to unlock the pandoras' box. It is our hope that this book
	will help the network administrator to rapidly deploy the new features with minimum effort. As you
	deploy and gain mileage from the new enablement, take the time to think through what may lie ahead.
	Above all, take stock of the freedom of choice that Samba provides in your world, and enjoy the new 
	potential for seamless interoperability.
	</para>

</preface>

	<?latex \cleardoublepage ?>

	<preface>
  <title>Attribution</title>
  <para><link linkend="install">How to Install and Test SAMBA</link>
	<itemizedlist>	<listitem><para>Andrew Tridgell<ulink noescape="1" url="mailto:tridge@samba.org"/></para></listitem>
	<listitem><para>Jelmer R.  Vernooij<ulink noescape="1" url="mailto:jelmer@samba.org"/></para></listitem>
	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	<listitem><para>Karl Auer<ulink noescape="1" url="mailto:kauer@biplane.com.au"/></para></listitem>
	<listitem><para>Dan Shearer<ulink noescape="1" url="mailto:dan@samba.org"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="FastStart">Fast Start: Cure for Impatience</link>
	<itemizedlist>	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="ServerType">Server Types and Security Modes</link>
	<itemizedlist>	<listitem><para>Andrew Tridgell<ulink noescape="1" url="mailto:tridge@samba.org"/></para></listitem>
	<listitem><para>Jelmer R.  Vernooij<ulink noescape="1" url="mailto:jelmer@samba.org"/></para></listitem>
	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="samba-pdc">Domain Control</link>
	<itemizedlist>	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	<listitem><para>Gerald (Jerry)  Carter<ulink noescape="1" url="mailto:jerry@samba.org"/></para></listitem>
	<listitem><para>David Bannon<ulink noescape="1" url="mailto:dbannon@samba.org"/></para></listitem>
	<listitem><para>Guenther Deschner<ulink noescape="1" url="mailto:gd@suse.de"/> (LDAP updates) </para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="samba-bdc">Backup Domain Control</link>
	<itemizedlist>	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	<listitem><para>Volker Lendecke<ulink noescape="1" url="mailto:Volker.Lendecke@SerNet.DE"/></para></listitem>
	<listitem><para>Guenther Deschner<ulink noescape="1" url="mailto:gd@suse.de"/> (LDAP updates) </para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="domain-member">Domain Membership</link>
	<itemizedlist>	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	<listitem><para>Jeremy Allison<ulink noescape="1" url="mailto:jra@samba.org"/></para></listitem>
	<listitem><para>Gerald (Jerry)  Carter<ulink noescape="1" url="mailto:jerry@samba.org"/></para></listitem>
	<listitem><para>Andrew Tridgell<ulink noescape="1" url="mailto:tridge@samba.org"/></para></listitem>
	<listitem><para>Jelmer R.  Vernooij<ulink noescape="1" url="mailto:jelmer@samba.org"/></para></listitem>
	<listitem><para>Guenther Deschner<ulink noescape="1" url="mailto:gd@suse.de"/> (LDAP updates) </para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="StandAloneServer">Standalone Servers</link>
	<itemizedlist>	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="ClientConfig">MS Windows Network Configuration Guide</link>
	<itemizedlist>	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="ChangeNotes">Important and Critical Change Notes for the Samba 3.x Series</link>
	<itemizedlist>	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	<listitem><para>Gerald (Jerry)  Carter<ulink noescape="1" url="mailto:jerry@samba.org"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="NetworkBrowsing">Network Browsing</link>
	<itemizedlist>	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	<listitem><para>Jelmer R.  Vernooij<ulink noescape="1" url="mailto:jelmer@samba.org"/></para></listitem>
	<listitem><para>Jonathan Johnson<ulink noescape="1" url="mailto:jon@sutinen.com"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="passdb">Account Information Databases</link>
	<itemizedlist>	<listitem><para>Jelmer R.  Vernooij<ulink noescape="1" url="mailto:jelmer@samba.org"/></para></listitem>
	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	<listitem><para>Gerald (Jerry)  Carter<ulink noescape="1" url="mailto:jerry@samba.org"/></para></listitem>
	<listitem><para>Jeremy Allison<ulink noescape="1" url="mailto:jra@samba.org"/></para></listitem>
	<listitem><para>Guenther Deschner<ulink noescape="1" url="mailto:gd@suse.de"/> (LDAP updates) </para></listitem>
	<listitem><para>Olivier (lem) Lemaire<ulink noescape="1" url="mailto:olem@IDEALX.org"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="groupmapping">Group Mapping: MS Windows and UNIX</link>
	<itemizedlist>	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	<listitem><para>Jean François Micouleau</para></listitem>
	<listitem><para>Gerald (Jerry)  Carter<ulink noescape="1" url="mailto:jerry@samba.org"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="NetCommand">Remote and Local Management: The Net Command</link>
	<itemizedlist>	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	<listitem><para>Volker Lendecke<ulink noescape="1" url="mailto:Volker.Lendecke@SerNet.DE"/></para></listitem>
	<listitem><para>Guenther Deschner<ulink noescape="1" url="mailto:gd@suse.de"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="idmapper">Identity Mapping (IDMAP)</link>
	<itemizedlist>	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="rights">User Rights and Privileges</link>
	<itemizedlist>	<listitem><para>Gerald (Jerry)  Carter<ulink noescape="1" url="mailto:jerry@samba.org"/></para></listitem>
	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="AccessControls">File, Directory, and Share Access Controls</link>
	<itemizedlist>	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	<listitem><para>Jeremy Allison<ulink noescape="1" url="mailto:jra@samba.org"/></para></listitem>
	<listitem><para>Jelmer R.  Vernooij<ulink noescape="1" url="mailto:jelmer@samba.org"/> (drawing) </para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="locking">File and Record Locking</link>
	<itemizedlist>	<listitem><para>Jeremy Allison<ulink noescape="1" url="mailto:jra@samba.org"/></para></listitem>
	<listitem><para>Jelmer R.  Vernooij<ulink noescape="1" url="mailto:jelmer@samba.org"/></para></listitem>
	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	<listitem><para>Eric Roseme<ulink noescape="1" url="mailto:eric.roseme@hp.com"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="securing-samba">Securing Samba</link>
	<itemizedlist>	<listitem><para>Andrew Tridgell<ulink noescape="1" url="mailto:tridge@samba.org"/></para></listitem>
	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="InterdomainTrusts">Interdomain Trust Relationships</link>
	<itemizedlist>	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	<listitem><para>Rafal Szczesniak<ulink noescape="1" url="mailto:mimir@samba.org"/></para></listitem>
	<listitem><para>Jelmer R.  Vernooij<ulink noescape="1" url="mailto:jelmer@samba.org"/> (drawing) </para></listitem>
	<listitem><para>Stephen Langasek<ulink noescape="1" url="mailto:vorlon@netexpress.net"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="msdfs">Hosting a Microsoft Distributed File System Tree</link>
	<itemizedlist>	<listitem><para>Shirish Kalele<ulink noescape="1" url="mailto:samba@samba.org"/></para></listitem>
	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="classicalprinting">Classical Printing Support</link>
	<itemizedlist>	<listitem><para>Kurt Pfeifle<ulink noescape="1" url="mailto:kpfeifle@danka.de"/></para></listitem>
	<listitem><para>Gerald (Jerry)  Carter<ulink noescape="1" url="mailto:jerry@samba.org"/></para></listitem>
	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="CUPS-printing">CUPS Printing Support</link>
	<itemizedlist>	<listitem><para>Kurt Pfeifle<ulink noescape="1" url="mailto:kpfeifle@danka.de"/></para></listitem>
	<listitem><para>Ciprian Vizitiu<ulink noescape="1" url="mailto:CVizitiu@gbif.org"/> (drawings) </para></listitem>
	<listitem><para>Jelmer R.  Vernooij<ulink noescape="1" url="mailto:jelmer@samba.org"/> (drawings) </para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="VFS">Stackable VFS modules</link>
	<itemizedlist>	<listitem><para>Jelmer R.  Vernooij<ulink noescape="1" url="mailto:jelmer@samba.org"/></para></listitem>
	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	<listitem><para>Tim Potter<ulink noescape="1" url="mailto:tpot@samba.org"/></para></listitem>
	<listitem><para>Simo Sorce (original vfs_skel README) </para></listitem>
	<listitem><para>Alexander Bokovoy (original vfs_netatalk docs) </para></listitem>
	<listitem><para>Stefan Metzmacher (Update for multiple modules) </para></listitem>
	<listitem><para>Ed Riddle (original shadow_copy docs) </para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="winbind">Winbind: Use of Domain Accounts</link>
	<itemizedlist>	<listitem><para>Tim Potter<ulink noescape="1" url="mailto:tpot@linuxcare.com.au"/></para></listitem>
	<listitem><para>Andrew Tridgell<ulink noescape="1" url="mailto:tridge@samba.org"/></para></listitem>
	<listitem><para>Naag Mummaneni<ulink noescape="1" url="mailto:getnag@rediffmail.com"/> (Notes for Solaris) </para></listitem>
	<listitem><para>John Trostel<ulink noescape="1" url="mailto:jtrostel@snapserver.com"/></para></listitem>
	<listitem><para>Jelmer R.  Vernooij<ulink noescape="1" url="mailto:jelmer@samba.org"/></para></listitem>
	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="AdvancedNetworkManagement">Advanced Network Management</link>
	<itemizedlist>	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="PolicyMgmt">System and Account Policies</link>
	<itemizedlist>	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="ProfileMgmt">Desktop Profile Management</link>
	<itemizedlist>	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="pam">PAM-Based Distributed Authentication</link>
	<itemizedlist>	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	<listitem><para>Stephen Langasek<ulink noescape="1" url="mailto:vorlon@netexpress.net"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="integrate-ms-networks">Integrating MS Windows Networks with Samba</link>
	<itemizedlist>	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="unicode">Unicode/Charsets</link>
	<itemizedlist>	<listitem><para>Jelmer R.  Vernooij<ulink noescape="1" url="mailto:jelmer@samba.org"/></para></listitem>
	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	<listitem><para>TAKAHASHI Motonobu<ulink noescape="1" url="mailto:monyo@home.monyo.com"/> (Japanese character support) </para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="Backup">Backup Techniques</link>
	<itemizedlist>	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="SambaHA">High Availability</link>
	<itemizedlist>	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	<listitem><para>Jeremy Allison<ulink noescape="1" url="mailto:jra@samba.org"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="largefile">Handling Large Directories</link>
	<itemizedlist>	<listitem><para>Jeremy Allison<ulink noescape="1" url="mailto:jra@samba.org"/></para></listitem>
	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="cfgsmarts">Advanced Configuration Techniques</link>
	<itemizedlist>	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="upgrading-to-3.0">Updating and Upgrading Samba</link>
	<itemizedlist>	<listitem><para>Jelmer R.  Vernooij<ulink noescape="1" url="mailto:jelmer@samba.org"/></para></listitem>
	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	<listitem><para>Gerald (Jerry)  Carter<ulink noescape="1" url="mailto:jerry@samba.org"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="NT4Migration">Migration from NT4 PDC to Samba-3 PDC</link>
	<itemizedlist>	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="SWAT">SWAT: The Samba Web Administration Tool</link>
	<itemizedlist>	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="diagnosis">The Samba Checklist</link>
	<itemizedlist>	<listitem><para>Andrew Tridgell<ulink noescape="1" url="mailto:tridge@samba.org"/></para></listitem>
	<listitem><para>Jelmer R.  Vernooij<ulink noescape="1" url="mailto:jelmer@samba.org"/></para></listitem>
	<listitem><para>Dan Shearer<ulink noescape="1" url="mailto:dan@samba.org"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="problems">Analyzing and Solving Samba Problems</link>
	<itemizedlist>	<listitem><para>Gerald (Jerry)  Carter<ulink noescape="1" url="mailto:jerry@samba.org"/></para></listitem>
	<listitem><para>Jelmer R.  Vernooij<ulink noescape="1" url="mailto:jelmer@samba.org"/></para></listitem>
	<listitem><para>David Bannon<ulink noescape="1" url="mailto:dbannon@samba.org"/></para></listitem>
	<listitem><para>Dan Shearer<ulink noescape="1" url="mailto:dan@samba.org"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="bugreport">Reporting Bugs</link>
	<itemizedlist>	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	<listitem><para>Jelmer R.  Vernooij<ulink noescape="1" url="mailto:jelmer@samba.org"/></para></listitem>
	<listitem><para>Andrew Tridgell<ulink noescape="1" url="mailto:tridge@samba.org"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="tdb">Managing TDB Files</link>
	<itemizedlist>	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="compiling">How to Compile Samba</link>
	<itemizedlist>	<listitem><para>Jelmer R.  Vernooij<ulink noescape="1" url="mailto:jelmer@samba.org"/></para></listitem>
	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	<listitem><para>Andrew Tridgell<ulink noescape="1" url="mailto:tridge@samba.org"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="Portability">Portability</link>
	<itemizedlist>	<listitem><para>Jelmer R.  Vernooij<ulink noescape="1" url="mailto:jelmer@samba.org"/></para></listitem>
	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="Other-Clients">Samba and Other CIFS Clients</link>
	<itemizedlist>	<listitem><para>Jelmer R.  Vernooij<ulink noescape="1" url="mailto:jelmer@samba.org"/></para></listitem>
	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	<listitem><para>Dan Shearer<ulink noescape="1" url="mailto:dan@samba.org"/></para></listitem>
	<listitem><para>Jim McDonough<ulink noescape="1" url="mailto:jmcd@us.ibm.com"/> (OS/2) </para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="speed">Samba Performance Tuning</link>
	<itemizedlist>	<listitem><para>Paul Cochrane<ulink noescape="1" url="mailto:paulc@dth.scot.nhs.uk"/></para></listitem>
	<listitem><para>Jelmer R.  Vernooij<ulink noescape="1" url="mailto:jelmer@samba.org"/></para></listitem>
	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="ch-ldap-tls">LDAP and Transport Layer Security</link>
	<itemizedlist>	<listitem><para>Gavin Henry<ulink noescape="1" url="mailto:ghenry@suretecsystems.com"/></para></listitem>
	</itemizedlist>
</para>
  <para><link linkend="DNSDHCP">DNS and DHCP Configuration Guide</link>
	<itemizedlist>	<listitem><para>John H.  Terpstra<ulink noescape="1" url="mailto:jht@samba.org"/></para></listitem>
	</itemizedlist>
</para>
</preface>

	<?latex \cleardoublepage ?>

<!-- Contents -->
	<toc/>

	<?latex \cleardoublepage ?>

	<?latex \listofexamples ?>

	<?latex \cleardoublepage ?>

	<lot/>

	<preface>
<title>Foreword</title>

<para>
When John first asked me to write an introductory piece for his latest book, I was somewhat mystified as to
why he chose me. A conversation with John provided some of the rationale, and he left it to me to fill in the
<emphasis>rest</emphasis> of the story. So, if you are willing to endure a little bit of background, I will
provide the part of the story that John wouldn't provide.
</para>

<para>
I am the Director of Corporate Standards at Sun Microsystems, and manage Sun's standards portfolio. Before
that, I was the Director of Standards at Netscape, which was when I met John. Before Sun, there was Digital
Equipment Corporation, also standards. I've written several books on standards, and tend to observe (and
occasionally help) the technical and business trends that drive standardization as a discipline. I tend to see
standardization as a management tool, not as a technical discipline and this is part of the rationale that
John provided.
</para>

<para>
The book that you have before you focuses on a particular standardized way of doing something hence, it is a
book about a standard. The most important thing to keep in mind about a standard is the rationale for its
creation. Standards are created not for technical reasons, not for business reasons, but for a deeper and much
more compelling reason. Standards are created and used to allow people to communicate in a meaningful way.
Every standard, if it is a true standard, has as its entire (and only) goal set the increasing of relevant
communication between people.
</para>

<para>
This primary goal cannot be met however, unless the standard is documented. I have been involved in too many
standardization efforts when it became apparent that <emphasis>everybody knows</emphasis> was the dominant
emotion of those providing documentation. <emphasis>They</emphasis> of the ever present <emphasis>they
say</emphasis> and <emphasis>they know</emphasis> are the bane of good standards. If <emphasis>they
know</emphasis>, why are you doing a standard?
</para>

<para>
A <emphasis>good standard</emphasis> survives because people know how to use it. People know how to use a
standard when it is so transparent, so obvious, and so easy that it become invisible. And a standard becomes
invisible only when the documentation describing how to deploy it is clear, unambiguous, and correct. These
three elements must be present for a standard to be useful, allowing communication and interaction between two
separate and distinct entities to occur without obvious effort. As you read this book, look for the evidence
of these three characteristics and notice how they are seamlessly woven into John's text. Clarity and
unambiguity without <emphasis>correctness</emphasis> provide a technical nightmare. Correctness and clarity
with ambiguity create <emphasis>maybe bits,</emphasis> and correctness and unambiguity without clarity provide
a <emphasis>muddle through</emphasis> scenario.
</para>

<para>
And this is <emphasis>the rest of the story</emphasis> that John couldn't (or wouldn't) bring himself to
state. This book provides a clear, concise, unambiguous, and technically valid presentation of Samba to make
it useful to a user to someone who wants to use the standard to increase communication and the capability
for communication between two or more entities whether person-machine, machine-machine, or person-person.
The intent of this book is not to convince anyone of any agenda political, technical, or social. The intent
is to provide documentation for users who need to know about Samba, how to use it, and how to get on with
their primary responsibilities. While there is pride on John's part because of the tremendous success of
the Samba documentation, he writes for the person who needs a tool to accomplish a particular job, and who has
selected Samba to be that tool.
</para>

<para>
The book is a monument to John's perseverance and dedication to Samba and in my opinion to the goal of
standardization. By writing this book, John has provided the users of Samba those that want to deploy it to
make things better a clear, easy, and ultimately valuable resource. Additionally, he has increased the
understanding and utility of a highly useful standard, and for this, as much as for the documentation, he is
owed a debt of gratitude by those of us who rely on standards to make our lives more manageable.
</para>

<para>
<simplelist type="vert">
	<member>Carl Cargill, Senior Director</member>
	<member>Corporate Standardization, The Office of the CTO</member>
	<member>Sun Microsystems</member>
</simplelist>
</para>
	
</preface>

	<?latex \cleardoublepage ?>

	<preface id="TOSHpreface">
<title>Preface</title>

<para>
The editors wish to thank you for your decision to purchase this book.
The Official Samba-3 HOWTO and Reference Guide is the result of many years
of accumulation of information, feedback, tips, hints, and happy solutions.
</para>

<para>
Please note that this book is a living document, the contents of which are
constantly being updated. We encourage you to contribute your tips, techniques,
helpful hints, and your special insight into the Windows networking world to
help make the next generation of this book even more valuable to Samba users.
</para>

<para>
We have made a concerted effort to document more comprehensively than has been
done previously the information that may help you to better deploy Samba and to
gain more contented network users.
</para>

<para>
This book provides example configurations, it documents key aspects of Microsoft
Windows networking, provides in-depth insight into the important configuration of
Samba-3, and helps to put all of these into a useful framework.
</para>

<para>
The most recent electronic versions of this document can be found at
<ulink noescape="1" url="http://www.samba.org/">http://www.samba.org/</ulink>
on the <quote>Documentation</quote> page.
</para>

<para>
Updates, patches and corrections are most welcome. Please email your contributions
to any one of the following:
</para>

<para>
<simplelist type="vert">
<member><ulink noescape="1" url="mailto:jelmer@samba.org">Jelmer Vernooij (jelmer@samba.org)</ulink></member>
<member><ulink noescape="1" url="mailto:jht@samba.org">John H. Terpstra (jht@samba.org)</ulink></member>
<member><ulink noescape="1" url="mailto:jerry@samba.org">Gerald (Jerry) Carter (jerry@samba.org)</ulink></member>
</simplelist>
</para>

<para>
We wish to advise that only original and unencumbered material can be published. Please do not submit
content that is not your own work unless proof of consent from the copyright holder accompanies your
submission.
</para>

  <!-- the conventions used in this book -->
  <sect1>
    <title>Conventions Used</title>

    <para>
    The following notation conventions are used throughout this book:
    </para>

    <itemizedlist>
      <listitem>
	<para>
	TOSHARG2 is used as an abbreviation for the book, <quote>The Official Samba-3
	HOWTO and Reference Guide, Second Edition</quote> Editors: John H. Terpstra and Jelmer R. Vernooij,
	Publisher: Prentice Hall, ISBN: 0131882228.
	</para>
      </listitem>

      <listitem>
	<para>
	S3bE2 is used as an abbreviation for the book, <quote>Samba-3 by Example, Second Edition</quote> 
	Editors: John H. Terpstra, Publisher: Prentice Hall, ISBN: 013188221X.
	</para>
      </listitem>

      <listitem>
        <para>
	Directories and filenames appear in mono-font. For example,
        <filename moreinfo="none">/etc/pam.conf</filename>.
        </para>
      </listitem>

      <listitem>
        <para>
	Executable names are bolded. For example, <literal>smbd</literal>.
	</para>
      </listitem>

      <listitem>
        <para>
	Menu items and buttons appear in bold. For example, click <guibutton moreinfo="none">Next</guibutton>.
        </para>
      </listitem>

      <listitem>
        <para>
	Selecting a menu item is indicated as:
	<menuchoice moreinfo="none">
		<guimenu moreinfo="none">Start</guimenu>
		<guimenuitem moreinfo="none">Control Panel</guimenuitem>
		<guimenuitem moreinfo="none">Administrative Tools</guimenuitem>
		<guimenuitem moreinfo="none">Active Directory Users and Computers</guimenuitem>
	</menuchoice>
       </para>
      </listitem>
    </itemizedlist>

  </sect1>


</preface>

	<?latex \cleardoublepage ?>
	<preface id="IntroSMB">
<prefaceinfo>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
	<pubdate>June 29, 2003</pubdate>
</prefaceinfo>

<title>Introduction</title>

<para><quote>
A man's gift makes room for him before great men. Gifts are like hooks that can catch
hold of the mind taking it beyond the reach of forces that otherwise might constrain it.
</quote> --- Anon.
</para>


<para>
This is a book about Samba. It is a tool, a derived work of the labors
of many and of the diligence and goodwill of more than a few.
This book contains material that has been contributed in a persistent belief
that each of us can add value to our neighbors as well as to those who will
follow us.
</para>

<para>
This book is designed to meet the needs of the Microsoft network administrator.
UNIX administrators will benefit from this book also, though they may complain
that it is hard to find the information they think they need. So if you are a
Microsoft certified specialist, this book should meet your needs rather well.
If you are a UNIX or Linux administrator, there is no need to feel badly <?latex --- ?> you
should have no difficulty finding answers to your current concerns also.
</para>

<sect1>
<title>What Is Samba?</title>

	<para>
	Samba is a big, complex project. The Samba project is ambitious and exciting.
	The team behind Samba is a group of some thirty individuals who are spread
	the world over and come from an interesting range of backgrounds. This team
	includes scientists, engineers, programmers, business people, and students.
	</para>

	<para>
	Team members were drawn into active participation through the desire to help
	deliver an exciting level of transparent interoperability between Microsoft
	Windows and the non-Microsoft information
	technology world. 
	</para>

	<para>
	The slogan that unites the efforts behind the Samba project says:
	<emphasis>Samba, Opening Windows to a Wider World!</emphasis> The goal
	behind the project is one of removing barriers to interoperability.
	</para>

	<para>
	Samba provides file and print services for Microsoft Windows clients. These
	services may be hosted off any TCP/IP-enabled platform. The original deployment
	platforms were UNIX and Linux, though today it is in common use across
	a broad variety of systems.
	</para>

	<para>
	The Samba project includes not only an impressive feature set in file and print
	serving capabilities, but has been extended to include client functionality,
	utilities to ease migration to Samba, tools to aid interoperability with
	Microsoft Windows, and administration tools.
	</para>

	<para>
	The real people behind Samba are users like you. You have inspired the
	developers (the Samba Team) to do more than any of them imagined could or should
	be done. User feedback drives Samba development. Samba-3 in particular incorporates
	a huge amount of work done as a result of user requests, suggestions and direct
	code contributions.
	</para>

</sect1>

<sect1>
<title>Why This Book?</title>

	<para>
	There is admittedly a large number of Samba books on the market today and
	each book has its place. Despite the apparent plethora of books, Samba
	as a project continues to receive much criticism for failing to provide
	sufficient documentation. Samba is also criticized for being too complex
	and too difficult to configure. In many ways this is evidence of the
	success of Samba as there would be no complaints if it was not successful.
	</para>

	<para>
	The Samba Team members work predominantly with UNIX and Linux, so
	it is hardly surprising that existing Samba documentation should reflect
	that orientation. The original HOWTO text documents were intended to provide
	some tips, a few golden nuggets, and if they helped anyone then that was
	just wonderful. But the HOWTO documents lacked structure and context. They were
	isolated snapshots of information that were written to pass information
	on to someone else who might benefit. They reflected a need to transmit
	more information that could be conveniently put into manual pages.
	</para>

	<para>
	The original HOWTO documents were written by different authors. Most HOWTO
	documents are the result of feedback and contributions from numerous
	authors. In this book we took care to preserve as much original content as
	possible. As you read this book you will note that chapters were written by
	multiple authors, each of whom has his own style. This demonstrates
	the nature of the Open Source software development process.
	</para>

	<para>
	Out of the original HOWTO documents sprang a collection of unofficial
	HOWTO documents that are spread over the Internet. It is sincerely intended
	that this work will <emphasis>not</emphasis> replace the valuable unofficial 
	HOWTO work that continues to flourish. If you are involved in unofficial 
	HOWTO production then please continue your work!
	</para>

	<para>
	Those of you who have dedicated your labors to the production of unofficial
	HOWTOs, to Web page information regarding Samba, or to answering questions
	on the mailing lists or elsewhere, may be aware that this is a labor
	of love. We would like to know about your contribution and willingly receive
	the precious pearls of wisdom you have collected. Please email your contribution to
	<ulink noescape="1" url="mailto:jht@samba.org">John H. Terpstra (jht@samba.org)</ulink>.
	As a service to other users we will gladly adopt material that is technically accurate.
	</para>

	<para>
	Existing Samba books are largely addressed to the UNIX administrator.
	From the perspective of this target group the existing books serve
	an adequate purpose, with one exception <?latex --- ?> now that Samba-3 is out
	they need to be updated!
	</para>

	<para>
	This book, the <emphasis>Official Samba-3 HOWTO and Reference Guide</emphasis>,
	includes the Samba-HOWTO-Collection.pdf that ships with Samba.
	These documents have been written with a new design intent and purpose.
	</para>

	<para>
	Over the past two years many Microsoft network administrators have adopted
	Samba and have become interested in its deployment. Their information needs
	are very different from that of the UNIX administrator. This book has been
	arranged and the information presented from the perspective of someone with previous
	Microsoft Windows network administrative training and experience.
	</para>

</sect1>

<sect1>
<title>Book Structure and Layout</title>

	<para>
	This book is presented in six parts:
	</para>

	<variablelist>
		<varlistentry><term>General Installation</term>
			<listitem><para>
			Designed to help you get Samba-3 running quickly.
			The Fast Start chapter is a direct response to requests from
			Microsoft network administrators for some sample configurations
			that <emphasis>just work</emphasis>.
			</para></listitem>
		</varlistentry>

		<varlistentry><term>Server Configuration Basics</term>
			<listitem><para>
			The purpose of this section is to aid the transition from existing
			Microsoft Windows network knowledge to Samba terminology and norms.
			The chapters in this part each cover the installation of one type of 
			Samba server.
			</para></listitem>
		</varlistentry>

		<varlistentry><term>Advanced Configuration</term>
			<listitem><para>
			The mechanics of network browsing have long been the Achilles heel of
			all Microsoft Windows users. Samba-3 introduces new user and machine
			account management facilities, a new way to map UNIX groups and Windows
			groups, Interdomain trusts, new loadable file system drivers (VFS), and
			more. New with this document is expanded printing documentation, as well
			as a wealth of information regarding desktop and user policy handling,
			use of desktop profiles, and techniques for enhanced network integration.
			This section makes up the core of the book. Read and enjoy.
			</para></listitem>
		</varlistentry>

		<varlistentry><term>Migration and Updating</term>
			<listitem><para>
			A much requested addition to the book is information on how to migrate
			from Microsoft Windows NT4 to Samba-3, as well as an overview of what the
			issues are when moving from Samba-2.x to Samba-3.
			</para></listitem>
		</varlistentry>

		<varlistentry><term>Troubleshooting</term>
			<listitem><para>
			This short section should help you when all else fails.
			</para></listitem>
		</varlistentry>

		<varlistentry><term>Reference Section</term>
			<listitem><para>
			Here you will find a collection of things that are either too peripheral
			for most users, or are a little left of field to be included in the
			main body of information.
			</para></listitem>
		</varlistentry>
	</variablelist>
		
<para>
Welcome to Samba-3 and the first published document to help you and your users to enjoy a whole
new world of interoperability between Microsoft Windows and the rest of the world.
</para>

</sect1>

</preface>

	<?latex \cleardoublepage ?>

<!-- Chapters -->
<part id="introduction">
<title>General Installation</title>
<?latex \pagenumbering{arabic} ?>

<partintro>
<title>Preparing Samba for Configuration</title>

<para>
This section of the Samba-HOWTO-Collection contains general info on how to install Samba 
and how to configure the parts of Samba you will most likely need.
PLEASE read this.
</para>

</partintro>

	<?latex \cleardoublepage ?>

	<chapter id="install">
<chapterinfo>
	
<author>
	<firstname>Andrew</firstname><surname>Tridgell</surname>
	<affiliation>
		<orgname>Samba Team</orgname>
		<address format="linespecific"><email>tridge@samba.org</email></address>
	</affiliation>
</author>
	<author>
<firstname>Jelmer</firstname><othername>R.</othername><surname>Vernooij</surname><othername>R.</othername>
<affiliation>
	<orgname>The Samba Team</orgname>
	<address format="linespecific"><email>jelmer@samba.org</email></address>
</affiliation></author>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
	<author>
<firstname>Karl</firstname><surname>Auer</surname>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>kauer@biplane.com.au</email></address>
</affiliation>
</author>
	<author>
<firstname>Dan</firstname><surname>Shearer</surname>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>dan@samba.org</email></address>
</affiliation>
</author>
	<!-- Isn't some of this written by others as well? -->

</chapterinfo>

<title>How to Install and Test SAMBA</title>

<sect1>
	<title>Obtaining and Installing Samba</title>

	<para>
	<indexterm significance="normal"><primary>packages</primary></indexterm>
	Binary packages of Samba are included in almost any Linux or UNIX distribution. There are also some
	packages available at <ulink url="http://samba.org/">the Samba home page</ulink>. Refer to the manual of your
	operating system for details on installing packages for your specific operating system.
	</para>

	<para>
	<indexterm significance="normal"><primary>compile</primary></indexterm>
	If you need to compile Samba from source, check <link linkend="compiling">How to Compile Samba</link>.
	</para>

</sect1>

<sect1>
	<title>Configuring Samba (smb.conf)</title>

	<para>
	<indexterm significance="normal"><primary>/etc/samba/smb.conf</primary></indexterm>
	<indexterm significance="normal"><primary>SWAT</primary></indexterm>
	Samba's configuration is stored in the <filename moreinfo="none">smb.conf</filename> file, which usually resides in
	<filename moreinfo="none">/etc/samba/smb.conf</filename> or <filename moreinfo="none">/usr/local/samba/lib/smb.conf</filename>. You can either
	edit this file yourself or do it using one of the many graphical tools that are available, such as the
	Web-based interface SWAT, that is included with Samba.
	</para>

	<sect2>
	<title>Configuration File Syntax</title>

	<para>
	<indexterm significance="normal"><primary>section name</primary></indexterm>
	The <filename moreinfo="none">smb.conf</filename> file uses the same syntax as the various old <filename moreinfo="none">.ini</filename> files in Windows
	3.1: Each file consists of various sections, which are started by putting the section name between brackets
	(<literal moreinfo="none">[]</literal>) on a new line. Each contains zero or more key/value pairs separated by an equality
	sign (<literal moreinfo="none">=</literal>). The file is just a plaintext file, so you can open and edit it with your favorite
	editing tool.
	</para>

	<para>
	<indexterm significance="normal"><primary>meta-service</primary></indexterm>
	<indexterm significance="normal"><primary>print</primary><secondary>queue</secondary></indexterm>
	<indexterm significance="normal"><primary>share</primary></indexterm>
	<indexterm significance="normal"><primary>spooler.</primary></indexterm>
	<indexterm significance="normal"><primary>print</primary><secondary>spooler</secondary></indexterm>
	<indexterm significance="normal"><primary>spool</primary><secondary>directory</secondary></indexterm>
	Each section in the <filename moreinfo="none">smb.conf</filename> file represents either a share or a meta-service on the Samba server. The
	section <literal moreinfo="none">[global]</literal> is special, since it contains settings that apply to the whole Samba
	server.  Samba supports a number of meta-services, each of which serves its own purpose. For example, the
	<literal moreinfo="none">[homes]</literal> share is a meta-service that causes Samba to provide a personal home share for
	each user. The <literal moreinfo="none">[printers]</literal> share is a meta-service that establishes print queue support
	and that specifies the location of the intermediate spool directory into which print jobs are received
	from Windows clients prior to being dispatched to the UNIX/Linux print spooler.
	</para>

	<para>
<indexterm significance="normal"><primary>printers</primary></indexterm>
<indexterm significance="normal"><primary>meta-service</primary></indexterm>
<indexterm significance="normal"><primary>printcap</primary></indexterm>
<indexterm significance="normal"><primary>lpstat</primary></indexterm>
<indexterm significance="normal"><primary>CUPS API</primary></indexterm>
<indexterm significance="normal"><primary>browseable</primary></indexterm>
	The <literal moreinfo="none">printers</literal> meta-service will cause every printer that is either specified in a
	<literal moreinfo="none">printcap</literal> file, via the <literal>lpstat</literal>,  or via the CUPS API, to be
	published as a shared print queue. The <literal moreinfo="none">printers</literal> stanza in the <filename moreinfo="none">smb.conf</filename> file can
	be set as not browseable. If it is set to be browseable, then it will be visible as if it is a share.
	That makes no sense given that this meta-service is responsible only for making UNIX system printers
	available as Windows print queues. If a <literal moreinfo="none">comment</literal> parameter is specified, the value
	of it will be displayed as part of the printer name in Windows Explorer browse lists.
	</para>

	<para>
	<indexterm significance="normal"><primary>stanza</primary></indexterm>
	Each section of the <filename moreinfo="none">smb.conf</filename> file that specifies a share, or a meta-service, is called a stanza.
	The <literal moreinfo="none">global</literal> stanza specifies settings that affect all the other stanzas in the
	<filename moreinfo="none">smb.conf</filename> file. Configuration parameters are documented in the <filename moreinfo="none">smb.conf</filename> man page. Some parameters
	can be used only in the <literal moreinfo="none">global</literal> stanza, some only in share or meta-service stanzas,
	and some can be used globally or just within a share or meta-service stanza.
	</para>

	<para>
	<indexterm significance="normal"><primary>minimal</primary><secondary>configuration</secondary></indexterm>
	<link linkend="smbconfminimal">A minimal smb.conf</link> contains a very minimal <filename moreinfo="none">smb.conf</filename>.
	<indexterm significance="normal"><primary>minimal configuration</primary></indexterm>
	</para>

	<example id="smbconfminimal">
		<title>A minimal smb.conf</title>
		<simplelist>

		<member> </member><member><parameter>[global]</parameter></member>
		<member><indexterm>workgroup</indexterm><parameter>workgroup = WKG</parameter></member>
		<member><indexterm>netbios name</indexterm><parameter>netbios name = MYNAME</parameter></member>
		<member> </member><member><parameter>[share1]</parameter></member>
		<member><indexterm>path</indexterm><parameter>path = /tmp</parameter></member>

		<member> </member><member><parameter>[share2]</parameter></member>
		<member><indexterm>path</indexterm><parameter>path = /my_shared_folder</parameter></member>
		<member><indexterm>comment</indexterm><parameter>comment = Some random files</parameter></member>
	</simplelist>
	</example>

</sect2>

<sect2 id="tdbdocs">
	<title>TDB Database File Information</title>

	<para>
	This section contains brief descriptions of the databases that are used by Samba-3.
	</para>

	<para>
<indexterm significance="normal"><primary>tdb file locations</primary></indexterm>
	The directory in which Samba stores the tdb files is determined by compile-time directives. Samba-3 stores
	tdb files in two locations. The best way to determine these locations is to execute the following
	command:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> smbd -b | grep PRIVATE_DIR
   PRIVATE_DIR: /etc/samba/private
</screen>
	This means that the confidential tdb files are stored in the <filename moreinfo="none">/etc/samba/private</filename>
	directory. Samba-3 also uses a number of tdb files that contain more mundane data. The location of
	these files can be found by executing:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> smbd -b | grep LOCKDIR
   LOCKDIR: /var/lib/samba
</screen>
	Therefore the remaining control files will, in the example shown, be stored in the
	<filename moreinfo="none">/var/lib/samba</filename> directory.
	</para>

	<para>
<indexterm significance="normal"><primary>tdb file descriptions</primary></indexterm>
	The persistent tdb files are described in <link linkend="tdbpermfiledesc">the Persistent TDB File
	Descriptions table</link>. All persistent tdb files should be regularly backed up. Use the
	<literal>tdbbackup</literal> utility to backup the tdb files. All persistent tdb files must be
	preserved during machine migrations, updates and upgrades.
	</para>

	<para>
	The temporary tdb files do not need to be backed up, nor do they need to be preseved across machine
	migrations, updates or upgrades. The temporary tdb files are described in <link linkend="tdbtempfiledesc">
	the Temporary TDB File Descriptions</link>.
	</para>

        <table frame="all" id="tdbpermfiledesc"><title>Persistent TDB File Descriptions</title>
        <tgroup cols="2">
			<colspec align="left"/>
			<colspec align="justify" colwidth="1*"/>
			<colspec align="left"/>
                <thead>
                <row>
                        <entry align="left">Name</entry>
                        <entry align="justify">Description</entry>
                </row>
                </thead>
                <tbody>
		<row>
			<entry>account_policy</entry>
			<entry><para>Samba/NT account policy settings, includes password expiration settings.</para></entry>
		</row>
		<row>
			<entry>group_mapping</entry>
			<entry><para>Mapping table from Windows groups/SID to UNIX groups.</para></entry>
		</row>
		<row>
			<entry>ntdrivers</entry>
			<entry><para>Stores per-printer installed driver information.</para></entry>
		</row>
		<row>
			<entry>ntforms</entry>
			<entry><para>Stores per-printer installed forms information.</para></entry>
		</row>
		<row>
			<entry>ntprinters</entry>
			<entry><para>Stores the per-printer devmode configuration settings.</para></entry>
		</row>
		<row>
			<entry>passdb</entry>
			<entry><para>
				Exists only when the tdbsam passwd backend is used. This file stores the
				SambaSAMAccount information. Note: This file requires that user POSIX account information is
				availble from either the /etc/passwd file, or from an alternative system source.
			</para></entry>
		</row>
		<row>
			<entry>registry</entry>
			<entry><para>
				Read-only Samba database of a Windows registry skeleton that provides support for exporting 
				various database tables via the winreg RPCs.
			</para></entry>
		</row>
		<row>
			<entry>secrets</entry>
			<entry><para>
				This file stores the Workgroup/Domain/Machine SID, the LDAP directory update password, and
				a further collection of critical environmental data that is necessary for Samba to operate
				correctly. This file contains very sensitive information that must be protected. It is stored
				in the PRIVATE_DIR directory.
			</para></entry>
		</row>
		<row>
			<entry>share_info</entry>
			<entry><para>Stores per-share ACL information.</para></entry>
		</row>
		<row>
			<entry>winbindd_idmap</entry>
			<entry><para>Winbindd's local IDMAP database.</para></entry>
		</row>
		</tbody>
	</tgroup>
	</table>

        <table frame="all" id="tdbtempfiledesc"><title>Temporary TDB File Descriptions</title>
        <tgroup cols="3">
			<colspec align="left"/>
			<colspec align="justify" colwidth="1*"/>
			<colspec align="left"/>
                <thead>
                <row>
                        <entry align="left">Name</entry>
                        <entry align="justify">Description</entry>
                        <entry align="center">Backup</entry>
                </row>
                </thead>
                <tbody>
		<row>
			<entry>brlock</entry>
			<entry><para>Byte-range locking information.</para></entry>
			<entry>No</entry>
		</row>
		<row>
			<entry>connections</entry>
			<entry><para>A temporary cache for current connection information used to enforce max connections.</para></entry>
			<entry>no</entry>
		</row>
		<row>
			<entry>eventlog/*tdb</entry>
			<entry><para>Records of eventlog entries. In most circumstances this is just a cache of system logs.</para></entry>
			<entry>no</entry>
		</row>
		<row>
			<entry>gencache</entry>
			<entry><para>Generic caching database for dead WINS servers and trusted domain data.</para></entry>
			<entry>no</entry>
		</row>
		<row>
			<entry>login_cache</entry>
			<entry><para>A temporary cache for login information, in particular bad password attempts.</para></entry>
			<entry>no</entry>
		</row>
		<row>
			<entry>messages</entry>
			<entry><para>Temporary storage of messages being processed by smbd.</para></entry>
			<entry>no</entry>
		</row>
		<row>
			<entry>netsamlogon_cache</entry>
			<entry><para>Caches user net_info_3 structure data from net_samlogon requests (as a domain member).</para></entry>
			<entry>no</entry>
		</row>
		<row>
			<entry>perfmon/*.tdb</entry>
			<entry><para>Performance counter information.</para></entry>
			<entry>no</entry>
		</row>
		<row>
			<entry>printing/*.tdb</entry>
			<entry><para>Cached output from lpq command created on a per-print-service basis.</para></entry>
			<entry>no</entry>
		</row>
		<row>
			<entry>schannel_store</entry>
			<entry><para>
				A confidential file, stored in the PRIVATE_DIR, containing crytographic connection
				information so that clients that have temporarily disconnected can reconnect without
				needing to renegotiate the connection setup process.
			</para></entry>
			<entry>no</entry>
		</row>
		<row>
			<entry>sessionid</entry>
			<entry><para>Temporary cache for miscellaneous session information and for utmp handling.</para></entry>
			<entry>no</entry>
		</row>
		<row>
			<entry>unexpected</entry>
			<entry><para>Stores packets received for which no process is actively listening.</para></entry>
			<entry>no</entry>
		</row>
		<row>
			<entry>winbindd_cache</entry>
			<entry><para>Cache of Identity information received from an NT4 domain or from ADS. Includes user
				lists, etc.</para></entry>
			<entry>yes</entry>
		</row>
		</tbody>
	</tgroup>
	</table>

</sect2>

<sect2>
	<title>Starting Samba</title>

	<para>
	<indexterm significance="normal"><primary>daemon</primary></indexterm>
	Samba essentially consists of two or three daemons. A daemon is a UNIX application that runs in the background and provides services.
	An example of a service is the Apache Web server for which the daemon is called <literal>httpd</literal>. In the case of Samba there
	are three daemons, two of which are needed as a minimum.
	</para>

	<para>
	The Samba server is made up of the following daemons:
	</para>

	<variablelist>
		<varlistentry><term>nmbd</term>
			<listitem><para>
			<indexterm significance="normal"><primary>smbd</primary></indexterm>
			<indexterm significance="normal"><primary>starting samba</primary><secondary>smbd</secondary></indexterm>
			This daemon handles all name registration and resolution requests. It is the primary vehicle involved
			in network browsing. It handles all UDP-based protocols. The <literal>nmbd</literal> daemon should
			be the first command started as part of the Samba startup process.
			</para></listitem>
		</varlistentry>

		<varlistentry><term>smbd</term>
			<listitem><para>
			<indexterm significance="normal"><primary>nmbd</primary></indexterm>
			<indexterm significance="normal"><primary>starting samba</primary><secondary>nmbd</secondary></indexterm>
			This daemon handles all TCP/IP-based connection services for file- and print-based operations. It also
			manages local authentication. It should be started immediately following the startup of <literal>nmbd</literal>.
			</para></listitem>
		</varlistentry>

		<varlistentry><term>winbindd</term>
			<listitem><para>
			<indexterm significance="normal"><primary>winbindd</primary></indexterm>
			<indexterm significance="normal"><primary>starting samba</primary><secondary>winbindd</secondary></indexterm>
			This daemon should be started when Samba is a member of a Windows NT4 or ADS domain. It is also needed when
			Samba has trust relationships with another domain. The <literal>winbindd</literal> daemon will check the
			<filename moreinfo="none">smb.conf</filename> file for the presence of the <parameter moreinfo="none">idmap uid</parameter> and <parameter moreinfo="none">idmap gid</parameter>
			parameters. If they are are found, <literal>winbindd</literal> will use the values specified for
			for UID and GID allocation. If these parameters are not specified, <literal>winbindd</literal>
			will start but it will not be able to allocate UIDs or GIDs.
			</para></listitem>
		</varlistentry>
	</variablelist>

	<para>
	<indexterm significance="normal"><primary>startup</primary><secondary>process</secondary></indexterm>
	When Samba has been packaged by an operating system vendor, the startup process is typically a custom feature of its
	integration into the platform as a whole. Please refer to your operating system platform administration manuals for
	specific information pertaining to correct management of Samba startup.
	</para>

</sect2>
	
<sect2>
	<title>Example Configuration</title>
	
	<para>
	<indexterm significance="normal"><primary>examples</primary></indexterm>
	<indexterm significance="normal"><primary>source code</primary></indexterm>
	<indexterm significance="normal"><primary>distribution</primary></indexterm>
	<indexterm significance="normal"><primary>tarball</primary></indexterm>
	<indexterm significance="normal"><primary>package</primary></indexterm>
	There are sample configuration files in the examples subdirectory in the source code distribution tarball
	package. It is suggested you read them carefully so you can see how the options go together in practice. See
	the man page for all the options.  It might be worthwhile to start out with the
	<filename moreinfo="none">smb.conf.default</filename> configuration file and adapt it to your needs. It contains plenty of comments.
	</para>

	<para>
	<indexterm significance="normal"><primary>simplest</primary><secondary>configuration</secondary></indexterm>
	The simplest useful configuration file would contain something like that shown in
	<link linkend="simple-example">Another simple smb.conf File</link>.
	<indexterm significance="normal"><primary>simple configuration</primary></indexterm>
	</para>

<example id="simple-example">
<title>Another simple smb.conf File</title>
<simplelist>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>workgroup</indexterm><parameter>workgroup = MIDEARTH</parameter></member>

<member> </member><member><parameter>[homes]</parameter></member>
<member><indexterm>guest ok</indexterm><parameter>guest ok = no</parameter></member>
<member><indexterm>read only</indexterm><parameter>read only = no</parameter></member>
</simplelist>
</example>
	
	<para>
	<indexterm significance="normal"><primary>connections</primary></indexterm>
	<indexterm significance="normal"><primary>account</primary></indexterm>
	<indexterm significance="normal"><primary>login name</primary></indexterm>
	<indexterm significance="normal"><primary>service name</primary></indexterm>
	This will allow connections by anyone with an account on the server, using either
	their login name or <parameter>homes</parameter> as the service name.
	(Note: The workgroup that Samba should appear in must also be set. The default
	workgroup name is WORKGROUP.)
	</para>
	
	<para>
	<indexterm significance="normal"><primary>smbd</primary></indexterm>
	Make sure you put the <filename moreinfo="none">smb.conf</filename> file in the correct place. Note, the correct location of this file
	depends on how the binary files were built. You can discover the correct location by executing from
	the directory that contains the <literal>smbd</literal> command file:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> smbd -b | grep smb.conf
</screen>
	</para>

	<para>
	<indexterm significance="normal"><primary>security</primary><secondary>settings</secondary></indexterm>
	For more information about security settings for the <parameter>[homes]</parameter> share, please refer to 
	<link linkend="securing-samba">Securing Samba</link>.
	</para>

<sect3>
	<title>Test Your Config File with <literal>testparm</literal></title>

	<para>
	<indexterm significance="normal"><primary>validate</primary></indexterm>
	<indexterm significance="normal"><primary>testparm</primary></indexterm>
	<indexterm significance="normal"><primary>misconfigurations</primary></indexterm>
	It's important to validate the contents of the <filename moreinfo="none">smb.conf</filename> file using the <application moreinfo="none">testparm</application> program.
	If testparm runs correctly, it will list the loaded services. If not, it will give an error message.
	Make sure it runs correctly and that the services look reasonable before proceeding. Enter the command: 
	<screen format="linespecific">
	<prompt moreinfo="none">root# </prompt> testparm /etc/samba/smb.conf
	</screen>
	Testparm will parse your configuration file and report any unknown parameters or incorrect syntax.
	It also performs a check for common misconfigurations and will issue a warning if one is found.
	</para>

	<para>
	Always run testparm again whenever the <filename moreinfo="none">smb.conf</filename> file is changed!
	</para>

	<para>
	<indexterm significance="normal"><primary>smbd</primary></indexterm>
	<indexterm significance="normal"><primary>nmbd</primary></indexterm>
	<indexterm significance="normal"><primary>winbindd</primary></indexterm>
	<indexterm significance="normal"><primary>configuration</primary><secondary>documentation</secondary></indexterm>
	The <filename moreinfo="none">smb.conf</filename> file is constantly checked by the Samba daemons <literal>smbd</literal> and every instance of
	itself that it spawns, <literal>nmbd</literal> and <literal>winbindd</literal>. It is good practice to
	keep this file as small as possible. Many administrators prefer to document Samba configuration settings
	and thus the need to keep this file small goes against good documentation wisdom. One solution that may
	be adopted is to do all documentation and configuration in a file that has another name, such as
	<filename moreinfo="none">smb.conf.master</filename>. The <literal>testparm</literal> utility can be used to generate a
	fully optimized <filename moreinfo="none">smb.conf</filename> file from this master configuration and documtenation file as shown here:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> testparm -s smb.conf.master &gt; smb.conf
</screen>
	This administrative method makes it possible to maintain detailed configuration change records while at
	the same time keeping the working <filename moreinfo="none">smb.conf</filename> file size to the minimum necessary.
	</para>

</sect3>
</sect2>

<sect2>
	<title>SWAT</title>

	<para>
	<indexterm significance="normal"><primary>swat</primary></indexterm>
	SWAT is a Web-based interface that can be used to facilitate the configuration of Samba.  SWAT might not
	be available in the Samba package that shipped with your platform, but in a separate package. If it is
	necesaary to built SWAT please read the SWAT man page regarding compilation, installation, and
	configuration of SWAT from the source code.
	</para>

	<para>
	To launch SWAT, just run your favorite Web browser and point it to
	<ulink url="http://localhost:901/" noescape="1">http://localhost:901/</ulink>.
	Replace <replaceable>localhost</replaceable> with the name of the computer on which
	Samba is running if that is a different computer than your browser.
	</para>

	<para>
	SWAT can be used from a browser on any IP-connected machine, but be aware that connecting from a remote
	machine leaves your connection open to password sniffing because passwords will be sent over the wire in the clear. 
	</para>

	<para>
	More information about SWAT can be found in <link linkend="SWAT">The Samba Web Administration Tool</link>.
	</para>

</sect2>

</sect1>

<sect1>
	<title>List Shares Available on the Server</title>

	<para>
	To list shares that are available from the configured Samba server, execute the
	following command:
	</para>

<para><screen format="linespecific">
<prompt moreinfo="none">$ </prompt><userinput moreinfo="none">smbclient -L <replaceable>yourhostname</replaceable></userinput>
</screen></para>

	<para>
	You should see a list of shares available on your server. If you do not, then
	something is incorrectly configured. This method can also be used to see what shares 
	are available on other SMB servers, such as Windows 2000.
	</para>

	<para>
	If you choose user-level security, you may find that Samba requests a password
	before it will list the shares. See the <literal>smbclient</literal> man page for details.
	You can force it to list the shares without a password by adding the option
	<option>-N</option> to the command line. 
	</para>
</sect1>

<sect1>
	<title>Connect with a UNIX Client</title>
	
	<para>
	Enter the following command:
<screen format="linespecific">
<prompt moreinfo="none">$ </prompt><userinput moreinfo="none">smbclient <replaceable> //yourhostname/aservice</replaceable></userinput>
</screen></para>
	
	<para>Typically <replaceable>yourhostname</replaceable> is the name of the host on which <application moreinfo="none">smbd</application>
	has been installed. The <replaceable>aservice</replaceable> is any service that has been defined in the <filename moreinfo="none">smb.conf</filename>
	file. Try your username if you just have a <parameter>[homes]</parameter> section in the <filename moreinfo="none">smb.conf</filename> file.</para>

	<para>Example: If the UNIX host is called <replaceable>bambi</replaceable> and a valid login name
	is <replaceable>fred</replaceable>, you would type:</para>

<para><screen format="linespecific">
<prompt moreinfo="none">$ </prompt><userinput moreinfo="none">smbclient //<replaceable>bambi</replaceable>/<replaceable>fred</replaceable></userinput>
</screen></para>
</sect1>

<sect1>
	<title>Connect from a Remote SMB Client</title>

	<para>
	Now that Samba is working correctly locally, you can try to access it from other clients. Within a few
	minutes, the Samba host should be listed in the Network Neighborhood on all Windows clients of its subnet.
	Try browsing the server from another client or "mounting" it.
	</para>

	<para>
	Mounting disks from a DOS, Windows, or OS/2 client can be done by running a command such as:
<screen format="linespecific">
<prompt moreinfo="none">C:\&gt; </prompt><userinput moreinfo="none">net use m: \\servername\service</userinput>
</screen>
	Where the drive letter m: is any available drive letter. It is important to double-check that the
	service (share) name that you used does actually exist.
	</para>

	<para>
	Try printing, for example,
<screen format="linespecific">
<prompt moreinfo="none">C:\&gt; </prompt><userinput moreinfo="none">net use lpt1:	\\servername\spoolservice</userinput>
</screen>
	The <literal moreinfo="none">spoolservice</literal> is the name of the printer (actually the print queue) on the target
	server. This will permit all print jobs that are captured by the lpt1: port on the Windows client to
	be sent to the printer that owns the spoolservice that has been specified.
	</para>

<para>
<screen format="linespecific"><prompt moreinfo="none">C:\&gt; </prompt><userinput moreinfo="none">print filename</userinput>
</screen></para>

	<sect2>
	<title>What If Things Don't Work?</title>
	
	<para>
	You might want to read <link linkend="diagnosis">The Samba Checklist</link>.  If you are still
	stuck, refer to <link linkend="problems">Analyzing and Solving Samba Problems</link>.  Samba has
	been successfully installed at thousands of sites worldwide.  It is unlikely that your particular problem is
	unique, so it might be productive to perform an Internet search to see if someone else has encountered your
	problem and has found a way to overcome it.
	</para>

	<para>
	If you are new to Samba, and particularly if you are new to Windows networking, or to UNIX/Linux,
	the book <quote>Samba-3 by Example</quote> will help you to create a validated network environment.
	Simply choose from the first five chapters the network design that most closely matches site needs,
	then follow the simple step-by-step procedure to deploy it. Later, when you have a working network
	you may well want to refer back to this book for further insight into opportunities for improvement.
	</para>

	</sect2>

	<sect2>
	<title>Still Stuck?</title>

	<para>
	The best advice under the stress of abject frustration is to cool down! That may be challenging
	of itself, but while you are angry or annoyed your ability to seek out a solution is somewhat
	undermined. A cool head clears the way to finding the answer you are looking for. Just remember,
	every problem has a solution <?latex --- ?> there is a good chance that someone else has found it
	even though you can't right now. That will change with time, patience and learning.
	</para>

	<para>
	Now that you have cooled down a bit, please refer to <link linkend="diagnosis">the Samba Checklist</link>
	for a process that can be followed to identify the cause of your problem.
	</para>

	</sect2>

</sect1>

<sect1>
<title>Common Errors</title>

<para>
The following questions and issues are raised repeatedly on the Samba mailing list.
</para>

<sect2>
	<title>Large Number of smbd Processes</title>

	<para>
	Samba consists of three core programs: <application moreinfo="none">nmbd</application>, <application moreinfo="none">smbd</application>, and <application moreinfo="none">winbindd</application>. <application moreinfo="none">nmbd</application> is the name server message daemon,
	<application moreinfo="none">smbd</application> is the server message daemon, and <application moreinfo="none">winbindd</application> is the daemon that handles communication with domain controllers.
	</para>

	<para>
	If Samba is <emphasis>not</emphasis> running as a WINS server, then there will be one single instance of
	 <application moreinfo="none">nmbd</application> running on your system. If it is running as a WINS server, then there will be
	two instances <?latex --- ?> one to handle the WINS requests.
	</para>

	<para>
	<application moreinfo="none">smbd</application> handles all connection requests. It spawns a new process for each client
	connection made. That is why you may see so many of them, one per client connection.
	</para>

	<para>
	<application moreinfo="none">winbindd</application> will run as one or two daemons, depending on whether or not it is being
	run in <emphasis>split mode</emphasis> (in which case there will be two instances).
	</para>

	</sect2>

	<sect2>
		<title>Error Message: open_oplock_ipc</title>

	<para>
	An error message is observed in the log files when <application moreinfo="none">smbd</application> is started: <quote>open_oplock_ipc: Failed to
	get local UDP socket for address 100007f. Error was Cannot assign requested.</quote>
	</para>

	<para>
	Your loopback device isn't working correctly. Make sure it is configured correctly. The loopback
	device is an internal (virtual) network device with the IP address <emphasis>127.0.0.1</emphasis>.
	Read your OS documentation for details on how to configure the loopback on your system.
	</para>

	</sect2>

	<sect2>
		<title><quote><errorname>The network name cannot be found</errorname></quote></title>

		<para>
		This error can be caused by one of these misconfigurations:
		</para>

		<itemizedlist>
			<listitem><para>You specified a nonexisting path
			for the share in <filename moreinfo="none">smb.conf</filename>.</para></listitem>

			<listitem><para>The user you are trying to access the share with does not 
			have sufficient permissions to access the path for
			the share. Both read (r) and access (x) should be possible.</para></listitem>

			<listitem><para>The share you are trying to access does not exist.</para></listitem>
	</itemizedlist>

	</sect2>
</sect1>

</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="FastStart">
<chapterinfo>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
</chapterinfo>

<title>Fast Start: Cure for Impatience</title>

<para>
When we first asked for suggestions for inclusion in the Samba HOWTO documentation,
someone wrote asking for example configurations <?latex --- ?> and lots of them. That is remarkably
difficult to do without losing a lot of value that can be derived from presenting
many extracts from working systems. That is what the rest of this document does.
It does so with extensive descriptions of the configuration possibilities within the
context of the chapter that covers it. We hope that this chapter is the medicine 
that has been requested.
</para>

<para>
The information in this chapter is very sparse compared with the book <quote>Samba-3 by Example</quote>
that was written after the original version of this book was nearly complete. <quote>Samba-3 by Example</quote>
was the result of feedback from reviewers during the final copy editing of the first edition. It
was interesting to see that reader feedback mirrored that given by the original reviewers.
In any case, a month and a half was spent in doing basic research to better understand what
new as well as experienced network administrators would best benefit from. The book <quote>Samba-3 by Example</quote>
is the result of that research. What is presented in the few pages of this book is covered
far more comprehensively in the second edition of <quote>Samba-3 by Example</quote>. The second edition
of both books will be released at the same time.
</para>

<para>
So in summary, the book <quote>The Official Samba-3 HOWTO &amp; Reference Guide</quote> is intended
as the equivalent of an auto mechanic's repair guide. The book <quote>Samba-3 by Example</quote> is the
equivalent of the driver's guide that explains how to drive the car. If you want complete network
configuration examples, go to <ulink url="http://www.samba.org/samba/docs/Samba3-ByExample.pdf">Samba-3 by
Example</ulink>.
</para>

<sect1>
<title>Features and Benefits</title>

<para>
Samba needs very little configuration to create a basic working system.
In this chapter we progress from the simple to the complex, for each providing
all steps and configuration file changes needed to make each work. Please note
that a comprehensively configured system will likely employ additional smart
features. These additional features are covered in the remainder of this document.
</para>

<para>
The examples used here have been obtained from a number of people who made
requests for example configurations. All identities have been obscured to protect
the guilty, and any resemblance to unreal nonexistent sites is deliberate.
</para>

</sect1>

<sect1>
<title>Description of Example Sites</title>

<para>
In the first set of configuration examples we consider the case of exceptionally simple system requirements.
There is a real temptation to make something that should require little effort much too complex.
</para>

<para>
<link linkend="anon-ro"/> documents the type of server that might be sufficient to serve CD-ROM images,
or reference document files for network client use. This configuration is also discussed in <link linkend="StandAloneServer"/>, <link linkend="RefDocServer"/>.  The purpose for this configuration
is to provide a shared volume that is read-only that anyone, even guests, can access.
</para>

<para>
The second example shows a minimal configuration for a print server that anyone can print to as long as they
have the correct printer drivers installed on their computer. This is a mirror of the system described in
<link linkend="StandAloneServer"/>, <link linkend="SimplePrintServer"/>.
</para>

<para>
The next example is of a secure office file and print server that will be accessible only to users who have an
account on the system. This server is meant to closely resemble a workgroup file and print server, but has to
be more secure than an anonymous access machine.  This type of system will typically suit the needs of a small
office. The server provides no network logon facilities, offers no domain control; instead it is just a
network-attached storage (NAS) device and a print server.
</para>

<para>
The later example consider more complex systems that will either integrate into existing MS Windows networks
or replace them entirely. These cover domain member servers as well as Samba domain control (PDC/BDC) and
finally describes in detail a large distributed network with branch offices in remote locations.
</para>

</sect1>

<sect1>
<title>Worked Examples</title>

<para>
The configuration examples are designed to cover everything necessary to get Samba 
running. They do not cover basic operating system platform configuration, which is
clearly beyond the scope of this text.
</para>

<para>
It is also assumed that Samba has been correctly installed, either by way of installation
of the packages that are provided by the operating system vendor or through other means.
</para>

	<sect2>
	<title>Standalone Server</title>

	<para>
	<indexterm significance="normal"><primary>Server Type</primary><secondary>Stand-alone</secondary></indexterm>
	A standalone server implies no more than the fact that it is not a domain controller
	and it does not participate in domain control. It can be a simple, workgroup-like
	server, or it can be a complex server that is a member of a domain security context.
	</para>

	<para>
	As the examples are developed, every attempt is made to progress the system toward greater capability, just as
	one might expect would happen in a real business office as that office grows in size and its needs change.
	</para>

		<sect3 id="anon-ro">
		<title>Anonymous Read-Only Document Server</title>

		<para>
		<indexterm significance="normal"><primary>read only</primary><secondary>server</secondary></indexterm>
		The purpose of this type of server is to make available to any user
		any documents or files that are placed on the shared resource. The
		shared resource could be a CD-ROM drive, a CD-ROM image, or a file
		storage area.
		</para>

		<itemizedlist>
			<listitem><para>
			The file system share point will be <filename moreinfo="none">/export</filename>.
			</para></listitem>

			<listitem><para>
			All files will be owned by a user called Jack Baumbach.
			Jack's login name will be <emphasis>jackb</emphasis>. His password will be
			<emphasis>m0r3pa1n</emphasis> <?latex --- ?> of course, that's just the example we are
			using; do not use this in a production environment because
			all readers of this document will know it.
			</para></listitem>
		</itemizedlist>

		<procedure>
		<title>Installation Procedure: Read-Only Server</title>
			<step performance="required"><para>
			Add user to system (with creation of the user's home directory):
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">useradd -c "Jack Baumbach" -m -g users -p m0r3pa1n jackb</userinput>
</screen>
			</para></step>

			<step performance="required"><para>
			Create directory, and set permissions and ownership:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">mkdir /export</userinput>
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">chmod u+rwx,g+rx,o+rx /export</userinput>
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">chown jackb.users /export</userinput>
</screen>
			</para></step>

			<step performance="required"><para>
			Copy the files that should be shared to the <filename moreinfo="none">/export</filename>
			directory.
			</para></step>

			<step performance="required"><para>
			Install the Samba configuration file (<filename moreinfo="none">/etc/samba/smb.conf</filename>)
			as shown in <link linkend="anon-example">Anonymous Read-Only Server Configuration</link>.
			</para></step>

<example id="anon-example">
<title>Anonymous Read-Only Server Configuration</title>
<simplelist>
<member># Global parameters</member>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>workgroup</indexterm><parameter>workgroup = MIDEARTH</parameter></member>
<member><indexterm>netbios name</indexterm><parameter>netbios name = HOBBIT</parameter></member>
<member><indexterm>security</indexterm><parameter>security = share</parameter></member>

<member> </member><member><parameter>[data]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = Data</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /export</parameter></member>
<member><indexterm>read only</indexterm><parameter>read only = Yes</parameter></member>
<member><indexterm>guest ok</indexterm><parameter>guest ok = Yes</parameter></member>
</simplelist>
</example>

			<step performance="required"><para>
			Test the configuration file by executing the following command:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">testparm</userinput>
</screen>
			Alternatively, where you are operating from a master configuration file called
			<filename moreinfo="none">smb.conf.master</filename>, the following sequence of commands might prove
			more appropriate:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> cd /etc/samba
<prompt moreinfo="none">root# </prompt> testparm -s smb.conf.master &gt; smb.conf
<prompt moreinfo="none">root# </prompt> testparm
</screen>
			Note any error messages that might be produced. Proceed only if error-free output has been
			obtained. An example of typical output that should be generated from the above configuration
			file is shown here:
<screen format="linespecific">
Load smb config files from /etc/samba/smb.conf
Processing section "[data]"
Loaded services file OK.
Server role: ROLE_STANDALONE
Press enter to see a dump of your service definitions
<userinput moreinfo="none">[Press enter]</userinput>

# Global parameters
[global]
	workgroup = MIDEARTH
	netbios name = HOBBIT
	security = share

[data]
	comment = Data
	path = /export
	read only = Yes
	guest only = Yes
</screen>
			</para></step>

			<step performance="required"><para>
			Start Samba using the method applicable to your operating system platform. The method that
			should be used is platform dependent. Refer to <link linkend="startingSamba">Starting Samba</link>
			for further information regarding the starting of Samba.
			</para></step>

			<step performance="required"><para>
			Configure your MS Windows client for workgroup <emphasis>MIDEARTH</emphasis>,
			set the machine name to ROBBINS, reboot, wait a few (2 - 5) minutes,
			then open Windows Explorer and visit the Network Neighborhood.
			The machine HOBBIT should be visible. When you click this machine
			icon, it should open up to reveal the <emphasis>data</emphasis> share. After
			you click the share, it should open up to reveal the files previously
			placed in the <filename moreinfo="none">/export</filename> directory.
			</para></step>
		</procedure>

		<para>
		The information above (following # Global parameters) provides the complete
		contents of the <filename moreinfo="none">/etc/samba/smb.conf</filename> file.
		</para>

		</sect3>

		<sect3>
		<title>Anonymous Read-Write Document Server</title>

		<para>
		<indexterm significance="normal"><primary>anonymous</primary><secondary>read-write server</secondary></indexterm>
		We should view this configuration as a progression from the previous example.
		The difference is that shared access is now forced to the user identity of jackb
		and to the primary group jackb belongs to. One other refinement we can make is to
		add the user <emphasis>jackb</emphasis> to the <filename moreinfo="none">smbpasswd</filename> file.
		To do this, execute:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">smbpasswd -a jackb</userinput>
New SMB password: <userinput moreinfo="none">m0r3pa1n</userinput>
Retype new SMB password: <userinput moreinfo="none">m0r3pa1n</userinput>
Added user jackb.
</screen>
		Addition of this user to the <filename moreinfo="none">smbpasswd</filename> file allows all files
		to be displayed in the Explorer Properties boxes as belonging to <emphasis>jackb</emphasis>
		instead of to <emphasis>User Unknown</emphasis>.
		</para>

		<para>
		The complete, modified <filename moreinfo="none">smb.conf</filename> file is as shown in <link linkend="anon-rw"/>.
		</para>

<example id="anon-rw">
<title>Modified Anonymous Read-Write smb.conf</title>
<simplelist>
<member># Global parameters</member>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>workgroup</indexterm><parameter>workgroup = MIDEARTH</parameter></member>
<member><indexterm>netbios name</indexterm><parameter>netbios name = HOBBIT</parameter></member>
<member><indexterm>security</indexterm><parameter>security = SHARE</parameter></member>

<member> </member><member><parameter>[data]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = Data</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /export</parameter></member>
<member><indexterm>force user</indexterm><parameter>force user = jackb</parameter></member>
<member><indexterm>force group</indexterm><parameter>force group = users</parameter></member>
<member><indexterm>read only</indexterm><parameter>read only = No</parameter></member>
<member><indexterm>guest ok</indexterm><parameter>guest ok = Yes</parameter></member>
</simplelist>
</example>

		</sect3>

		<sect3>
		<title>Anonymous Print Server</title>

		<para>
		<indexterm significance="normal"><primary>anonymous</primary><secondary>print server</secondary></indexterm>
		An anonymous print server serves two purposes:
		</para>

		<itemizedlist>
			<listitem><para>
			It allows printing to all printers from a single location.
			</para></listitem>

			<listitem><para>
			It reduces network traffic congestion due to many users trying
			to access a limited number of printers.
			</para></listitem>
		</itemizedlist>

		<para>
		In the simplest of anonymous print servers, it is common to require the installation
		of the correct printer drivers on the Windows workstation. In this case the print
		server will be designed to just pass print jobs through to the spooler, and the spooler
		should be configured to do raw pass-through to the printer. In other words, the print
		spooler should not filter or process the data stream being passed to the printer.
		</para>

		<para>
		In this configuration, it is undesirable to present the Add Printer Wizard, and we do
		not want to have automatic driver download, so we disable it in the following
		configuration. <link linkend="anon-print"/> is the resulting <filename moreinfo="none">smb.conf</filename> file.
		</para>

<example id="anon-print">
<title>Anonymous Print Server smb.conf</title>
<simplelist>
<member># Global parameters</member>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>workgroup</indexterm><parameter>workgroup = MIDEARTH</parameter></member>
<member><indexterm>netbios name</indexterm><parameter>netbios name = LUTHIEN</parameter></member>
<member><indexterm>security</indexterm><parameter>security = share</parameter></member>
<member><indexterm>printcap name</indexterm><parameter>printcap name = cups</parameter></member>
<member><indexterm>disable spoolss</indexterm><parameter>disable spoolss = Yes</parameter></member>
<member><indexterm>show add printer wizard</indexterm><parameter>show add printer wizard = No</parameter></member>
<member><indexterm>printing</indexterm><parameter>printing = cups</parameter></member>

<member> </member><member><parameter>[printers]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = All Printers</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /var/spool/samba</parameter></member>
<member><indexterm>guest ok</indexterm><parameter>guest ok = Yes</parameter></member>
<member><indexterm>printable</indexterm><parameter>printable = Yes</parameter></member>
<member><indexterm>use client driver</indexterm><parameter>use client driver = Yes</parameter></member>
<member><indexterm>browseable</indexterm><parameter>browseable = No</parameter></member>
</simplelist>
</example>

		<para>
		The above configuration is not ideal. It uses no smart features, and it deliberately
		presents a less than elegant solution. But it is basic, and it does print. Samba makes
		use of the direct printing application program interface that is provided by CUPS.
		When Samba has been compiled and linked with the CUPS libraries, the default printing
		system will be CUPS. By specifying that the printcap name is CUPS, Samba will use
		the CUPS library API to communicate directly with CUPS for all printer functions.
		It is possible to force the use of external printing commands by setting the value
		of the <parameter moreinfo="none">printing</parameter> to either SYSV or BSD, and thus the value of
		the parameter <parameter moreinfo="none">printcap name</parameter> must be set to something other than
		CUPS. In such case, it could be set to the name of any file that contains a list
		of printers that should be made available to Windows clients.
		</para>

		<note><para>
		Windows users will need to install a local printer and then change the print
		to device after installation of the drivers. The print to device can then be set to
		the network printer on this machine.
		</para></note>

		<para>
		Make sure that the directory <filename moreinfo="none">/var/spool/samba</filename> is capable of being used
		as intended. The following steps must be taken to achieve this:
		</para>

		<itemizedlist>
			<listitem><para>
			The directory must be owned by the superuser (root) user and group:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">chown root.root /var/spool/samba</userinput>
</screen>
			</para></listitem>

			<listitem><para>
			Directory permissions should be set for public read-write with the
			sticky bit set as shown:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">chmod a+twrx /var/spool/samba</userinput>
</screen>
		The purpose of setting the sticky bit is to prevent who does not own the temporary print file
		from being able to take control of it with the potential for devious misuse.
			</para></listitem>
		</itemizedlist>


		<note><para>
		<indexterm significance="normal"><primary>MIME</primary><secondary>raw</secondary></indexterm>
		<indexterm significance="normal"><primary>raw printing</primary></indexterm>
		On CUPS-enabled systems there is a facility to pass raw data directly to the printer without
		intermediate processing via CUPS print filters. Where use of this mode of operation is desired,
		it is necessary to configure a raw printing device. It is also necessary to enable the raw mime
		handler in the <filename moreinfo="none">/etc/mime.conv</filename> and <filename moreinfo="none">/etc/mime.types</filename>
		files. Refer to <link linkend="cups-raw"/>.
		</para></note>

		</sect3>

		<sect3>

		<title>Secure Read-Write File and Print Server</title>

		<para>
		We progress now from simple systems to a server that is slightly more complex.
		</para>

		<para>
		Our new server will require a public data storage area in which only authenticated
		users (i.e., those with a local account) can store files, as well as a home directory.
		There will be one printer that should be available for everyone to use.
		</para>

		<para>
		In this hypothetical environment (no espionage was conducted to obtain this data),
		the site is demanding a simple environment that is <emphasis>secure enough</emphasis>
		but not too difficult to use. 
		</para>

		<para>
		Site users will be Jack Baumbach, Mary Orville, and Amed Sehkah. Each will have
		a password (not shown in further examples). Mary will be the printer administrator and will
		own all files in the public share.
		</para>

		<para>
		This configuration will be based on <emphasis>user-level security</emphasis> that
		is the default, and for which the default is to store Microsoft Windows-compatible
		encrypted passwords in a file called <filename moreinfo="none">/etc/samba/smbpasswd</filename>.
		The default <filename moreinfo="none">smb.conf</filename> entry that makes this happen is
		<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PASSDBBACKEND" xlink:href="smb.conf.5.html#PASSDBBACKEND">passdb backend = smbpasswd, guest</link>. Since this is the default,
		it is not necessary to enter it into the configuration file. Note that the guest backend is
		added to the list of active passdb backends no matter whether it specified directly in Samba configuration
		file or not.
		</para>


		<procedure>
		<title>Installing the Secure Office Server</title>
			<step performance="required"><para>
		<indexterm significance="normal"><primary>office server</primary></indexterm>
			Add all users to the operating system:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">useradd -c "Jack Baumbach" -m -g users -p m0r3pa1n jackb</userinput>
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">useradd -c "Mary Orville" -m -g users -p secret maryo</userinput>
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">useradd -c "Amed Sehkah" -m -g users -p secret ameds</userinput>
</screen>
			</para></step>

			<step performance="required"><para>
			Configure the Samba <filename moreinfo="none">smb.conf</filename> file as shown in <link linkend="OfficeServer"/>.
			</para></step>

<example id="OfficeServer">
<title>Secure Office Server smb.conf</title>
<simplelist>
<member># Global parameters</member>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>workgroup</indexterm><parameter>workgroup = MIDEARTH</parameter></member>
<member><indexterm>netbios name</indexterm><parameter>netbios name = OLORIN</parameter></member>
<member><indexterm>printcap name</indexterm><parameter>printcap name = cups</parameter></member>
<member><indexterm>disable spoolss</indexterm><parameter>disable spoolss = Yes</parameter></member>
<member><indexterm>show add printer wizard</indexterm><parameter>show add printer wizard = No</parameter></member>
<member><indexterm>printing</indexterm><parameter>printing = cups</parameter></member>

<member> </member><member><parameter>[homes]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = Home Directories</parameter></member>
<member><indexterm>valid users</indexterm><parameter>valid users = %S</parameter></member>
<member><indexterm>read only</indexterm><parameter>read only = No</parameter></member>
<member><indexterm>browseable</indexterm><parameter>browseable = No</parameter></member>

<member> </member><member><parameter>[public]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = Data</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /export</parameter></member>
<member><indexterm>force user</indexterm><parameter>force user = maryo</parameter></member>
<member><indexterm>force group</indexterm><parameter>force group = users</parameter></member>
<member><indexterm>read only</indexterm><parameter>read only = No</parameter></member>

<member> </member><member><parameter>[printers]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = All Printers</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /var/spool/samba</parameter></member>
<member><indexterm>printer admin</indexterm><parameter>printer admin = root, maryo</parameter></member>
<member><indexterm>create mask</indexterm><parameter>create mask = 0600</parameter></member>
<member><indexterm>guest ok</indexterm><parameter>guest ok = Yes</parameter></member>
<member><indexterm>printable</indexterm><parameter>printable = Yes</parameter></member>
<member><indexterm>use client driver</indexterm><parameter>use client driver = Yes</parameter></member>
<member><indexterm>browseable</indexterm><parameter>browseable = No</parameter></member>
</simplelist>
</example>

			<step performance="required"><para>
			Initialize the Microsoft Windows password database with the new users:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">smbpasswd -a root</userinput>
New SMB password: <userinput moreinfo="none">bigsecret</userinput>
Reenter smb password: <userinput moreinfo="none">bigsecret</userinput>
Added user root.

<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">smbpasswd -a jackb</userinput>
New SMB password: <userinput moreinfo="none">m0r3pa1n</userinput>
Retype new SMB password: <userinput moreinfo="none">m0r3pa1n</userinput>
Added user jackb.

<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">smbpasswd -a maryo</userinput>
New SMB password: <userinput moreinfo="none">secret</userinput>
Reenter smb password: <userinput moreinfo="none">secret</userinput>
Added user maryo.

<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">smbpasswd -a ameds</userinput>
New SMB password: <userinput moreinfo="none">mysecret</userinput>
Reenter smb password: <userinput moreinfo="none">mysecret</userinput>
Added user ameds.
</screen>
			</para></step>

			<step performance="required"><para>
			Install printer using the CUPS Web interface. Make certain that all
			printers that will be shared with Microsoft Windows clients are installed
			as raw printing devices.
			</para></step>

			<step performance="required"><para>
			Start Samba using the operating system administrative interface.
			Alternately, this can be done manually by executing:
			<indexterm significance="normal"><primary>smbd</primary></indexterm>
			<indexterm significance="normal"><primary>nmbd</primary></indexterm>
			<indexterm significance="normal"><primary>starting samba</primary><secondary>smbd</secondary></indexterm>
			<indexterm significance="normal"><primary>starting samba</primary><secondary>nmbd</secondary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none"> nmbd; smbd;</userinput>
</screen>
			Both applications automatically execute as daemons. Those who are paranoid about
			maintaining control can add the <constant>-D</constant> flag to coerce them to start
			up in daemon mode.
			</para></step>

			<step performance="required"><para>
			Configure the <filename moreinfo="none">/export</filename> directory:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">mkdir /export</userinput>
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">chown maryo.users /export</userinput>
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">chmod u=rwx,g=rwx,o-rwx /export</userinput>
</screen>
			</para></step>

			<step performance="required"><para>
			Check that Samba is running correctly:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">smbclient -L localhost -U%</userinput>
Domain=[MIDEARTH] OS=[UNIX] Server=[Samba-3.0.20]

Sharename      Type      Comment
---------      ----      -------
public         Disk      Data
IPC$           IPC       IPC Service (Samba-3.0.20)
ADMIN$         IPC       IPC Service (Samba-3.0.20)
hplj4          Printer   hplj4

Server               Comment
---------            -------
OLORIN               Samba-3.0.20

Workgroup            Master
---------            -------
MIDEARTH             OLORIN
</screen>
			The following error message indicates that Samba was not running:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> smbclient -L olorin -U%
Error connecting to 192.168.1.40 (Connection refused)
Connection to olorin failed
</screen>
			</para></step>

			<step performance="required"><para>
			Connect to OLORIN as maryo:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">smbclient //olorin/maryo -Umaryo%secret</userinput>
OS=[UNIX] Server=[Samba-3.0.20]
smb: \&gt; <userinput moreinfo="none">dir</userinput>
.                              D        0  Sat Jun 21 10:58:16 2003
..                             D        0  Sat Jun 21 10:54:32 2003
Documents                      D        0  Fri Apr 25 13:23:58 2003
DOCWORK                        D        0  Sat Jun 14 15:40:34 2003
OpenOffice.org                 D        0  Fri Apr 25 13:55:16 2003
.bashrc                        H     1286  Fri Apr 25 13:23:58 2003
.netscape6                    DH        0  Fri Apr 25 13:55:13 2003
.mozilla                      DH        0  Wed Mar  5 11:50:50 2003
.kermrc                        H      164  Fri Apr 25 13:23:58 2003
.acrobat                      DH        0  Fri Apr 25 15:41:02 2003

		55817 blocks of size 524288. 34725 blocks available
smb: \&gt; <userinput moreinfo="none">q</userinput>
</screen>
			</para></step>
		</procedure>

			<para>
			By now you should be getting the hang of configuration basics. Clearly, it is time to
			explore slightly more complex examples. For the remainder of this chapter we abbreviate
			instructions, since there are previous examples.
			</para>

		</sect3>

	</sect2>

	<sect2>
	<title>Domain Member Server</title>

	<para>
	<indexterm significance="normal"><primary>Server Type</primary><secondary>Domain Member</secondary></indexterm>
	In this instance we consider the simplest server configuration we can get away with
	to make an accounting department happy. Let's be warned, the users are accountants and they
	do have some nasty demands. There is a budget for only one server for this department.
	</para>

	<para>
	The network is managed by an internal Information Services Group (ISG), to which we belong.
	Internal politics are typical of a medium-sized organization; Human Resources is of the
	opinion that they run the ISG because they are always adding and disabling users. Also,
	departmental managers have to fight tooth and nail to gain basic network resources access for
	their staff. Accounting is different, though, they get exactly what they want. So this should
	set the scene.
	</para>

	<para>
	We use the users from the last example. The accounting department
	has a general printer that all departmental users may use. There is also a check printer
	that may be used only by the person who has authority to print checks. The chief financial
	officer (CFO) wants that printer to be completely restricted and for it to be located in the
	private storage area in her office. It therefore must be a network printer.
	</para>

	<para>
	The accounting department uses an accounting application called <emphasis>SpytFull</emphasis>
	that must be run from a central application server. The software is licensed to run only off
	one server, there are no workstation components, and it is run off a mapped share. The data
	store is in a UNIX-based SQL backend. The UNIX gurus look after that, so this is not our
	problem.
	</para>

	<para>
	The accounting department manager (maryo) wants a general filing system as well as a separate
	file storage area for form letters (nastygrams). The form letter area should be read-only to
	all accounting staff except the manager. The general filing system has to have a structured
	layout with a general area for all staff to store general documents as well as a separate
	file area for each member of her team that is private to that person, but she wants full
	access to all areas. Users must have a private home share for personal work-related files
	and for materials not related to departmental operations.
	</para>
	
		<sect3>
		<title>Example Configuration</title>
		
		<para>
		The server <emphasis>valinor</emphasis> will be a member server of the company domain.
		Accounting will have only a local server. User accounts will be on the domain controllers,
		as will desktop profiles and all network policy files.
		</para>

		<procedure>
			<step performance="required"><para>
			Do not add users to the UNIX/Linux server; all of this will run off the
			central domain.
			</para></step>

			<step performance="required"><para>
			Configure <filename moreinfo="none">smb.conf</filename> according to <link linkend="fast-member-server">Member server smb.conf
			(globals)</link> and <link linkend="fast-memberserver-shares">Member server smb.conf (shares
			and services)</link>.
			</para></step>

<example id="fast-member-server">
<title>Member Server smb.conf (Globals)</title>
<simplelist>
<member># Global parameters</member>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>workgroup</indexterm><parameter>workgroup = MIDEARTH</parameter></member>
<member><indexterm>netbios name</indexterm><parameter>netbios name = VALINOR</parameter></member>
<member><indexterm>security</indexterm><parameter>security = DOMAIN</parameter></member>
<member><indexterm>printcap name</indexterm><parameter>printcap name = cups</parameter></member>
<member><indexterm>disable spoolss</indexterm><parameter>disable spoolss = Yes</parameter></member>
<member><indexterm>show add printer wizard</indexterm><parameter>show add printer wizard = No</parameter></member>
<member><indexterm>idmap uid</indexterm><parameter>idmap uid = 15000-20000</parameter></member>
<member><indexterm>idmap gid</indexterm><parameter>idmap gid = 15000-20000</parameter></member>
<member><indexterm>winbind use default domain</indexterm><parameter>winbind use default domain = Yes</parameter></member>
<member><indexterm>printing</indexterm><parameter>printing = cups</parameter></member>
</simplelist>
</example>

<example id="fast-memberserver-shares">
<title>Member Server smb.conf (Shares and Services)</title>
<simplelist>
<member> </member><member><parameter>[homes]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = Home Directories</parameter></member>
<member><indexterm>valid users</indexterm><parameter>valid users = %S</parameter></member>
<member><indexterm>read only</indexterm><parameter>read only = No</parameter></member>
<member><indexterm>browseable</indexterm><parameter>browseable = No</parameter></member>

<member> </member><member><parameter>[spytfull]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = Accounting Application Only</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /export/spytfull</parameter></member>
<member><indexterm>valid users</indexterm><parameter>valid users = @Accounts</parameter></member>
<member><indexterm>admin users</indexterm><parameter>admin users = maryo</parameter></member>
<member><indexterm>read only</indexterm><parameter>read only = Yes</parameter></member>

<member> </member><member><parameter>[public]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = Data</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /export/public</parameter></member>
<member><indexterm>read only</indexterm><parameter>read only = No</parameter></member>

<member> </member><member><parameter>[printers]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = All Printers</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /var/spool/samba</parameter></member>
<member><indexterm>printer admin</indexterm><parameter>printer admin = root, maryo</parameter></member>
<member><indexterm>create mask</indexterm><parameter>create mask = 0600</parameter></member>
<member><indexterm>guest ok</indexterm><parameter>guest ok = Yes</parameter></member>
<member><indexterm>printable</indexterm><parameter>printable = Yes</parameter></member>
<member><indexterm>use client driver</indexterm><parameter>use client driver = Yes</parameter></member>
<member><indexterm>browseable</indexterm><parameter>browseable = No</parameter></member>
</simplelist>
</example>

			<step performance="required"><para>
			<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary></indexterm>
			Join the domain. Note: Do not start Samba until this step has been completed!
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">net rpc join -Uroot%'bigsecret'</userinput>
Joined domain MIDEARTH.
</screen>
			</para></step>

			<step performance="required"><para>
			Make absolutely certain that you disable (shut down) the <literal>nscd</literal>
			daemon on any system on which <literal>winbind</literal> is configured to run.
			</para></step>

			<step performance="required"><para>
			Start Samba following the normal method for your operating system platform.
			If you wish to do this manually, execute as root:
			<indexterm significance="normal"><primary>smbd</primary></indexterm>
			<indexterm significance="normal"><primary>nmbd</primary></indexterm>
			<indexterm significance="normal"><primary>winbindd</primary></indexterm>
			<indexterm significance="normal"><primary>starting samba</primary><secondary>smbd</secondary></indexterm>
			<indexterm significance="normal"><primary>starting samba</primary><secondary>nmbd</secondary></indexterm>
			<indexterm significance="normal"><primary>starting samba</primary><secondary>winbindd</secondary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">nmbd; smbd; winbindd;</userinput>
</screen>
			</para></step>

			<step performance="required"><para>
			Configure the name service switch (NSS) control file on your system to resolve user and group names
			via winbind. Edit the following lines in <filename moreinfo="none">/etc/nsswitch.conf</filename>:
<programlisting format="linespecific">
passwd: files winbind
group:  files winbind
hosts:  files dns winbind
</programlisting>
			</para></step>

			<step performance="required"><para>
			Set the password for <literal>wbinfo</literal> to use:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">wbinfo --set-auth-user=root%'bigsecret'</userinput>
</screen>
			</para></step>

			<step performance="required"><para>
			Validate that domain user and group credentials can be correctly resolved by executing:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">wbinfo -u</userinput>
MIDEARTH\maryo
MIDEARTH\jackb
MIDEARTH\ameds
...
MIDEARTH\root

<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">wbinfo -g</userinput>
MIDEARTH\Domain Users
MIDEARTH\Domain Admins
MIDEARTH\Domain Guests
...
MIDEARTH\Accounts
</screen>
			</para></step>

			<step performance="required"><para>
			Check that <literal>winbind</literal> is working. The following demonstrates correct
			username resolution via the <literal>getent</literal> system utility:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">getent passwd maryo</userinput>
maryo:x:15000:15003:Mary Orville:/home/MIDEARTH/maryo:/bin/false
</screen>
			</para></step>

			<step performance="required"><para>
			A final test that we have this under control might be reassuring:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">touch /export/a_file</userinput>
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">chown maryo /export/a_file</userinput>
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">ls -al /export/a_file</userinput>
...
-rw-r--r--    1 maryo    users       11234 Jun 21 15:32 a_file
...

<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">rm /export/a_file</userinput>
</screen>
			</para></step>

			<step performance="required"><para>
			Configuration is now mostly complete, so this is an opportune time
			to configure the directory structure for this site:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">mkdir -p /export/{spytfull,public}</userinput>
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">chmod ug=rwxS,o=x /export/{spytfull,public}</userinput>
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">chown maryo.Accounts /export/{spytfull,public}</userinput>
</screen>
			</para></step>
		</procedure>

		</sect3>

	</sect2>

	<sect2>
	<title>Domain Controller</title>


	<para>
	<indexterm significance="normal"><primary>Server Type</primary><secondary>Domain Controller</secondary></indexterm>
	For the remainder of this chapter the focus is on the configuration of domain control.
	The examples that follow are for two implementation strategies. Remember, our objective is
	to create a simple but working solution. The remainder of this book should help to highlight
	opportunity for greater functionality and the complexity that goes with it.
	</para>

	<para>
	A domain controller configuration can be achieved with a simple configuration using the new
	tdbsam password backend. This type of configuration is good for small
	offices, but has limited scalability (cannot be replicated), and performance can be expected
	to fall as the size and complexity of the domain increases.
	</para>

	<para>
	The use of tdbsam is best limited to sites that do not need
	more than a Primary Domain Controller (PDC). As the size of a domain grows the need
	for additional domain controllers becomes apparent. Do not attempt to under-resource
	a Microsoft Windows network environment; domain controllers provide essential
	authentication services. The following are symptoms of an under-resourced domain control
	environment:
	</para>

	<itemizedlist>	
		<listitem><para>
		 Domain logons intermittently fail.
		</para></listitem>

		<listitem><para>
		File access on a domain member server intermittently fails, giving a permission denied
		error message.
		</para></listitem>
	</itemizedlist>

	<para>
	A more scalable domain control authentication backend option might use
	Microsoft Active Directory or an LDAP-based backend. Samba-3 provides
	for both options as a domain member server. As a PDC, Samba-3 is not able to provide
	an exact alternative to the functionality that is available with Active Directory.
	Samba-3 can provide a scalable LDAP-based PDC/BDC solution.
	</para>

	<para>
	The tdbsam authentication backend provides no facility to replicate
	the contents of the database, except by external means (i.e., there is no self-contained protocol
	in Samba-3 for Security Account Manager database [SAM] replication).
	</para>

	<note><para>
	If you need more than one domain controller, do not use a tdbsam authentication backend.
	</para></note>

		<sect3>
		<title>Example: Engineering Office</title>

		<para>
		The engineering office network server we present here is designed to demonstrate use
		of the new tdbsam password backend. The tdbsam
		facility is new to Samba-3. It is designed to provide many user and machine account controls
		that are possible with Microsoft Windows NT4. It is safe to use this in smaller networks.
		</para>

		<procedure>
			<step performance="required"><para>
			A working PDC configuration using the tdbsam
			password backend can be found in <link linkend="fast-engoffice-global">Engineering Office smb.conf
			(globals)</link> together with <link linkend="fast-engoffice-shares">Engineering Office smb.conf
			(shares and services)</link>:
			<indexterm significance="normal"><primary>pdbedit</primary></indexterm>
			</para></step>

<example id="fast-engoffice-global">
<title>Engineering Office smb.conf (globals)</title>
<simplelist>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>workgroup</indexterm><parameter>workgroup = MIDEARTH</parameter></member>
<member><indexterm>netbios name</indexterm><parameter>netbios name = FRODO</parameter></member>
<member><indexterm>passdb backend</indexterm><parameter>passdb backend = tdbsam</parameter></member>
<member><indexterm>printcap name</indexterm><parameter>printcap name = cups</parameter></member>
<member><indexterm>add user script</indexterm><parameter>add user script = /usr/sbin/useradd -m %u</parameter></member>
<member><indexterm>delete user script</indexterm><parameter>delete user script = /usr/sbin/userdel -r %u</parameter></member>
<member><indexterm>add group script</indexterm><parameter>add group script = /usr/sbin/groupadd %g</parameter></member>
<member><indexterm>delete group script</indexterm><parameter>delete group script = /usr/sbin/groupdel %g</parameter></member>
<member><indexterm>add user to group script</indexterm><parameter>add user to group script = /usr/sbin/groupmod -A %u %g</parameter></member>
<member><indexterm>delete user from group script</indexterm><parameter>delete user from group script = /usr/sbin/groupmod -R %u %g</parameter></member>
<member><indexterm>add machine script</indexterm><parameter>add machine script = /usr/sbin/useradd -s /bin/false -d /var/lib/nobody %u</parameter></member>
<member># Note: The following specifies the default logon script.</member>
<member># Per user logon scripts can be specified in the user account using pdbedit </member>
<member><indexterm>logon script</indexterm><parameter>logon script = scripts\logon.bat</parameter></member>
<member># This sets the default profile path. Set per user paths with pdbedit</member>
<member><indexterm>logon path</indexterm><parameter>logon path = \\%L\Profiles\%U</parameter></member>
<member><indexterm>logon drive</indexterm><parameter>logon drive = H:</parameter></member>
<member><indexterm>logon home</indexterm><parameter>logon home = \\%L\%U</parameter></member>
<member><indexterm>domain logons</indexterm><parameter>domain logons = Yes</parameter></member>
<member><indexterm>os level</indexterm><parameter>os level = 35</parameter></member>
<member><indexterm>preferred master</indexterm><parameter>preferred master = Yes</parameter></member>
<member><indexterm>domain master</indexterm><parameter>domain master = Yes</parameter></member>
<member><indexterm>idmap uid</indexterm><parameter>idmap uid = 15000-20000</parameter></member>
<member><indexterm>idmap gid</indexterm><parameter>idmap gid = 15000-20000</parameter></member>
<member><indexterm>printing</indexterm><parameter>printing = cups</parameter></member>
</simplelist>
</example>

<example id="fast-engoffice-shares">
<title>Engineering Office smb.conf (shares and services)</title>
<simplelist>
<member> </member><member><parameter>[homes]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = Home Directories</parameter></member>
<member><indexterm>valid users</indexterm><parameter>valid users = %S</parameter></member>
<member><indexterm>read only</indexterm><parameter>read only = No</parameter></member>
<member><indexterm>browseable</indexterm><parameter>browseable = No</parameter></member>

<member># Printing auto-share (makes printers available thru CUPS)</member>
<member> </member><member><parameter>[printers]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = All Printers</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /var/spool/samba</parameter></member>
<member><indexterm>printer admin</indexterm><parameter>printer admin = root, maryo</parameter></member>
<member><indexterm>create mask</indexterm><parameter>create mask = 0600</parameter></member>
<member><indexterm>guest ok</indexterm><parameter>guest ok = Yes</parameter></member>
<member><indexterm>printable</indexterm><parameter>printable = Yes</parameter></member>
<member><indexterm>browseable</indexterm><parameter>browseable = No</parameter></member>

<member> </member><member><parameter>[print$]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = Printer Drivers Share</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /var/lib/samba/drivers</parameter></member>
<member><indexterm>write list</indexterm><parameter>write list = maryo, root</parameter></member>
<member><indexterm>printer admin</indexterm><parameter>printer admin = maryo, root</parameter></member>

<member># Needed to support domain logons</member>
<member> </member><member><parameter>[netlogon]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = Network Logon Service</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /var/lib/samba/netlogon</parameter></member>
<member><indexterm>admin users</indexterm><parameter>admin users = root, maryo</parameter></member>
<member><indexterm>guest ok</indexterm><parameter>guest ok = Yes</parameter></member>
<member><indexterm>browseable</indexterm><parameter>browseable = No</parameter></member>

<member># For profiles to work, create a user directory under the path</member>
<member>#  shown. i.e., mkdir -p /var/lib/samba/profiles/maryo</member>
<member> </member><member><parameter>[Profiles]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = Roaming Profile Share</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /var/lib/samba/profiles</parameter></member>
<member><indexterm>read only</indexterm><parameter>read only = No</parameter></member>
<member><indexterm>profile acls</indexterm><parameter>profile acls = Yes</parameter></member>

<member># Other resource (share/printer) definitions would follow below.</member>
</simplelist>
</example>

			<step performance="required"><para>
			Create UNIX group accounts as needed using a suitable operating system tool:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">groupadd ntadmins</userinput>
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">groupadd designers</userinput>
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">groupadd engineers</userinput>
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">groupadd qateam</userinput>
</screen>
			</para></step>

			<step performance="required"><para>
			Create user accounts on the system using the appropriate tool
			provided with the operating system. Make sure all user home directories
			are created also. Add users to groups as required for access control
			on files, directories, printers, and as required for use in the Samba
			environment.
			</para></step>


			<step performance="required"><para>
			<indexterm significance="normal"><primary>net</primary><secondary>groupmap</secondary></indexterm>
			<indexterm significance="normal"><primary>initGroups.sh</primary></indexterm>
			Assign each of the UNIX groups to NT groups by executing this shell script
			(You could name the script <filename moreinfo="none">initGroups.sh</filename>):
<screen format="linespecific">
#!/bin/bash
#### Keep this as a shell script for future re-use
			
# First assign well known groups
net groupmap add ntgroup="Domain Admins" unixgroup=ntadmins rid=512 type=d
net groupmap add ntgroup="Domain Users"  unixgroup=users rid=513 type=
net groupmap add ntgroup="Domain Guests" unixgroup=nobody rid=514 type=d

# Now for our added Domain Groups
net groupmap add ntgroup="Designers" unixgroup=designers type=d
net groupmap add ntgroup="Engineers" unixgroup=engineers type=d
net groupmap add ntgroup="QA Team"   unixgroup=qateam    type=d
</screen>
			</para></step>

			<step performance="required"><para>
			Create the <filename moreinfo="none">scripts</filename> directory for use in the 
			<parameter>[NETLOGON]</parameter> share:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">mkdir -p /var/lib/samba/netlogon/scripts</userinput>
</screen>
			Place the logon scripts that will be used (batch or cmd scripts)
			in this directory.
			</para></step>
		</procedure>

		<para>
		The above configuration provides a functional PDC
		system to which must be added file shares and printers as required.
		</para>

		</sect3>

		<sect3>
		<title>A Big Organization</title>

		<para>
		In this section we finally get to review in brief a Samba-3 configuration that
		uses a Lightweight Directory Access (LDAP)-based authentication backend. The
		main reasons for this choice are to provide the ability to host primary
		and Backup Domain Control (BDC), as well as to enable a higher degree of
		scalability to meet the needs of a very distributed environment.
		</para>

			<sect4>
			<title>The Primary Domain Controller</title>

			<para>
			This is an example of a minimal configuration to run a Samba-3 PDC
			using an LDAP authentication backend. It is assumed that the operating system
			has been correctly configured.
			</para>

			<para>
			The Idealx scripts (or equivalent) are needed to manage LDAP-based POSIX and/or
			SambaSamAccounts. The Idealx scripts may be downloaded from the <ulink url="http://www.idealx.org">
			Idealx</ulink> Web site. They may also be obtained from the Samba tarball. Linux
			distributions tend to install the Idealx scripts in the 
			<filename moreinfo="none">/usr/share/doc/packages/sambaXXXXXX/examples/LDAP/smbldap-tools</filename> directory.
			Idealx scripts version <constant>smbldap-tools-0.9.1</constant> are known to work well.
			</para>

			<procedure>
				<step performance="required"><para>
				Obtain from the Samba sources <filename moreinfo="none">~/examples/LDAP/samba.schema</filename>
				and copy it to the <filename moreinfo="none">/etc/openldap/schema/</filename> directory.
				</para></step>

				<step performance="required"><para>
				Set up the LDAP server. This example is suitable for OpenLDAP 2.1.x.
				The <filename moreinfo="none">/etc/openldap/slapd.conf</filename> file.
				<indexterm significance="normal"><primary>/etc/openldap/slapd.conf</primary></indexterm>
<title>Example slapd.conf File</title>
<screen format="linespecific">
# Note commented out lines have been removed
include         /etc/openldap/schema/core.schema
include         /etc/openldap/schema/cosine.schema
include         /etc/openldap/schema/inetorgperson.schema
include         /etc/openldap/schema/nis.schema
include         /etc/openldap/schema/samba.schema

pidfile         /var/run/slapd/slapd.pid
argsfile        /var/run/slapd/slapd.args

database        bdb
suffix          "dc=quenya,dc=org"
rootdn          "cn=Manager,dc=quenya,dc=org"
rootpw          {SSHA}06qDkonA8hk6W6SSnRzWj0/pBcU3m0/P
# The password for the above is 'nastyon3'

directory     /var/lib/ldap

index   objectClass     eq
index cn                      pres,sub,eq
index sn                      pres,sub,eq
index uid                     pres,sub,eq
index displayName             pres,sub,eq
index uidNumber               eq
index gidNumber               eq
index memberUid               eq
index   sambaSID              eq
index   sambaPrimaryGroupSID  eq
index   sambaDomainName       eq
index   default               sub
</screen>
				</para></step>

				<step performance="required"><para>
				Create the following file <filename moreinfo="none">initdb.ldif</filename>:
				<indexterm significance="normal"><primary>initdb.ldif</primary></indexterm>
<programlisting format="linespecific">
# Organization for SambaXP Demo
dn: dc=quenya,dc=org
objectclass: dcObject
objectclass: organization
dc: quenya
o: SambaXP Demo
description: The SambaXP Demo LDAP Tree

# Organizational Role for Directory Management
dn: cn=Manager,dc=quenya,dc=org
objectclass: organizationalRole
cn: Manager
description: Directory Manager

# Setting up the container for users
dn: ou=People, dc=quenya, dc=org
objectclass: top
objectclass: organizationalUnit
ou: People

# Set up an admin handle for People OU
dn: cn=admin, ou=People, dc=quenya, dc=org
cn: admin
objectclass: top
objectclass: organizationalRole
objectclass: simpleSecurityObject
userPassword: {SSHA}0jBHgQ1vp4EDX2rEMMfIudvRMJoGwjVb
# The password for above is 'mordonL8'
</programlisting>
				</para></step>

				<step performance="required"><para>
				Load the initial data above into the LDAP database:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">slapadd -v -l initdb.ldif</userinput>
</screen>
				</para></step>

				<step performance="required"><para>
				Start the LDAP server using the appropriate tool or method for
				the operating system platform on which it is installed.
				</para></step>

				<step performance="required"><para>
				Install the Idealx script files in the <filename moreinfo="none">/usr/local/sbin</filename> directory,
				then configure the smbldap_conf.pm file to match your system configuration.
				</para></step>

				<step performance="required"><para>
				The <filename moreinfo="none">smb.conf</filename> file that drives this backend can be found in example <link linkend="fast-ldap">LDAP backend smb.conf for PDC</link>. Add additional stanzas
				as required.
				</para></step>

<example id="fast-ldap">
<title>LDAP backend smb.conf for PDC</title>
<simplelist>
<member># Global parameters</member>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>workgroup</indexterm><parameter>workgroup = MIDEARTH</parameter></member>
<member><indexterm>netbios name</indexterm><parameter>netbios name = FRODO</parameter></member>
<member><indexterm>passdb backend</indexterm><parameter>passdb backend = ldapsam:ldap://localhost</parameter></member>
<member><indexterm>username map</indexterm><parameter>username map = /etc/samba/smbusers</parameter></member>
<member><indexterm>printcap name</indexterm><parameter>printcap name = cups</parameter></member>
<member><indexterm>add user script</indexterm><parameter>add user script = /usr/local/sbin/smbldap-useradd -m '%u'</parameter></member>
<member><indexterm>delete user script</indexterm><parameter>delete user script = /usr/local/sbin/smbldap-userdel %u</parameter></member>
<member><indexterm>add group script</indexterm><parameter>add group script = /usr/local/sbin/smbldap-groupadd -p '%g'</parameter></member>
<member><indexterm>delete group script</indexterm><parameter>delete group script = /usr/local/sbin/smbldap-groupdel '%g'</parameter></member>
<member><indexterm>add user to group script</indexterm><parameter>add user to group script = /usr/local/sbin/smbldap-groupmod -m '%u' '%g'</parameter></member>
<member><indexterm>delete user from group script</indexterm><parameter>delete user from group script = /usr/local/sbin/smbldap-groupmod -x '%u' '%g'</parameter></member>
<member><indexterm>set primary group script</indexterm><parameter>set primary group script = /usr/local/sbin/smbldap-usermod -g '%g' '%u'</parameter></member>
<member><indexterm>add machine script</indexterm><parameter>add machine script = /usr/local/sbin/smbldap-useradd -w '%u'</parameter></member>
<member><indexterm>logon script</indexterm><parameter>logon script = scripts\logon.bat</parameter></member>
<member><indexterm>logon path</indexterm><parameter>logon path = \\%L\Profiles\%U</parameter></member>
<member><indexterm>logon drive</indexterm><parameter>logon drive = H:</parameter></member>
<member><indexterm>logon home</indexterm><parameter>logon home = \\%L\%U</parameter></member>
<member><indexterm>domain logons</indexterm><parameter>domain logons = Yes</parameter></member>
<member><indexterm>os level</indexterm><parameter>os level = 35</parameter></member>
<member><indexterm>preferred master</indexterm><parameter>preferred master = Yes</parameter></member>
<member><indexterm>domain master</indexterm><parameter>domain master = Yes</parameter></member>
<member><indexterm>ldap suffix</indexterm><parameter>ldap suffix = dc=quenya,dc=org</parameter></member>
<member><indexterm>ldap machine suffix</indexterm><parameter>ldap machine suffix = ou=People</parameter></member>
<member><indexterm>ldap user suffix</indexterm><parameter>ldap user suffix = ou=People</parameter></member>
<member><indexterm>ldap group suffix</indexterm><parameter>ldap group suffix = ou=People</parameter></member>
<member><indexterm>ldap idmap suffix</indexterm><parameter>ldap idmap suffix = ou=People</parameter></member>
<member><indexterm>ldap admin dn</indexterm><parameter>ldap admin dn = cn=Manager</parameter></member>
<member><indexterm>ldap ssl</indexterm><parameter>ldap ssl = no</parameter></member>
<member><indexterm>ldap passwd sync</indexterm><parameter>ldap passwd sync = Yes</parameter></member>
<member><indexterm>idmap uid</indexterm><parameter>idmap uid = 15000-20000</parameter></member>
<member><indexterm>idmap gid</indexterm><parameter>idmap gid = 15000-20000</parameter></member>
<member><indexterm>printing</indexterm><parameter>printing = cups</parameter></member>
</simplelist>
</example>

				<step performance="required"><para>
				Add the LDAP password to the <filename moreinfo="none">secrets.tdb</filename> file so Samba can update
				the LDAP database:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">smbpasswd -w mordonL8</userinput>
</screen>
				</para></step>

				<step performance="required"><para>
				Add users and groups as required. Users and groups added using Samba tools
				will automatically be added to both the LDAP backend and the operating
				system as required.
				</para></step>

			</procedure>

			</sect4>

			<sect4>
			<title>Backup Domain Controller</title>

			<para>
			<link linkend="fast-bdc"/> shows the example configuration for the BDC. Note that
			the <filename moreinfo="none">smb.conf</filename> file does not specify the smbldap-tools scripts <?latex --- ?> they are
			not needed on a BDC. Add additional stanzas for shares and printers as required.
			</para>

			<procedure>
				<step performance="required"><para>
				Decide if the BDC should have its own LDAP server or not. If the BDC is to be
				the LDAP server, change the following <filename moreinfo="none">smb.conf</filename> as indicated. The default
				configuration in <link linkend="fast-bdc">Remote LDAP BDC smb.conf</link>
				uses a central LDAP server.
				</para></step>

<example id="fast-bdc">
<title>Remote LDAP BDC smb.conf</title>
<simplelist>
<member># Global parameters</member>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>workgroup</indexterm><parameter>workgroup = MIDEARTH</parameter></member>
<member><indexterm>netbios name</indexterm><parameter>netbios name = GANDALF</parameter></member>
<member><indexterm>passdb backend</indexterm><parameter>passdb backend = ldapsam:ldap://frodo.quenya.org</parameter></member>
<member><indexterm>username map</indexterm><parameter>username map = /etc/samba/smbusers</parameter></member>
<member><indexterm>printcap name</indexterm><parameter>printcap name = cups</parameter></member>
<member><indexterm>logon script</indexterm><parameter>logon script = scripts\logon.bat</parameter></member>
<member><indexterm>logon path</indexterm><parameter>logon path = \\%L\Profiles\%U</parameter></member>
<member><indexterm>logon drive</indexterm><parameter>logon drive = H:</parameter></member>
<member><indexterm>logon home</indexterm><parameter>logon home = \\%L\%U</parameter></member>
<member><indexterm>domain logons</indexterm><parameter>domain logons = Yes</parameter></member>
<member><indexterm>os level</indexterm><parameter>os level = 33</parameter></member>
<member><indexterm>preferred master</indexterm><parameter>preferred master = Yes</parameter></member>
<member><indexterm>domain master</indexterm><parameter>domain master = No</parameter></member>
<member><indexterm>ldap suffix</indexterm><parameter>ldap suffix = dc=quenya,dc=org</parameter></member>
<member><indexterm>ldap machine suffix</indexterm><parameter>ldap machine suffix = ou=People</parameter></member>
<member><indexterm>ldap user suffix</indexterm><parameter>ldap user suffix = ou=People</parameter></member>
<member><indexterm>ldap group suffix</indexterm><parameter>ldap group suffix = ou=People</parameter></member>
<member><indexterm>ldap idmap suffix</indexterm><parameter>ldap idmap suffix = ou=People</parameter></member>
<member><indexterm>ldap admin dn</indexterm><parameter>ldap admin dn = cn=Manager</parameter></member>
<member><indexterm>ldap ssl</indexterm><parameter>ldap ssl = no</parameter></member>
<member><indexterm>ldap passwd sync</indexterm><parameter>ldap passwd sync = Yes</parameter></member>
<member><indexterm>idmap uid</indexterm><parameter>idmap uid = 15000-20000</parameter></member>
<member><indexterm>idmap gid</indexterm><parameter>idmap gid = 15000-20000</parameter></member>
<member><indexterm>printing</indexterm><parameter>printing = cups</parameter></member>
</simplelist>
</example>

				<step performance="required"><para>
				Configure the NETLOGON and PROFILES directory as for the PDC in <link linkend="fast-bdc"/>.
				</para></step>
			</procedure>

			</sect4>

		</sect3>

	</sect2>

</sect1>

</chapter>
	<?latex \cleardoublepage ?>

</part>

<part id="type">
<title>Server Configuration Basics</title>
<partintro>
<title>First Steps in Server Configuration</title>

<para>
Samba can operate in various modes within SMB networks. This HOWTO section contains information on
configuring Samba to function as the type of server your network requires. Please read this
section carefully.
</para>

</partintro>

	<?latex \cleardoublepage ?>
	<chapter id="ServerType">
<chapterinfo>
	
<author>
	<firstname>Andrew</firstname><surname>Tridgell</surname>
	<affiliation>
		<orgname>Samba Team</orgname>
		<address format="linespecific"><email>tridge@samba.org</email></address>
	</affiliation>
</author>
	<author>
<firstname>Jelmer</firstname><othername>R.</othername><surname>Vernooij</surname><othername>R.</othername>
<affiliation>
	<orgname>The Samba Team</orgname>
	<address format="linespecific"><email>jelmer@samba.org</email></address>
</affiliation></author>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
</chapterinfo>

<title>Server Types and Security Modes</title>

<para>
<indexterm significance="normal"><primary>migrate</primary></indexterm>
<indexterm significance="normal"><primary>security mode</primary></indexterm>
This chapter provides information regarding the types of server that Samba may be configured to be. A
Microsoft network administrator who wishes to migrate to or use Samba will want to know the meaning, within a
Samba context, of terms familiar to the MS Windows administrator. This means that it is essential also to
define how critical security modes function before we get into the details of how to configure the server
itself.
</para>

<para>
This chapter provides an overview of the security modes of which Samba is capable and how they relate to MS
Windows servers and clients.
</para>

<para>
A question often asked is, <quote>Why would I want to use Samba?</quote> Most chapters contain a section that
highlights features and benefits. We hope that the information provided will help to answer this question. Be
warned though, we want to be fair and reasonable, so not all features are positive toward Samba. The benefit
may be on the side of our competition.
</para>

<sect1>
<title>Features and Benefits</title>

<para>
Two men were walking down a dusty road, when one suddenly kicked up a small red stone. It
hurt his toe and lodged in his sandal. He took the stone out and cursed it with a passion
and fury befitting his anguish. The other looked at the stone and said, <quote>This is a garnet.
I can turn that into a precious gem and some day it will make a princess very happy!</quote>
</para>

<para>
The moral of this tale: Two men, two very different perspectives regarding the same stone.
Like it or not, Samba is like that stone. Treat it the right way and it can bring great
pleasure, but if you are forced to use it and have no time for its secrets, then it can be
a source of discomfort.
</para>

<para>
<indexterm significance="normal"><primary>UNIX</primary><secondary>server</secondary></indexterm>
<indexterm significance="normal"><primary>interoperability</primary></indexterm>
Samba started out as a project that sought to provide interoperability for MS Windows 3.x
clients with a UNIX server. It has grown up a lot since its humble beginnings and now provides
features and functionality fit for large-scale deployment. It also has some warts. In sections
like this one, we tell of both.
</para>

<para>
So, what are the benefits of the features mentioned in this chapter?
</para>

<itemizedlist>
	<listitem><para>
	<indexterm significance="normal"><primary>domain</primary><secondary>controller</secondary></indexterm>
	Samba-3 can replace an MS Windows NT4 domain controller.
	</para></listitem>

	<listitem><para>
	<indexterm significance="normal"><primary>active directory</primary></indexterm>
	Samba-3 offers excellent interoperability with MS Windows NT4-style
	domains as well as natively with Microsoft Active Directory domains.
	</para></listitem>

	<listitem><para>
	<indexterm significance="normal"><primary>interdomain</primary><secondary>trustrs</secondary></indexterm>
	Samba-3 permits full NT4-style interdomain trusts.
	</para></listitem>

	<listitem><para>
	<indexterm significance="normal"><primary>authentication</primary></indexterm>
	<indexterm significance="normal"><primary>security</primary><secondary>modes</secondary></indexterm>
	Samba has security modes that permit more flexible authentication
	than is possible with MS Windows NT4 domain controllers.
	</para></listitem>

	<listitem><para>
	<indexterm significance="normal"><primary>account</primary><secondary>database</secondary><tertiary>backends</tertiary></indexterm>
	<indexterm significance="normal"><primary>encrypted</primary></indexterm>
	Samba-3 permits use of multiple concurrent account database backends.
	(Encrypted passwords that are stored in the account database are in
	formats that are unique to Windows networking).
	</para></listitem>

	<listitem><para>
	<indexterm significance="normal"><primary>replicated</primary></indexterm>
	The account database backends can be distributed
	and replicated using multiple methods. This gives Samba-3
	greater flexibility than MS Windows NT4 and in many cases a
	significantly higher utility than Active Directory domains
	with MS Windows 200x.
	</para></listitem>
</itemizedlist>

</sect1>

<sect1>
<title>Server Types</title>


<para>
<indexterm significance="normal"><primary>Server Type</primary></indexterm>
Administrators of Microsoft networks often refer to three different types of servers:
</para>

<itemizedlist>
	<listitem><para>Domain Controller</para>
		<itemizedlist>
			<listitem><para>Primary Domain Controller (PDC)</para></listitem>
			<listitem><para>Backup Domain Controller (BDC)</para></listitem>
			<listitem><para>ADS Domain Controller</para></listitem>
		</itemizedlist>
	</listitem>
	<listitem><para>Domain Member Server</para>
		<itemizedlist>
			<listitem><para>Active Directory Domain Server</para></listitem>
			<listitem><para>NT4 Style Domain Domain Server</para></listitem>
		</itemizedlist>
	</listitem>
	<listitem><para>Standalone Server</para></listitem>
</itemizedlist>

<para>
<indexterm significance="normal"><primary>domain</primary><secondary>control</secondary></indexterm>
<indexterm significance="normal"><primary>domain</primary><secondary>member</secondary></indexterm>
<indexterm significance="normal"><primary>domain control</primary><secondary>primary</secondary></indexterm>
<indexterm significance="normal"><primary>domain control</primary><secondary>backup</secondary></indexterm>
The chapters covering domain control (<link linkend="samba-pdc">Domain Control</link>), 
backup domain control (<link linkend="samba-bdc">Backup Domain Control</link>), and 
domain membership (<link linkend="domain-member">Domain Membership</link>) provide
pertinent information regarding Samba configuration for each of these server roles.
You are strongly encouraged to become intimately familiar with these chapters because
they lay the foundation for deployment of Samba domain security.
</para>

<para>
<indexterm significance="normal"><primary>standalone</primary></indexterm>
A Standalone server is autonomous in respect of the source of its account backend.
Refer to <link linkend="StandAloneServer">Standalone Servers</link> to gain a wider appreciation
of what is meant by a server being configured as a <emphasis>standalone</emphasis> server.
</para>

</sect1>

<sect1>
<title>Samba Security Modes</title>


<para>
<indexterm significance="normal"><primary>Security Mode</primary></indexterm>
<indexterm significance="normal"><primary>security</primary></indexterm>
In this section, the function and purpose of Samba's security modes are described. An accurate understanding of
how Samba implements each security mode as well as how to configure MS Windows clients for each mode will
significantly reduce user complaints and administrator heartache.
</para>

<para>
<indexterm significance="normal"><primary>Server Message Block</primary><see>SMB</see></indexterm>
<indexterm significance="normal"><primary>Common Internet Filesystem</primary><see>CIFS</see></indexterm>
Microsoft Windows networking uses a protocol that was originally called the Server Message Block (SMB)
protocol. Since some time around 1996 the protocol has been better known as the Common Internet Filesystem
(CIFS) protocol.
</para>

<para>
<indexterm significance="normal"><primary>security levels</primary></indexterm>
<indexterm significance="normal"><primary>security modes</primary></indexterm>
<indexterm significance="normal"><primary>user-level</primary></indexterm>
<indexterm significance="normal"><primary>share-level</primary></indexterm>
In the SMB/CIFS networking world, there are only two types of security: <emphasis>user-level</emphasis> and
<emphasis>share level</emphasis>. We refer to these collectively as <emphasis>security levels</emphasis>.  In
implementing these two security levels, Samba provides flexibilities that are not available with MS Windows
NT4/200x servers. In fact, Samba implements <emphasis>share-level</emphasis> security only one way, but has
four ways of implementing <emphasis>user-level</emphasis> security. Collectively, we call the Samba
implementations of the security levels <emphasis>security modes</emphasis>. They are known as
<emphasis>share</emphasis>, <emphasis>user</emphasis>, <emphasis>domain</emphasis>, <emphasis>ADS</emphasis>,
and <emphasis>server</emphasis> modes.  They are documented in this chapter.
</para>

<para>
An SMB server informs the client, at the time of a session setup, the security level the server is running.
There are two options: share-level and user-level. Which of these two the client receives affects the way the
client then tries to authenticate itself. It does not directly affect (to any great extent) the way the Samba
server does security. This may sound strange, but it fits in with the client/server approach of SMB.  In SMB
everything is initiated and controlled by the client, and the server can only tell the client what is
available and whether an action is allowed.
</para>

<para>
The term <literal moreinfo="none">client</literal> refers to all agents whether it is a Windows workstation, a Windows server,
another Samba server, or any vanilla SMB or CIFS client application (e.g., <literal>smbclient</literal>) that
make use of services provided by an SMB/CIFS server.
</para>

<sect2>
<title>User Level Security</title>

<para>
<indexterm significance="normal"><primary>user-level</primary></indexterm>
We describe user-level security first because its simpler.  In user-level security, the client sends a session
setup request directly following protocol negotiation.  This request provides a username and password. The
server can either accept or reject that username/password combination. At this stage the server has no idea
what share the client will eventually try to connect to, so it can't base the
<emphasis>accept/reject</emphasis> on anything other than:
</para>

<orderedlist inheritnum="ignore" continuation="restarts">
<listitem><para>the username/password.</para></listitem>
<listitem><para>the name of the client machine.</para></listitem>
</orderedlist>

<para>
<indexterm significance="normal"><primary>credentials</primary></indexterm>
If the server accepts the username/password credentials, the client expects to be able to mount shares (using
a <emphasis>tree connection</emphasis>) without further specifying a password. It expects that all access
rights will be as the username/password credentials set that was specified in the initial <emphasis>session
setup</emphasis>.
</para>

<para>
<indexterm significance="normal"><primary>session setup</primary></indexterm>
It is also possible for a client to send multiple <emphasis>session setup</emphasis>
requests. When the server responds, it gives the client a <emphasis>uid</emphasis> to use
as an authentication tag for that username/password. The client can maintain multiple
authentication contexts in this way (WinDD is an example of an application that does this).
</para>

<para>
<indexterm significance="normal"><primary>LanManager</primary></indexterm>
<indexterm significance="normal"><primary>case-preserving</primary></indexterm>
<indexterm significance="normal"><primary>case-insensitive</primary></indexterm>
<indexterm significance="normal"><primary>upper-case</primary></indexterm>
<indexterm significance="normal"><primary>lower-case</primary></indexterm>
Windows networking user account names are case-insensitive, meaning that upper-case and lower-case characters
in the account name are considered equivalent. They are said to be case-preserving, but not case significant.
Windows and LanManager systems previous to Windows NT version 3.10 have case-insensitive passwords that were
not necessarily case-preserving. All Windows NT family systems treat passwords as case-preserving and
case-sensitive.
</para>

<sect3>
<title>Example Configuration</title>

<para>
The <filename moreinfo="none">smb.conf</filename> parameter that sets user-level security is:
</para>

<para><simplelist>
<member><indexterm>security</indexterm><parameter>security = user</parameter></member>
</simplelist></para>

<para>
This is the default setting since Samba-2.2.x.
</para>

</sect3>

</sect2>
<sect2>
<title>Share-Level Security</title>

<para>
<indexterm significance="normal"><primary>share-level</primary></indexterm>
<indexterm significance="normal"><primary>mount</primary></indexterm>
In share-level security, the client authenticates itself separately for each share. It sends a password along
with each tree connection request (share mount), but it does not explicitly send a username with this
operation. The client expects a password to be associated with each share, independent of the user. This means
that Samba has to work out what username the client probably wants to use,
because the username is not explicitly sent to the SMB server. Some commercial SMB servers such as NT actually associate passwords directly with shares
in share-level security, but Samba always uses the UNIX authentication scheme where it is a username/password
pair that is authenticated, not a share/password pair.
</para>

<para>
To understand the MS Windows networking parallels, think in terms of MS Windows 9x/Me where you can create a
shared folder that provides read-only or full access, with or without a password.
</para>

<para>
Many clients send a session setup request even if the server is in share-level security. They normally send a valid
username but no password. Samba records this username in a list of possible usernames. When the client then
issues a tree connection request, it also adds to this list the name of the share they try to connect to (useful for
home directories) and any users listed in the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="USER" xlink:href="smb.conf.5.html#USER">user</link> parameter in the <filename moreinfo="none">smb.conf</filename> file.
The password is then checked in turn against these possible usernames. If a match is found, then the client is
authenticated as that user.
</para>

<para>
<indexterm significance="normal"><primary>name service switch</primary><see>NSS</see></indexterm>
<indexterm significance="normal"><primary>/etc/passwd</primary></indexterm>
<indexterm significance="normal"><primary>nsswitch.conf</primary></indexterm>
Where the list of possible user names is not provided, Samba makes a UNIX system call to find the user
account that has a password that matches the one provided from the standard account database. On a system that
has no name service switch (NSS) facility, such lookups will be from the <filename moreinfo="none">/etc/passwd</filename>
database. On NSS enabled systems, the lookup will go to the libraries that have been specified in the
<filename moreinfo="none">nsswitch.conf</filename> file. The entries in that file in which the libraries are specified are:
<screen format="linespecific">
passwd: files nis ldap
shadow: files nis ldap
group: files nis ldap
</screen>
<indexterm significance="normal"><primary>/etc/passwd</primary></indexterm>
<indexterm significance="normal"><primary>/etc/group</primary></indexterm>
<indexterm significance="normal"><primary>NIS</primary></indexterm>
In the example shown here (not likely to be used in practice) the lookup will check
<filename moreinfo="none">/etc/passwd</filename> and <filename moreinfo="none">/etc/group</filename>, if not found it will check NIS, then
LDAP.
</para>

<sect3>
<title>Example Configuration</title>

<para>
The <filename moreinfo="none">smb.conf</filename> parameter that sets share-level security is:
</para>

<para><simplelist>
<member><indexterm>security</indexterm><parameter>security = share</parameter></member>
</simplelist></para>

</sect3>
</sect2>

<sect2>
<title>Domain Security Mode (User-Level Security)</title>

<para>
<indexterm significance="normal"><primary>domain</primary><secondary>controllers</secondary></indexterm>
<indexterm significance="normal"><primary>security</primary><secondary>controllers</secondary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>BDC</primary></indexterm>
<indexterm significance="normal"><primary>logon</primary></indexterm>
<indexterm significance="normal"><primary>authentication</primary></indexterm>
Domain security provides a mechanism for storing all user and group accounts in a central, shared, account
repository. The centralized account repository is shared between domain (security) controllers. Servers that
act as domain controllers provide authentication and validation services to all machines that participate in
the security context for the domain. A primary domain controller (PDC) is a server that is responsible for
maintaining the integrity of the security account database. Backup domain controllers (BDCs) provide only domain
logon and authentication services. Usually, BDCs will answer network logon requests more responsively than
will a PDC.
</para>

<para>
<indexterm significance="normal"><primary>domain member</primary></indexterm>
<indexterm significance="normal"><primary>trust account</primary></indexterm>
<indexterm significance="normal"><primary>trust</primary><secondary>account</secondary></indexterm>
<indexterm significance="normal"><primary>domain</primary><secondary>security</secondary></indexterm>
<indexterm significance="normal"><primary>domain</primary><secondary>controller</secondary></indexterm>
When Samba is operating in <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SECURITY" xlink:href="smb.conf.5.html#SECURITY">security = domain</link> mode, the Samba server has a
domain security trust account (a machine account) and causes all authentication requests to be passed through
to the domain controllers.  In other words, this configuration makes the Samba server a domain member server,
even when it is in fact acting as a domain controller. All machines that participate in domain security must
have a machine account in the security database.
</para>

<para>
<indexterm significance="normal"><primary>account</primary><secondary>database</secondary></indexterm>
<indexterm significance="normal"><primary>machine</primary><secondary>account</secondary></indexterm>
<indexterm significance="normal"><primary>NetBIOS</primary><secondary>name</secondary></indexterm>
<indexterm significance="normal"><primary>NetBIOS</primary></indexterm>
Within the domain security environment, the underlying security architecture uses user-level security. Even
machines that are domain members must authenticate on startup. The machine account consists of an account
entry in the accounts database, the name of which is the NetBIOS name of the machine and of which the password
is randomly generated and known to both the domain controllers and the member machine. If the machine account
cannot be validated during startup, users will not be able to log on to the domain using this machine because
it cannot be trusted. The machine account is referred to as a machine trust account.
</para>

<para>
There are three possible domain member configurations:
</para>

<orderedlist inheritnum="ignore" continuation="restarts">
	<listitem><para>Primary domain controller (PDC) - of which there is one per domain.</para></listitem>
	<listitem><para>Backup domain controller (BDC) - of which there can be any number per domain.</para></listitem>
	<listitem><para>Domain member server (DMS) - of which there can be any number per domain.</para></listitem>
</orderedlist>

<para>
<indexterm significance="normal"><primary>DMS</primary></indexterm>
We will discuss each of these in separate chapters. For now, we are most interested in basic DMS
configuration.
</para>

<sect3>
<title>Example Configuration</title>
<para><emphasis>
Samba as a Domain Member Server
</emphasis></para>


<para>
<indexterm significance="normal"><primary>server type</primary><secondary>domain member</secondary></indexterm>
This method involves addition of the following parameters in the <filename moreinfo="none">smb.conf</filename> file:
<simplelist>
<member><indexterm>security</indexterm><parameter>security = domain</parameter></member>
<member><indexterm>workgroup</indexterm><parameter>workgroup = MIDEARTH</parameter></member>
</simplelist>
</para>

<para>
In order for this method to work, the Samba server needs to join the MS Windows NT
security domain. This is done as follows:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary></indexterm>
<indexterm significance="normal"><primary>Domain Member</primary><secondary>joining</secondary></indexterm>
</para>


<procedure>
        <step performance="required"><para>On the MS Windows NT domain controller, using
        the Server Manager, add a machine account for the Samba server.
        </para></step>

        <step performance="required"><para>On the UNIX/Linux system execute:</para>
	
			<para><screen format="linespecific"><prompt moreinfo="none">root# </prompt><userinput moreinfo="none">net rpc join -U administrator%password</userinput></screen></para>
		</step>
</procedure>

<note><para>
<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
Samba-2.2.4 and later Samba 2.2.x series releases can autojoin a Windows NT4-style domain just by executing:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">smbpasswd -j <replaceable>DOMAIN_NAME</replaceable> -r <replaceable>PDC_NAME</replaceable> \
	 -U Administrator%<replaceable>password</replaceable></userinput>
</screen>
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>join</tertiary></indexterm>
Samba-3 can do the same by executing:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">net rpc join -U Administrator%<replaceable>password</replaceable></userinput>
</screen>
It is not necessary with Samba-3 to specify the <replaceable>DOMAIN_NAME</replaceable> or the
<replaceable>PDC_NAME</replaceable>, as it figures this out from the <filename moreinfo="none">smb.conf</filename> file settings.
</para></note>

<para>
<indexterm significance="normal"><primary>invalid shell</primary></indexterm>
<indexterm significance="normal"><primary>/etc/passwd</primary></indexterm>
<indexterm significance="normal"><primary>/bin/false</primary></indexterm>
Use of this mode of authentication requires there to be a standard UNIX account for each user in order to
assign a UID once the account has been authenticated by the Windows domain controller. This account can be
blocked to prevent logons by clients other than MS Windows through means such as setting an invalid shell in
the <filename moreinfo="none">/etc/passwd</filename> entry. The best way to allocate an invalid shell to a user account is to
set the shell to the file <filename moreinfo="none">/bin/false</filename>.
</para>

<para>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>BDC</primary></indexterm>
Domain controllers can be located anywhere that is convenient. The best advice is to have a BDC on every
physical network segment, and if the PDC is on a remote network segment the use of WINS (see <link linkend="NetworkBrowsing">Network Browsing</link> for more information) is almost essential.
</para>

<para>
An alternative to assigning UIDs to Windows users on a Samba member server is presented in <link linkend="winbind">Winbind</link>, <link linkend="winbind">Winbind: Use of Domain Accounts</link>.
</para>

<para>
For more information regarding domain membership, <link linkend="domain-member">Domain Membership</link>.
</para>

</sect3>
</sect2>

<sect2>
<title>ADS Security Mode (User-Level Security)</title>

<para>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
<indexterm significance="normal"><primary>native mode</primary></indexterm>
Both Samba-2.2, and Samba-3 can join an Active Directory domain using NT4 style RPC based security.  This is
possible if the domain is run in native mode. Active Directory in native mode perfectly allows NT4-style
domain members. This is contrary to popular belief.
</para>

<para>
If you are using Active Directory, starting with Samba-3 you can join as a native AD member. Why would you
want to do that?  Your security policy might prohibit the use of NT-compatible authentication protocols. All
your machines are running Windows 2000 and above and all use Kerberos. In this case, Samba, as an NT4-style
domain, would still require NT-compatible authentication data. Samba in AD-member mode can accept Kerberos
tickets.
</para>

<para>
<indexterm significance="normal"><primary>realm</primary></indexterm>
<indexterm significance="normal"><primary>mixed mode</primary></indexterm>
Sites that use Microsoft Windows active directory services (ADS) should be aware of the significance of the
terms: <literal moreinfo="none">native mode</literal> and <literal moreinfo="none">mixed mode</literal> ADS operation. The term
<literal moreinfo="none">realm</literal> is used to describe a Kerberos-based security architecture (such as is used by
Microsoft ADS).
</para>

<sect3>
<title>Example Configuration</title>

<para><simplelist>
<member><indexterm>realm</indexterm><parameter>realm = your.kerberos.REALM</parameter></member>
<member><indexterm>security</indexterm><parameter>security = ADS</parameter></member>
</simplelist></para>

<para>
The following parameter may be required:
</para>

<para><simplelist>
<member><indexterm>password server</indexterm><parameter>password server = your.kerberos.server</parameter></member>
</simplelist></para>

<para>
Please refer to <link linkend="domain-member">Domain Membership</link>, and <link linkend="ads-member">Samba
ADS Domain Membership</link> for more information regarding this configuration option.
</para>

</sect3>
</sect2>

<sect2>
<title>Server Security (User Level Security)</title>

<para>
Server security mode is left over from the time when Samba was not capable of acting
as a domain member server. It is highly recommended not to use this feature. Server
security mode has many drawbacks that include:
</para>

<itemizedlist>
	<listitem><para>Potential account lockout on MS Windows NT4/200x password servers.</para></listitem>
	<listitem><para>Lack of assurance that the password server is the one specified.</para></listitem>
	<listitem><para>Does not work with Winbind, which is particularly needed when storing profiles remotely.</para></listitem>
	<listitem><para>This mode may open connections to the password server and keep them open for extended periods.</para></listitem>
	<listitem><para>Security on the Samba server breaks badly when the remote password server suddenly shuts down.</para></listitem>
	<listitem><para>With this mode there is NO security account in the domain that the password server belongs to for the Samba server.</para></listitem>
</itemizedlist>

<para>
<indexterm significance="normal"><primary>session setup</primary></indexterm>
<indexterm significance="normal"><primary>SMB</primary></indexterm>
In server security mode the Samba server reports to the client that it is in user-level security. The client
then does a session setup as described earlier.  The Samba server takes the username/password that the client
sends and attempts to log into the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PASSWORDSERVER" xlink:href="smb.conf.5.html#PASSWORDSERVER">password server</link> by sending exactly the same
username/password that it got from the client. If that server is in user-level security and accepts the
password, then Samba accepts the client's connection. This parameter allows the Samba server to use another
SMB server as the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PASSWORDSERVER" xlink:href="smb.conf.5.html#PASSWORDSERVER">password server</link>.
</para>

<para>
<indexterm significance="normal"><primary>security level</primary></indexterm>
<indexterm significance="normal"><primary>encryption</primary></indexterm>
You should also note that at the start of all this, when the server tells the client
what security level it is in, it also tells the client if it supports encryption. If it
does, it supplies the client with a random cryptkey. The client will then send all
passwords in encrypted form. Samba supports this type of encryption by default.
</para>

<para>
The parameter <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SECURITY" xlink:href="smb.conf.5.html#SECURITY">security = server</link> means that Samba reports to clients that
it is running in <emphasis>user mode</emphasis> but actually passes off all authentication requests to another
user mode server. This requires an additional parameter <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PASSWORDSERVER" xlink:href="smb.conf.5.html#PASSWORDSERVER">password server</link> that points to
the real authentication server.  The real authentication server can be another Samba server, or it can be a
Windows NT server, the latter being natively capable of encrypted password support.
</para>

<note><para>
<indexterm significance="normal"><primary>password server</primary></indexterm>
<indexterm significance="normal"><primary>workgroup</primary></indexterm>
When Samba is running in <emphasis>server security mode</emphasis>, it is essential that the parameter
<emphasis>password server</emphasis> is set to the precise NetBIOS machine name of the target authentication
server. Samba cannot determine this from NetBIOS name lookups because the choice of the target authentication
server is arbitrary and cannot be determined from a domain name. In essence, a Samba server that is in
<emphasis>server security mode</emphasis> is operating in what used to be known as workgroup mode.
</para></note>

<sect3>
<title>Example Configuration</title>
<para><emphasis>
Using MS Windows NT as an Authentication Server
</emphasis></para>

<para>
This method involves the additions of the following parameters in the <filename moreinfo="none">smb.conf</filename> file:
</para>

<para><simplelist>
<member><indexterm>encrypt passwords</indexterm><parameter>encrypt passwords = Yes</parameter></member>
<member><indexterm>security</indexterm><parameter>security = server</parameter></member>
<member><indexterm>password server</indexterm><parameter>password server = "NetBIOS_name_of_a_DC"</parameter></member>
</simplelist></para>


<para>
There are two ways of identifying whether or not a username and password pair is valid.
One uses the reply information provided as part of the authentication messaging
process, the other uses just an error code.
</para>

<para>
<indexterm significance="normal"><primary>bogus</primary></indexterm>
<indexterm significance="normal"><primary>lockout</primary></indexterm>
The downside of this mode of configuration is that for security reasons Samba
will send the password server a bogus username and a bogus password, and if the remote
server fails to reject the bogus username and password pair, then an alternative mode of
identification or validation is used. Where a site uses password lockout, after a
certain number of failed authentication attempts, this will result in user lockouts.
</para>

<para>
Use of this mode of authentication requires a standard UNIX account for the user.
This account can be blocked to prevent logons by non-SMB/CIFS clients.
</para>

</sect3>
</sect2>

</sect1>

<sect1>
<title>Password Checking</title>

<para>
MS Windows clients may use encrypted passwords as part of a challenge/response
authentication model (a.k.a. NTLMv1 and NTLMv2) or alone, or clear-text strings for simple
password-based authentication. It should be realized that with the SMB protocol,
the password is passed over the network either in plaintext or encrypted, but
not both in the same authentication request.
</para>

<para>
<indexterm significance="normal"><primary>encrypted passwords</primary></indexterm>
<indexterm significance="normal"><primary>encrypted</primary></indexterm>
When encrypted passwords are used, a password that has been entered by the user
is encrypted in two ways:
</para>

<itemizedlist>
        <listitem><para>An MD4 hash of the unicode of the password
        string. This is known as the NT hash.
        </para></listitem>

        <listitem><para>The password is converted to uppercase,
        and then padded or truncated to 14 bytes. This string is
        then appended with 5 bytes of NULL characters and split to
        form two 56-bit DES keys to encrypt a "magic" 8-byte value.
        The resulting 16 bytes form the LanMan hash.
        </para></listitem>
</itemizedlist>

<para>
<indexterm significance="normal"><primary>plain-text</primary><secondary>passwords</secondary></indexterm>
MS Windows 95 pre-service pack 1 and MS Windows NT versions 3.x and version 4.0 pre-service pack 3 will use
either mode of password authentication. All versions of MS Windows that follow these versions no longer
support plain-text passwords by default.
</para>

<para>
<indexterm significance="normal"><primary>cached</primary><secondary>password</secondary></indexterm>
MS Windows clients have a habit of dropping network mappings that have been idle
for 10 minutes or longer. When the user attempts to use the mapped drive
connection that has been dropped, the client re-establishes the connection using
a cached copy of the password.
</para>

<para>
When Microsoft changed the default password mode, support was dropped for caching
of the plaintext password. This means that when the registry parameter is changed
to re-enable use of plaintext passwords, it appears to work, but when a dropped
service connection mapping attempts to revalidate, this will fail if the remote
authentication server does not support encrypted passwords. It is definitely not
a good idea to re-enable plaintext password support in such clients.
</para>

<para>
The following parameters can be used to work around the issue of Windows 9x/Me clients
uppercasing usernames and passwords before transmitting them to the SMB server
when using clear-text authentication:
</para>


<?latex \newpage ?>
<simplelist>
<member><indexterm>password level</indexterm></member>
<member><indexterm>username level</indexterm></member>
</simplelist>

<para>
By default Samba will convert to lowercase the username before attempting to lookup the user
in the database of local system accounts. Because UNIX usernames conventionally
only contain lowercase characters, the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="USERNAME-LEVEL" xlink:href="smb.conf.5.html#USERNAME-LEVEL">username-level</link> parameter
is rarely needed.
</para>

<para>
<indexterm significance="normal"><primary>clear-text</primary></indexterm>
However, passwords on UNIX systems often make use of mixed-case characters.  This means that in order for a
user on a Windows 9x/Me client to connect to a Samba server using clear-text authentication, the
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PASSWORDLEVEL" xlink:href="smb.conf.5.html#PASSWORDLEVEL">password level</link> must be set to the maximum number of uppercase letters that
<emphasis>could</emphasis> appear in a password. Note that if the Server OS uses the traditional DES version
of crypt(), a <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PASSWORDLEVEL" xlink:href="smb.conf.5.html#PASSWORDLEVEL">password level</link> of 8 will result in case-insensitive passwords as seen
from Windows users. This will also result in longer login times because Samba has to compute the permutations
of the password string and try them one by one until a match is located (or all combinations fail).
</para>

<para>
The best option to adopt is to enable support for encrypted passwords wherever
Samba is used. Most attempts to apply the registry change to re-enable plaintext
passwords will eventually lead to user complaints and unhappiness.
</para>

</sect1>

<sect1>
<title>Common Errors</title>

<para>
We all make mistakes. It is okay to make mistakes, as long as they are made in the right places
and at the right time. A mistake that causes lost productivity is seldom tolerated; however, a mistake
made in a developmental test lab is expected.
</para>

<para>
Here we look at common mistakes and misapprehensions that have been the subject of discussions
on the Samba mailing lists. Many of these are avoidable by doing your homework before attempting
a Samba implementation. Some are the result of a misunderstanding of the English language,
which has many phrases that are potentially vague and may be highly confusing
to those for whom English is not their native tongue.
</para>

<sect2>
<title>What Makes Samba a Server?</title>

<para>
To some, the nature of the Samba security mode is obvious, but entirely
wrong all the same. It is assumed that <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SECURITY" xlink:href="smb.conf.5.html#SECURITY">security = server</link> means that Samba
will act as a server. Not so! This setting means that Samba will <emphasis>try</emphasis>
to use another SMB server as its source for user authentication alone.
</para>

<para>
Samba is a server regardless of which security mode is chosen. When Samba is used outside of a domain security
context, it is best to leave the security mode at the default setting. By default Samba-3 uses user-mode
security.
</para>

</sect2>

<sect2>
<title>What Makes Samba a Domain Controller?</title>

<para>
<indexterm significance="normal"><primary>server-mode</primary></indexterm>
The <filename moreinfo="none">smb.conf</filename> parameter <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SECURITY" xlink:href="smb.conf.5.html#SECURITY">security = domain</link> does not really make Samba behave
as a domain controller. This setting means we want Samba to be a domain member. See <link linkend="samba-pdc">Samba as a PDC</link> for more information.
</para>

</sect2>

<sect2>
<title>What Makes Samba a Domain Member?</title>

<para>
Guess! So many others do. But whatever you do, do not think that <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SECURITY" xlink:href="smb.conf.5.html#SECURITY">security = user</link>
makes Samba act as a domain member. Read the manufacturer's manual before the warranty expires. See 
<link linkend="domain-member">Domain Membership</link>, for more information.
</para>

</sect2>


<sect2>
<title>Constantly Losing Connections to Password Server</title>

<para><quote>
Why does server_validate() simply give up rather than re-establish its connection to the
password server?  Though I am not fluent in the SMB protocol, perhaps the cluster server
process passes along to its client workstation the session key it receives from the password
server, which means the password hashes submitted by the client would not work on a subsequent
connection whose session key would be different. So server_validate() must give up.
</quote></para>

<para>
Indeed. That's why <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SECURITY" xlink:href="smb.conf.5.html#SECURITY">security = server</link>
is at best a nasty hack. Please use <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SECURITY" xlink:href="smb.conf.5.html#SECURITY">security = domain</link>;
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SECURITY" xlink:href="smb.conf.5.html#SECURITY">security = server</link> mode is also known as pass-through authentication.
</para>

</sect2>

<sect2>
<title>Stand-alone Server is converted to Domain Controller <?latex --- ?> Now User accounts don't work</title>

<para><quote>
When I try to log in to the DOMAIN, the eventlog shows <emphasis>tried credentials DOMAIN/username; effective
credentials SERVER/username</emphasis>
</quote></para>

<para>
Usually this is due to a user or machine account being created before the Samba server is configured to be a
domain controller. Accounts created before the server becomes a domain controller will be
<emphasis>local</emphasis> accounts and authenticated as what looks like a member in the SERVER domain, much
like local user accounts in Windows 2000 and later.  Accounts created after the Samba server becomes a domain
controller will be <emphasis>domain</emphasis> accounts and will be authenticated as a member of the DOMAIN
domain.
</para>

<para>
This can be verified by issuing the command <literal>pdbedit -L -v username</literal>.  If this reports DOMAIN
then the account is a domain account, if it reports SERVER then the account is a local account.
</para>

<para>
The easiest way to resolve this is to remove and recreate the account; however this may cause problems with
established user profiles. You can also use <literal>pdbedit -u username -I DOMAIN</literal>. You may also
need to change the User SID and Primary Group SID to match the domain.
</para>

</sect2>

</sect1>

</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="samba-pdc">

<chapterinfo>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
	<author>
<firstname>Gerald</firstname><surname>Carter</surname><othername>(Jerry)</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jerry@samba.org</email></address>
</affiliation></author>
	
<author>
	<firstname>David</firstname><surname>Bannon</surname>
	<affiliation>
		<orgname>Samba Team</orgname>
		<address format="linespecific"><email>dbannon@samba.org</email></address>
	</affiliation>
</author>
	<author>
<firstname>Guenther</firstname><surname>Deschner</surname>
<affiliation>
	<orgname>SuSE</orgname>
	<address format="linespecific"><email>gd@suse.de</email></address>
</affiliation> <contrib>LDAP updates</contrib></author>
</chapterinfo>

<title>Domain Control</title>

<para>
There are many who approach MS Windows networking with incredible misconceptions.
That's okay, because it gives the rest of us plenty of opportunity to be of assistance.
Those who really want help are well advised to become familiar with information
that is already available.
</para>

<para>
<indexterm significance="normal"><primary>domain</primary><secondary>controller</secondary></indexterm>
You are advised not to tackle this section without having first understood
and mastered some basics. MS Windows networking is not particularly forgiving of
misconfiguration. Users of MS Windows networking are likely to complain 
of persistent niggles that may be caused by a broken network configuration.
To a great many people, however, MS Windows networking starts with a domain controller
that in some magical way is expected to solve all network operational ills.
</para>

<para>
<link linkend="domain-example">The Example Domain Illustration</link> shows a typical MS Windows domain security
network environment. Workstations A, B, and C are representative of many physical MS Windows
network clients.
</para>

<figure id="domain-example" float="0">
	<title>An Example Domain.</title>
	<mediaobject><imageobject role="html"><imagedata fileref="images/domain.png" scale="40" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/domain.png" scale="40" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/domain"/></imageobject></mediaobject>
</figure>

<para>
From the Samba mailing list we can readily identify many common networking issues.
If you are not clear on the following subjects, then it will do much good to read the 
sections of this HOWTO that deal with it. These are the most common causes of MS Windows
networking problems:
</para>

<itemizedlist>
	<listitem><para>Basic TCP/IP configuration.</para></listitem>
	<listitem><para>NetBIOS name resolution.</para></listitem>
	<listitem><para>Authentication configuration.</para></listitem>
	<listitem><para>User and group configuration.</para></listitem>
	<listitem><para>Basic file and directory permission control in UNIX/Linux.</para></listitem>
	<listitem><para>Understanding how MS Windows clients interoperate in a network environment.</para></listitem>
</itemizedlist>

<para>
Do not be put off; on the surface of it MS Windows networking seems so simple that anyone 
can do it. In fact, it is not a good idea to set up an MS Windows network with
inadequate training and preparation. But let's get our first indelible principle out of the
way: <emphasis>It is perfectly okay to make mistakes!</emphasis> In the right place and at
the right time, mistakes are the essence of learning. It is very much not okay to make
mistakes that cause loss of productivity and impose an avoidable financial burden on an
organization.
</para>

<para>
Where is the right place to make mistakes? Only out of harms way. If you are going to
make mistakes, then please do it on a test network, away from users, and in such a way as
to not inflict pain on others. Do your learning on a test network.
</para>

<sect1>
<title>Features and Benefits</title>

<para>
<indexterm significance="normal"><primary>domain security</primary></indexterm>
<emphasis>What is the key benefit of Microsoft Domain Security?</emphasis>
</para>

<para>
<indexterm significance="normal">single sign-on<primary/><see>SSO</see></indexterm>
<indexterm significance="normal"><primary>trust</primary></indexterm>
<indexterm significance="normal"><primary>account</primary></indexterm>
<indexterm significance="normal"><primary>domain</primary><secondary>security</secondary><tertiary>protocols</tertiary></indexterm>
In a word, <emphasis>single sign-on</emphasis>, or SSO for short. To many, this is the Holy Grail of MS
Windows NT and beyond networking. SSO allows users in a well-designed network to log onto any workstation that
is a member of the domain that contains their user account (or in a domain that has an appropriate trust
relationship with the domain they are visiting) and they will be able to log onto the network and access
resources (shares, files, and printers) as if they are sitting at their home (personal) workstation. This is a
feature of the domain security protocols.
</para>

<para>
<indexterm significance="normal"><primary>SID</primary></indexterm>
<indexterm significance="normal"><primary>RID</primary></indexterm>
<indexterm significance="normal"><primary>relative identifier</primary><see>RID</see></indexterm>
<indexterm significance="normal"><primary>security identifier</primary><see>SID</see></indexterm>
<indexterm significance="normal"><primary>access control</primary></indexterm>
The benefits of domain security are available to those sites that deploy a Samba PDC.  A domain provides a
unique network security identifier (SID). Domain user and group security identifiers are comprised of the
network SID plus a relative identifier (RID) that is unique to the account. User and group SIDs (the network
SID plus the RID) can be used to create access control lists (ACLs) attached to network resources to provide
organizational access control. UNIX systems recognize only local security identifiers.
</para>

<para>
<indexterm significance="normal"><primary>SID</primary></indexterm>
A SID represents a security context. For example, every Windows machine has local accounts within the security
context of the local machine which has a unique SID. Every domain (NT4, ADS, Samba) contains accounts that
exist within the domain security context which is defined by the domain SID.
</para>

<para>
<indexterm significance="normal"><primary>SID</primary></indexterm>
<indexterm significance="normal"><primary>RID</primary></indexterm>
A domain member server will have a SID that differs from the domain SID.  The domain member server can be
configured to regard all domain users as local users. It can also be configured to recognize domain users and
groups as non-local. SIDs are persistent. A typical domain of user SID looks like this:
<screen format="linespecific">
S-1-5-21-726309263-4128913605-1168186429
</screen>
Every account (user, group, machine, trust, etc.) is assigned a RID. This is done automatically as an account
is created. Samba produces the RID algorithmically. The UNIX operating system uses a separate name space for
user and group identifiers (the UID and GID) but Windows allocates the RID from a single name space. A Windows
user and a Windows group can not have the same RID. Just as the UNIX user <literal moreinfo="none">root</literal> has the
UID=0, the Windows Administrator has the well-known RID=500. The RID is catenated to the Windows domain SID,
so Administrator account for a domain that has the above SID will have the user SID
<screen format="linespecific">
S-1-5-21-726309263-4128913605-1168186429-500
</screen>
The result is that every account in the Windows networking world has a globally unique security identifier.
</para>

<note><para>
<indexterm significance="normal"><primary>domain</primary><secondary>member</secondary></indexterm>
<indexterm significance="normal"><primary>machine account</primary></indexterm>
<indexterm significance="normal"><primary>domain</primary><secondary>trust account</secondary></indexterm>
Network clients of an MS Windows domain security environment must be domain members to be able to gain access
to the advanced features provided. Domain membership involves more than just setting the workgroup name to the
domain name. It requires the creation of a domain trust account for the workstation (called a machine
account). Refer to <link linkend="domain-member">Domain Membership</link> for more information.
</para></note>

<para>
The following functionalities are new to the Samba-3 release:
</para>

<itemizedlist>
	<listitem><para>
	<indexterm significance="normal"><primary>account</primary><secondary>backend</secondary></indexterm>
	Samba-3 supports the use of a choice of backends that may be used in which user, group and machine
	accounts may be stored. Multiple passwd backends can be used in combination, either as additive backend
	data sets, or as fail-over data sets.
	</para>

	<para>
	<indexterm significance="normal"><primary>LDAP</primary></indexterm>
	<indexterm significance="normal"><primary>replicated</primary></indexterm>
	<indexterm significance="normal"><primary>distributed</primary></indexterm>
	<indexterm significance="normal"><primary>scalability</primary></indexterm>
	<indexterm significance="normal"><primary>reliability</primary></indexterm>
	An LDAP passdb backend confers the benefit that the account backend can be distributed and replicated,
	which is of great value because it confers scalability and provides a high degree of reliability. 
	</para></listitem>

	<listitem><para>
	<indexterm significance="normal"><primary>interdomain</primary><secondary>trust</secondary><tertiary>account</tertiary></indexterm>
	<indexterm significance="normal"><primary>trust account</primary><secondary>interdomain</secondary></indexterm>
	<indexterm significance="normal"><primary>interoperability</primary></indexterm>
	Windows NT4 domain trusts. Samba-3 supports workstation and server (machine) trust accounts. It also
	supports Windows NT4 style interdomain trust accounts, which further assists in network scalability
	and interoperability.
	</para></listitem>
	
	<listitem><para>
	<indexterm significance="normal"><primary>NetBIOS</primary></indexterm>
	<indexterm significance="normal"><primary>raw SMB</primary></indexterm>
	<indexterm significance="normal"><primary>active directory</primary></indexterm>
	<indexterm significance="normal"><primary>domain</primary><secondary>member server</secondary></indexterm>
	<indexterm significance="normal"><primary>domain</primary><secondary>controller</secondary></indexterm>
	<indexterm significance="normal"><primary>network</primary><secondary>browsing</secondary></indexterm>
	Operation without NetBIOS over TCP/IP, rather using the raw SMB over TCP/IP. Note, this is feasible
	only when operating as a Microsoft active directory domain member server. When acting as a Samba domain
	controller the use of NetBIOS is necessary to provide network browsing support.
	</para></listitem>

	<listitem><para>
	<indexterm significance="normal"><primary>WINS</primary></indexterm>
	<indexterm significance="normal"><primary>TCP port</primary></indexterm>
	<indexterm significance="normal"><primary>session services</primary></indexterm>
	Samba-3 provides NetBIOS name services (WINS), NetBIOS over TCP/IP (TCP port 139) session services, SMB over
	TCP/IP (TCP port 445) session services, and Microsoft compatible ONC DCE RPC services (TCP port 135)
	services.
	</para></listitem>

	<listitem><para>
	<indexterm significance="normal"><primary>Nexus.exe</primary></indexterm>
	Management of users and groups via the User Manager for Domains. This can be done on any MS Windows client
	using the <filename moreinfo="none">Nexus.exe</filename> toolkit for Windows 9x/Me, or using the SRVTOOLS.EXE package for MS
	Windows NT4/200x/XP platforms. These packages are available from Microsoft's Web site.
	</para></listitem>

	<listitem><para>
	Implements full Unicode support. This simplifies cross-locale internationalization support. It also opens up
	the use of protocols that Samba-2.2.x had but could not use due to the need to fully support Unicode.
	</para></listitem>
</itemizedlist>

<para>
The following functionalities are not provided by Samba-3:
</para>

<itemizedlist>
	<listitem><para>
	<indexterm significance="normal"><primary>SAM</primary></indexterm>
	<indexterm significance="normal"><primary>replication</primary></indexterm>
	SAM replication with Windows NT4 domain controllers (i.e., a Samba PDC and a Windows NT BDC, or vice versa).
	This means Samba cannot operate as a BDC when the PDC is Microsoft-based Windows NT PDC. Samba-3 can not
	participate in replication of account data to Windows PDCs and BDCs.
	</para></listitem>
	
	<listitem><para>
	<indexterm significance="normal"><primary>kerberos</primary></indexterm>
	<indexterm significance="normal"><primary>active directory</primary></indexterm>
	Acting as a Windows 2000 active directory domain controller (i.e., Kerberos and Active Directory). In point of
	fact, Samba-3 does have some Active Directory domain control ability that is at this time purely experimental.
	Active directory domain control is one of the features that is being developed in Samba-4, the next
	generation Samba release. At this time there are no plans to enable active directory domain control
	support during the Samba-3 series life-cycle.
	</para></listitem>

	<listitem><para>
	<indexterm significance="normal"><primary>MMC</primary></indexterm>
	<indexterm significance="normal"><primary>SVRTOOLS.EXE</primary></indexterm>
	<indexterm significance="normal"><primary>Microsoft management console</primary><see>MMC</see></indexterm>
	The Windows 200x/XP Microsoft Management Console (MMC) cannot be used to manage a Samba-3 server. For this you
	can use only the MS Windows NT4 Domain Server Manager and the MS Windows NT4 Domain User Manager. Both are
	part of the SVRTOOLS.EXE package mentioned later.
	</para></listitem>
</itemizedlist>

<para>
<indexterm significance="normal"><primary>Windows XP Home edition</primary></indexterm>
<indexterm significance="normal"><primary>LanMan</primary></indexterm>
Windows 9x/Me/XP Home clients are not true members of a domain for reasons outlined in this chapter. The
protocol for support of Windows 9x/Me-style network (domain) logons is completely different from NT4/Windows
200x-type domain logons and has been officially supported for some time. These clients use the old LanMan
network logon facilities that are supported in Samba since approximately the Samba-1.9.15 series.
</para>

<para>
<indexterm significance="normal"><primary>group</primary><secondary>mapping</secondary></indexterm>
Samba-3 implements group mapping between Windows NT groups and UNIX groups (this is really quite complicated
to explain in a short space). This is discussed more fully in <link linkend="groupmapping">Group Mapping: MS
Windows and UNIX</link>.
</para>

<para>
<indexterm significance="normal"><primary>machine trust account</primary></indexterm>
<indexterm significance="normal"><primary>trust account</primary><secondary>machine</secondary></indexterm>
<indexterm significance="normal"><primary>machine account</primary></indexterm>
Samba-3, like an MS Windows NT4 PDC or a Windows 200x Active Directory, needs to store user and Machine Trust
Account information in a suitable backend data-store.  Refer to <link linkend="machine-trust-accounts">MS
Windows Workstation/Server Machine Trust Accounts</link>. With Samba-3 there can be multiple backends for
this. A complete discussion of account database backends can be found in <link linkend="passdb">Account
Information Databases</link>.
</para>

</sect1>

<sect1>
<title>Single Sign-On and Domain Security</title>

<para>
<indexterm significance="normal"><primary>single sign-on</primary><see>SSO</see></indexterm>
<indexterm significance="normal"><primary>SSO</primary></indexterm>
<indexterm significance="normal"><primary>active directory</primary></indexterm>
<indexterm significance="normal"><primary>authentication</primary></indexterm>
<indexterm significance="normal"><primary>validation</primary></indexterm>
<indexterm significance="normal"><primary>password uniqueness</primary></indexterm>
<indexterm significance="normal"><primary>password history</primary></indexterm>
When network administrators  are asked to describe the benefits of Windows NT4 and active directory networking
the most often mentioned feature is that of single sign-on (SSO). Many companies have implemented SSO
solutions. The mode of implementation of a single sign-on solution is an important factor in the practice of
networking in general, and is critical in respect of Windows networking. A company may have a wide variety of
information systems, each of which requires a form of user authentication and validation, thus it is not
uncommon that users may need to remember more than ten login IDs and passwords. This problem is compounded
when the password for each system must be changed at regular intervals, and particularly so where password
uniqueness and history limits are applied.
</para>

<para>
<indexterm significance="normal"><primary>management overheads</primary></indexterm>
There is a broadly held perception that SSO is the answer to the problem of users having to deal with too many
information system access credentials (username/password pairs). Many elaborate schemes have been devised to
make it possible to deliver a user-friendly SSO solution. The trouble is that if this implementation is not
done correctly, the site may end up paying dearly by way of complexity and management overheads. Simply put,
many SSO solutions are an administrative nightmare.
</para>

<para>
<indexterm significance="normal"><primary>identity management</primary></indexterm>
<indexterm significance="normal"><primary>authentication system</primary></indexterm>
<indexterm significance="normal"><primary>SSO</primary></indexterm>
SSO implementations utilize centralization of all user account information. Depending on environmental
complexity and the age of the systems over which a SSO solution is implemented, it may not be possible to
change the solution architecture so as to accomodate a new identity management and user authentication system.
Many SSO solutions involving legacy systems consist of a new super-structure that handles authentication on
behalf of the user. The software that gets layered over the old system may simply implement a proxy
authentication system. This means that the addition of SSO increases over-all information systems complexity.
Ideally, the implementation of SSO should reduce complexity and reduce administative overheads.
</para>

<para>
<indexterm significance="normal"><primary>centralized identity management</primary></indexterm>
<indexterm significance="normal"><primary>identity management</primary><secondary>centralized</secondary></indexterm>
<indexterm significance="normal"><primary>centralized</primary><secondary>authentication</secondary></indexterm>
<indexterm significance="normal"><primary>legacy systems</primary></indexterm>
<indexterm significance="normal"><primary>access control</primary></indexterm>
The initial goal of many network administrators is often to create and use a centralized identity management
system. It is often assumed that such a centralized system will use a single authentication infrastructure
that can be used by all information systems. The Microsoft Windows NT4 security domain architecture and the
Micrsoft active directory service are often put forward as the ideal foundation for such a system. It is
conceptually simple to install an external authentication agent on each of the disparate infromation systems
that can then use the Microsoft (NT4 domain or ads service) for user authentication and access control. The
wonderful dream of a single centralized authentication service is commonly broken when realities are realized.
The problem with legacy systems is often the inability to externalize the authentication and access control
system it uses because its implementation will be excessively invasive from a re-engineering perspective, or
because application software has built-in dependencies on particular elements of the way user authentication
and access control were designed and built.
</para>

<para>
<indexterm significance="normal"><primary>meta-directory</primary></indexterm>
<indexterm significance="normal"><primary>credentials</primary></indexterm>
<indexterm significance="normal"><primary>disparate information systems</primary></indexterm>
<indexterm significance="normal"><primary>management procedures</primary></indexterm>
<indexterm significance="normal"><primary>work-flow protocol</primary></indexterm>
<indexterm significance="normal"><primary>rights</primary></indexterm>
<indexterm significance="normal"><primary>privileges</primary></indexterm>
<indexterm significance="normal"><primary>provisioned</primary></indexterm>
Over the past decade an industry has been developed around the various methods that have been built to get
around the key limitations of legacy information technology systems. One approach that is often used involves
the use of a meta-directory. The meta-directory stores user credentials for all disparate information systems
in the format that is particular to each system. An elaborate set of management procedures is coupled with a
rigidly enforced work-flow protocol for managing user rights and privileges within the maze of systems that
are provisioned by the new infrastructure makes possible user access to all systems using a single set of user
credentials.
</para>

<para>
<indexterm significance="normal"><primary>Organization for the Advancement of Structured Information Standards</primary><see>OASIS</see></indexterm>
<indexterm significance="normal"><primary>Security Assertion Markup Language</primary><see>SAML</see></indexterm>
<indexterm significance="normal"><primary>Federated Identity Management</primary><see>FIM</see></indexterm>
<indexterm significance="normal"><primary>secure access</primary></indexterm>
The Organization for the Advancement of Structured Information Standards (OASIS) has developed the Security
Assertion Markup Language (SAML), a structured method for communication of authentication information. The
over-all umbrella name for the technologies and methods that deploy SAML is called Federated Identity
Management (FIM). FIM depends on each system in the complex maze of disparate information systems to
authenticate their respective users and vouch for secure access to the services each provides.
</para>

<para>
<indexterm significance="normal"><primary>Simple Object Access Protocol</primary><see>SOAP</see></indexterm>
<indexterm significance="normal"><primary>federated organizations</primary></indexterm>
<indexterm significance="normal"><primary>Liberty Alliance</primary></indexterm>
<indexterm significance="normal"><primary>federated-identity</primary></indexterm>
<indexterm significance="normal"><primary/></indexterm>
<indexterm significance="normal"><primary/></indexterm>
SAML documents can be wrapped in a Simple Object Access Protocol (SOAP) message for the computer-to-computer
communications needed for Web services. Or they may be passed between Web servers of federated organizations
that share live services. The Liberty Alliance, an industry group formed to promote federated-identity
standards, has adopted SAML 1.1 as part of its application framework. Microsoft and IBM have proposed an
alternative specification called WS-Security. Some believe that the competing technologies and methods may
converge when the SAML 2.0 standard is introduced. A few Web access-management products support SAML today,
but implemention of the technology mostly requires customization to integrate applications and develop user
interfaces. In a nust-shell, that is why FIM is a big and growing industry.
</para>

<para>
<indexterm significance="normal"><primary>interoperability</primary></indexterm>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
<indexterm significance="normal"><primary>GSSAPI</primary></indexterm>
<indexterm significance="normal"><primary>general security service application programming interface</primary><see>GSSAPI</see></indexterm>
Ignoring the bigger picture, which is beyond the scope of this book, the migration of all user and group
management to a centralized system is a step in the right direction. It is essential for interoperability
reasons to locate the identity management system data in a directory such as Microsoft Active Directory
Service (ADS), or any proprietary or open source system that provides a standard protocol for information
access (such as LDAP) and that can be coupled with a flexible array of authentication mechanisms (such as
kerberos) that use the protocols that are defined by the various general security service application
programming interface (GSSAPI) services.
</para>

<para>
<indexterm significance="normal"><primary>OpenLDAP</primary></indexterm>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
<indexterm significance="normal"><primary>authentication agents</primary></indexterm>
A growing number of companies provide authentication agents for disparate legacy platforms to permit the use
of LDAP systems. Thus the use of OpenLDAP, the dominant open source software implementation of the light
weight directory access protocol standard. This fact, means that by providing support in Samba for the use of
LDAP and Microsoft ADS make Samba a highly scalable and forward reaching organizational networking technology.
</para>

<para>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
<indexterm significance="normal"><primary>authentication architecture</primary></indexterm>
<indexterm significance="normal"><primary>ntlm_auth</primary></indexterm>
<indexterm significance="normal"><primary>SQUID</primary></indexterm>
<indexterm significance="normal"><primary>FIM</primary></indexterm>
Microsoft ADS provides purely proprietary services that, with limitation, can be extended to provide a
centralized authentication infrastructure. Samba plus LDAP provides a similar opportunity for extension of a
centralized authentication architecture, but it is the fact that the Samba Team are pro-active in introducing
the extension of authentication services, using LDAP or otherwise, to applications such as SQUID (the open
source proxy server) through tools such as the <literal>ntlm_auth</literal> utility, that does much to create
sustainable choice and competition in the FIM market place.
</para>

<para>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
<indexterm significance="normal"><primary>OpenLDAP</primary></indexterm>
<indexterm significance="normal"><primary>identity information</primary></indexterm>
Primary domain control, if it is to be scalable to meet the needs of large sites, must therefore be capable of
using LDAP. The rapid adoption of OpenLDAP, and Samba configurations that use it, is ample proof that the era
of the directory has started. Samba-3 does not demand the use of LDAP, but the demand for a mechanism by which
user and group identity information can be distributed makes it an an unavoidable option.
</para>

<para>
<indexterm significance="normal"><primary>BDC</primary></indexterm>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
<indexterm significance="normal"><primary>e-Directory</primary></indexterm>
At this time, the use of Samba based BDCs, necessitates the use of LDAP. The most commonly used LDAP
implementation used by Samba sites is OpenLDAP. It is possible to use any standards compliant LDAP server.
Those known to work includes those manufactured by: IBM, CA, Novell (e-Directory), and others.
</para>

</sect1>

<sect1>
<title>Basics of Domain Control</title>

<para>
<indexterm significance="normal"><primary>domain control</primary></indexterm>
Over the years, public perceptions of what domain control really is has taken on an almost mystical nature.
Before we branch into a brief overview of domain control, there are three basic types of domain controllers.
</para>

<sect2>
<title>Domain Controller Types</title>

<itemizedlist>
	<listitem><para>NT4 style Primary Domain Controller</para></listitem>
	<listitem><para>NT4 style Backup Domain Controller</para></listitem>
	<listitem><para>ADS Domain Controller</para></listitem>
</itemizedlist>

<para>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>powerful</primary></indexterm>
<indexterm significance="normal"><primary>network</primary><secondary>performance</secondary></indexterm>
<indexterm significance="normal"><primary>domain</primary><secondary>member</secondary><secondary>server</secondary></indexterm>
The <emphasis>Primary Domain Controller</emphasis> or PDC plays an important role in MS Windows NT4. In
Windows 200x domain control architecture, this role is held by domain controllers.  Folklore dictates that
because of its role in the MS Windows network, the domain controller should be the most powerful and most
capable machine in the network.  As strange as it may seem to say this here, good overall network performance
dictates that the entire infrastructure needs to be balanced. It is advisable to invest more in standalone
(domain member) servers than in the domain controllers.
</para>

<para>
<indexterm significance="normal"><primary>SAM</primary></indexterm>
<indexterm significance="normal"><primary>BDC</primary></indexterm>
<indexterm significance="normal"><primary>authenticatior</primary></indexterm>
<indexterm significance="normal"><primary>synchronization</primary></indexterm>
<indexterm significance="normal"><primary>Security Account Manager</primary><see>SAM</see></indexterm>
In the case of MS Windows NT4-style domains, it is the PDC that initiates a new domain control database.
This forms a part of the Windows registry called the Security Account Manager (SAM). It plays a key
part in NT4-type domain user authentication and in synchronization of the domain authentication
database with BDCs. 
</para>

<para>
<indexterm significance="normal"><primary>domain</primary><secondary>controller</secondary><tertiary>hierarchy</tertiary></indexterm>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
<indexterm significance="normal">account<primary/><secondary>backend</secondary></indexterm>
<indexterm significance="normal"><primary>machine account</primary></indexterm>
With MS Windows 200x Server-based Active Directory domains, one domain controller initiates a potential
hierarchy of domain controllers, each with its own area of delegated control. The master domain
controller has the ability to override any downstream controller, but a downline controller has
control only over its downline. With Samba-3, this functionality can be implemented using an
LDAP-based user and machine account backend.
</para>

<para>
<indexterm significance="normal"><primary>backend database</primary></indexterm>
<indexterm significance="normal"><primary>registry</primary></indexterm>
New to Samba-3 is the ability to use a backend database that holds the same type of data as the NT4-style SAM
database (one of the registry files)<footnote><para>See also <link linkend="passdb">Account Information
Databases</link>.</para>.</footnote>
</para>

<para>
<indexterm significance="normal"><primary>BDC</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
<indexterm significance="normal"><primary>authentication</primary></indexterm>
<indexterm significance="normal"><primary>netlogon</primary></indexterm>
<indexterm significance="normal"><primary>name lookup</primary></indexterm>
The <emphasis>Backup Domain Controller</emphasis> or BDC plays a key role in servicing network authentication
requests. The BDC is biased to answer logon requests in preference to the PDC.  On a network segment that has
a BDC and a PDC, the BDC will most likely service network logon requests. The PDC will answer network logon
requests when the BDC is too busy (high load). When a user logs onto a Windows domain member client the
workstation will query the network to locate the nearest network logon server. Where a WINS server is used,
this is done via a query to the WINS server. If a netlogon server can not be found from the WINS query, or in
the absence of a WINS server, the workstation will perform a NetBIOS name lookup via a mailslot broadcast over
the UDP broadcast protocol. This means that the netlogon server that the windows client will use is influenced
by a number of variables, thus there is no simple determinant of whether a PDC or a BDC will serve a
particular logon authentication request.
</para>

<para>
<indexterm significance="normal"><primary>promote</primary></indexterm>
<indexterm significance="normal"><primary>demote</primary></indexterm>
A Windows NT4 BDC can be promoted to a PDC. If the PDC is online at the time that a BDC is promoted to PDC,
the previous PDC is automatically demoted to a BDC. With Samba-3, this is not an automatic operation; the PDC
and BDC must be manually configured, and other appropriate changes also need to be made.
</para>

<para>
<indexterm significance="normal"><primary>domain</primary><secondary>controller</secondary><tertiary>convert</tertiary></indexterm>
With MS Windows NT4, a decision is made at installation to determine what type of machine the server will be.
It is possible to promote a BDC to a PDC, and vice versa. The only method Microsoft provide to convert a
Windows NT4 domain controller to a domain member server or a standalone server is to reinstall it. The install
time choices offered are:
</para>

<itemizedlist>
	<listitem><para><emphasis>Primary Domain Controller</emphasis> <?latex --- ?> the one that seeds the domain SAM.</para></listitem>
	<listitem><para><emphasis>Backup Domain Controller</emphasis> <?latex --- ?> one that obtains a copy of the domain SAM.</para></listitem>
	<listitem><para><emphasis>Domain Member Server</emphasis> <?latex --- ?> one that has no copy of the domain SAM; rather
		it obtains authentication from a domain controller for all access controls.</para></listitem>
	<listitem><para><emphasis>Standalone Server</emphasis> <?latex --- ?> one that plays no part in SAM synchronization,
		has its own authentication database, and plays no role in domain security.</para></listitem>
</itemizedlist>

<note><para>
<indexterm significance="normal"><primary>promote</primary></indexterm>
Algin Technology LLC provide a commercial tool that makes it possible to promote a Windows NT4 standalone
server to a PDC or a BDC, and also permits this process to be reversed. Refer to the <ulink url="http://utools.com/UPromote.asp">Algin</ulink> web site for further information.
</para></note>

<para>
<indexterm significance="normal"><primary>domain</primary><secondary>control</secondary><tertiary>role</tertiary></indexterm>
<indexterm significance="normal"><primary>native member</primary></indexterm>
Samba-3 servers can readily be converted to and from domain controller roles through simple changes to the
<filename moreinfo="none">smb.conf</filename> file. Samba-3 is capable of acting fully as a native member of a Windows 200x server Active
Directory domain.
</para>

<para>
<indexterm significance="normal"><primary>convert</primary><secondary>domain member server</secondary></indexterm>
For the sake of providing a complete picture, MS Windows 2000 domain control configuration is done after the server has been
installed. Please refer to Microsoft documentation for the procedures that should be followed to convert a
domain member server to or from a domain control, and to install or remove active directory service support.
</para>

<para>
<indexterm significance="normal"><primary>replication</primary><secondary>SAM</secondary></indexterm>
<indexterm significance="normal"><primary>SAM</primary><secondary>replication</secondary></indexterm>
New to Samba-3 is the ability to function fully as an MS Windows NT4-style domain controller,
excluding the SAM replication components. However, please be aware that Samba-3 also supports the
MS Windows 200x domain control protocols.
</para>

<para>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
At this time any appearance that Samba-3 is capable of acting as a <emphasis>domain controller</emphasis> in
native ADS mode is limited and experimental in nature.  This functionality should not be used until the Samba
Team offers formal support for it.  At such a time, the documentation will be revised to duly reflect all
configuration and management requirements. Samba can act as a NT4-style domain controller in a Windows 2000/XP
environment. However, there are certain compromises:
</para>

<itemizedlist>
	<listitem><para>No machine policy files.</para></listitem>
	<listitem><para>No Group Policy Objects.</para></listitem>
	<listitem><para>No synchronously executed Active Directory logon scripts.</para></listitem>
	<listitem><para>Can't use Active Directory management tools to manage users and machines.</para></listitem>
	<listitem><para>Registry changes tattoo the main registry, while with Active Directory they do not leave
		permanent changes in effect.</para></listitem>
	<listitem><para>Without Active Directory you cannot perform the function of exporting specific
		applications to specific users or groups.</para></listitem>
</itemizedlist>

</sect2>

<sect2>
<title>Preparing for Domain Control</title>

<para>
<indexterm significance="normal"><primary>standalone</primary></indexterm>
<indexterm significance="normal"><primary>workgroup</primary></indexterm>
<indexterm significance="normal"><primary>member</primary></indexterm>
<indexterm significance="normal"><primary>security</primary></indexterm>
There are two ways that MS Windows machines may interact with each other, with other servers,
and with domain controllers: either as <emphasis>standalone</emphasis> systems, more commonly
called <emphasis>workgroup</emphasis> members, or as full participants in a security system,
more commonly called <emphasis>domain</emphasis> members.
</para>

<para>
<indexterm significance="normal"><primary>workgroup</primary></indexterm>
<indexterm significance="normal"><primary>workgroup</primary><secondary>membership</secondary></indexterm>
<indexterm significance="normal"><primary>machine trust account</primary></indexterm>
It should be noted that workgroup membership involves no special configuration other than the machine being
configured so the network configuration has a commonly used name for its workgroup entry. It is not uncommon
for the name WORKGROUP to be used for this. With this mode of configuration, there are no Machine Trust
Accounts, and any concept of membership as such is limited to the fact that all machines appear in the network
neighborhood to be logically grouped together. Again, just to be clear: <emphasis>workgroup mode does not
involve security machine accounts</emphasis>.
</para>

<para>
<indexterm significance="normal"><primary>domain membership</primary></indexterm>
<indexterm significance="normal"><primary>machine trust account</primary><secondary>password</secondary></indexterm>
<indexterm significance="normal"><primary>trigger</primary></indexterm>
Domain member machines have a machine trust account in the domain accounts database. A special procedure
must be followed on each machine to effect domain membership. This procedure, which can be done
only by the local machine Administrator account, creates the domain machine account (if it does
not exist), and then initializes that account. When the client first logs onto the
domain, a machine trust account password change will be automatically triggered.
</para>

<note><para>
<indexterm significance="normal"><primary>domain member</primary></indexterm>
When Samba is configured as a domain controller, secure network operation demands that
all MS Windows NT4/200x/XP Professional clients should be configured as domain members.
If a machine is not made a member of the domain, then it will operate like a workgroup
(standalone) machine. Please refer to <link linkend="domain-member">Domain Membership</link>, for
information regarding domain membership.
</para></note>

<para>
The following are necessary for configuring Samba-3 as an MS Windows NT4-style PDC for MS Windows
NT4/200x/XP clients:
</para>

<itemizedlist>
	<listitem><para>Configuration of basic TCP/IP and MS Windows networking.</para></listitem>
	<listitem><para>Correct designation of the server role (<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SECURITY" xlink:href="smb.conf.5.html#SECURITY">security = user</link>).</para></listitem>
	<listitem><para>Consistent configuration of name resolution.<footnote><para>See <link linkend="NetworkBrowsing">Network Browsing</link>, and 
		<link linkend="integrate-ms-networks">Integrating MS Windows Networks with Samba</link>.</para></footnote></para></listitem>
	<listitem><para>Domain logons for Windows NT4/200x/XP Professional clients.</para></listitem>
	<listitem><para>Configuration of roaming profiles or explicit configuration to force local profile usage.</para></listitem>
	<listitem><para>Configuration of network/system policies.</para></listitem>
	<listitem><para>Adding and managing domain user accounts.</para></listitem>
	<listitem><para>Configuring MS Windows NT4/2000 Professional and Windows XP Professional client machines to become domain members.</para></listitem>
</itemizedlist>

<para>
The following provisions are required to serve MS Windows 9x/Me clients:
</para>

<itemizedlist>
	<listitem><para>Configuration of basic TCP/IP and MS Windows networking.</para></listitem>
	<listitem><para>Correct designation of the server role (<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SECURITY" xlink:href="smb.conf.5.html#SECURITY">security = user</link>).</para></listitem>
	<listitem><para>Network logon configuration (since Windows 9x/Me/XP Home are not technically domain
	members, they do not really participate in  the security aspects of Domain logons as such).</para></listitem>
	<listitem><para>Roaming profile configuration.</para></listitem>
	<listitem><para>Configuration of system policy handling.</para></listitem>
	<listitem><para>Installation of the network driver <quote>Client for MS Windows Networks</quote> and configuration
	to log onto the domain.</para></listitem>
	<listitem><para>Placing Windows 9x/Me clients in user-level security <?latex --- ?> if it is desired to allow
	all client-share access to be controlled according to domain user/group identities.</para></listitem>
	<listitem><para>Adding and managing domain user accounts.</para></listitem>
</itemizedlist>

<note><para>
<indexterm significance="normal"><primary>roaming profiles</primary></indexterm>
<indexterm significance="normal"><primary>account policies</primary></indexterm>
Roaming profiles and system/network policies are advanced network administration topics
that are covered in <link linkend="ProfileMgmt">Desktop Profile Management</link> and
<link linkend="PolicyMgmt">System and Account Policies</link> of this document. However, these are not
necessarily specific to a Samba PDC as much as they are related to Windows NT networking concepts.
</para></note>

<para>
A domain controller is an SMB/CIFS server that:
</para>

<itemizedlist>
	<listitem><para>
	<indexterm significance="normal"><primary>NetBIOS</primary><secondary>brooadcast</secondary></indexterm>
	<indexterm significance="normal"><primary>WINS</primary></indexterm>
	<indexterm significance="normal"><primary>UDP</primary></indexterm>
	<indexterm significance="normal"><primary>DNS</primary></indexterm>
	<indexterm significance="normal"><primary>active directory</primary></indexterm>
	Registers and advertises itself as a domain controller (through NetBIOS broadcasts
	as well as by way of name registrations either by Mailslot Broadcasts over UDP broadcast,
	to a WINS server over UDP unicast, or via DNS and Active Directory).
	</para></listitem>

	<listitem><para>
	<indexterm significance="normal"><primary>NETLOGON</primary></indexterm>
	<indexterm significance="normal"><primary>LanMan logon service</primary></indexterm>
	Provides the NETLOGON service. (This is actually a collection of services that runs over
	multiple protocols. These include the LanMan logon service, the Netlogon service,
	the Local Security Account service, and variations of them.)
	</para></listitem>

	<listitem><para>
	Provides a share called NETLOGON.
	</para></listitem>
</itemizedlist>

<para>
<indexterm significance="normal"><primary>domain</primary><secondary>master</secondary><tertiary>browser</tertiary></indexterm>
<indexterm significance="normal"><primary>local</primary><secondary>master</secondary><tertiary>browser</tertiary></indexterm>
<indexterm significance="normal"><primary>DMB</primary></indexterm>
<indexterm significance="normal"><primary>LMB</primary></indexterm>
<indexterm significance="normal"><primary>browse list</primary></indexterm>
It is rather easy to configure Samba to provide these. Each Samba domain controller must provide the NETLOGON
service that Samba calls the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="DOMAINLOGONS" xlink:href="smb.conf.5.html#DOMAINLOGONS">domain logons</link> functionality (after the name of the
parameter in the <filename moreinfo="none">smb.conf</filename> file). Additionally, one server in a Samba-3 domain must advertise itself as the
domain master browser.<footnote><para>See <link linkend="NetworkBrowsing">Network
Browsing</link>.</para></footnote> This causes the PDC to claim a domain-specific NetBIOS name that identifies
it as a DMB for its given domain or workgroup. Local master browsers (LMBs) in the same domain or workgroup on
broadcast-isolated subnets then ask for a complete copy of the browse list for the whole wide-area network.
Browser clients then contact their LMB, and will receive the domain-wide browse list instead of just the list
for their broadcast-isolated subnet.
</para>

</sect2>
</sect1>

<sect1>
<title>Domain Control: Example Configuration</title>

<para>
The first step in creating a working Samba PDC is to understand the parameters necessary
in <filename moreinfo="none">smb.conf</filename>. An example <filename moreinfo="none">smb.conf</filename> for acting as a PDC can be found in <link linkend="pdc-example">the
smb.conf file for an example PDC</link>. 
</para>

<example id="pdc-example">
<title>smb.conf for being a PDC</title>
<simplelist>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>netbios name</indexterm></member>
<member><indexterm>workgroup</indexterm></member>
<member><indexterm>passdb backend</indexterm><parameter>passdb backend = tdbsam</parameter></member>
<member><indexterm>os level</indexterm><parameter>os level = 33</parameter></member>
<member><indexterm>preferred master</indexterm><parameter>preferred master = auto</parameter></member>
<member><indexterm>domain master</indexterm><parameter>domain master = yes</parameter></member>
<member><indexterm>local master</indexterm><parameter>local master = yes</parameter></member>
<member><indexterm>security</indexterm><parameter>security = user</parameter></member>
<member><indexterm>domain logons</indexterm><parameter>domain logons = yes</parameter></member>
<member><indexterm>logon path</indexterm><parameter>logon path = \\%N\profiles\%U</parameter></member>
<member><indexterm>logon drive</indexterm><parameter>logon drive = H:</parameter></member>
<member><indexterm>logon home</indexterm><parameter>logon home = \\homeserver\%U\winprofile</parameter></member>
<member><indexterm>logon script</indexterm><parameter>logon script = logon.cmd</parameter></member>

<member> </member><member><parameter>[netlogon]</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /var/lib/samba/netlogon</parameter></member>
<member><indexterm>read only</indexterm><parameter>read only = yes</parameter></member>
<member><indexterm>write list</indexterm></member>

<member> </member><member><parameter>[profiles]</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /var/lib/samba/profiles</parameter></member>
<member><indexterm>read only</indexterm><parameter>read only = no</parameter></member>
<member><indexterm>create mask</indexterm><parameter>create mask = 0600</parameter></member>
<member><indexterm>directory mask</indexterm><parameter>directory mask = 0700</parameter></member>
</simplelist>
</example>

<para>
The basic options shown in <link linkend="pdc-example">this example</link> are explained as follows:
</para>

<variablelist>
	<varlistentry><term>passdb backend </term>
		<listitem><para>
		<indexterm significance="normal"><primary>group</primary><secondary>account</secondary></indexterm>
		<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
		<indexterm significance="normal"><primary>tdbsam</primary></indexterm>
		<indexterm significance="normal"><primary>ldapsam</primary></indexterm>
		<indexterm significance="normal"><primary>guest</primary></indexterm>
		<indexterm significance="normal"><primary>default accounts</primary></indexterm>
		This contains all the user and group account information. Acceptable values for a PDC
		are: <emphasis>smbpasswd, tdbsam, and ldapsam</emphasis>. The <quote>guest</quote> entry provides
		default accounts and is included by default; there is no need to add it explicitly.
		</para>

		<para>
		<indexterm significance="normal"><primary>passdb backend</primary></indexterm>
		<indexterm significance="normal"><primary>distributed</primary></indexterm>
		<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
		<indexterm significance="normal"><primary>tdbsam</primary></indexterm>
		Where use of BDCs is intended, the only logical choice is
		to use LDAP so the passdb backend can be distributed. The tdbsam and smbpasswd files
		cannot effectively be distributed and therefore should not be used.
		</para></listitem>
	</varlistentry>

	<varlistentry><term>Domain Control Parameters </term>
		<listitem><para>
		<indexterm significance="normal"><primary>os level</primary></indexterm>
		<indexterm significance="normal"><primary>preferred master</primary></indexterm>
		<indexterm significance="normal"><primary>domain master</primary></indexterm>
		<indexterm significance="normal"><primary>network</primary><secondary>logon</secondary></indexterm>
		The parameters <emphasis>os level, preferred master, domain master, security, 
		encrypt passwords</emphasis>, and <emphasis>domain logons</emphasis> play a central role in assuring domain
		control and network logon support.
		</para>

		<para>
		<indexterm significance="normal"><primary>DMB</primary></indexterm>
		<indexterm significance="normal"><primary>encryped password</primary></indexterm>
		The <emphasis>os level</emphasis> must be set at or above a value of 32. A domain controller
		must be the DMB, must be set in <emphasis>user</emphasis> mode security,
		must support Microsoft-compatible encrypted passwords, and must provide the network logon
		service (domain logons). Encrypted passwords must be enabled. For more details on how 
		to do this, refer to <link linkend="passdb">Account Information Databases</link>.
		</para></listitem>
	</varlistentry>

	<varlistentry><term>Environment Parameters </term>
		<listitem><para>
		<indexterm significance="normal"><primary>logon path</primary></indexterm>
		<indexterm significance="normal"><primary>logon home</primary></indexterm>
		<indexterm significance="normal"><primary>logon drive</primary></indexterm>
		<indexterm significance="normal"><primary>logon script</primary></indexterm>
		The parameters <emphasis>logon path, logon home, logon drive</emphasis>, and <emphasis>logon script</emphasis> are
		environment support settings that help to facilitate client logon operations and that help
		to provide automated control facilities to ease network management overheads. Please refer
		to the man page information for these parameters.
		</para></listitem>
	</varlistentry>

	<varlistentry><term>NETLOGON Share </term>
		<listitem><para>
		<indexterm significance="normal"><primary>NETLOGON</primary></indexterm>
		<indexterm significance="normal"><primary>logon processing</primary></indexterm>
		<indexterm significance="normal"><primary>domain logon</primary></indexterm>
		<indexterm significance="normal"><primary>domain membership</primary></indexterm>
		<indexterm significance="normal"><primary>group policy</primary></indexterm>
		<indexterm significance="normal"><primary>NTConfig.POL</primary></indexterm>
		The NETLOGON share plays a central role in domain logon and domain membership support.
		This share is provided on all Microsoft domain controllers. It is used to provide logon
		scripts, to store group policy files (NTConfig.POL), as well as to locate other common
		tools that may be needed for logon processing. This is an essential share on a domain controller.
		</para></listitem>
	</varlistentry>

	<varlistentry><term>PROFILE Share </term>
		<listitem><para>
		<indexterm significance="normal"><primary>desktop profile</primary></indexterm>
		<indexterm significance="normal"><primary>VFS</primary></indexterm>
		<indexterm significance="normal"><primary>fake_permissions</primary></indexterm>
		<indexterm significance="normal"><primary>profile</primary></indexterm>
		<indexterm significance="normal"><primary/></indexterm>
		This share is used to store user desktop profiles. Each user must have a directory at the root
		of this share. This directory must be write-enabled for the user and must be globally read-enabled.
		Samba-3 has a VFS module called <quote>fake_permissions</quote> that may be installed on this share. This will
		allow a Samba administrator to make the directory read-only to everyone. Of course this is useful
		only after the profile has been properly created.
		</para></listitem>
	</varlistentry>
</variablelist>

<note><para>
The above parameters make for a full set of functionality that may define the server's mode
of operation. The following <filename moreinfo="none">smb.conf</filename> parameters are the essentials alone:
</para>

<para>
<simplelist>
<member><indexterm>netbios name</indexterm><parameter>netbios name = BELERIAND</parameter></member>
<member><indexterm>workgroup</indexterm><parameter>workgroup = MIDEARTH</parameter></member>
<member><indexterm>domain logons</indexterm><parameter>domain logons = Yes</parameter></member>
<member><indexterm>domain master</indexterm><parameter>domain master = Yes</parameter></member>
<member><indexterm>security</indexterm><parameter>security = User</parameter></member>
</simplelist>
</para>

<para>
The additional parameters shown in the longer listing in this section just make for
a more complete explanation.
</para></note>

</sect1>

<sect1>
<title>Samba ADS Domain Control</title>

<para>
<indexterm significance="normal"><primary>active directory</primary></indexterm>
Samba-3 is not, and cannot act as, an Active Directory server. It cannot truly function as an Active Directory
PDC. The protocols for some of the functionality of Active Directory domain controllers has been partially
implemented on an experimental only basis. Please do not expect Samba-3 to support these protocols. Do not
depend on any such functionality either now or in the future. The Samba Team may remove these experimental
features or may change their behavior. This is mentioned for the benefit of those who have discovered secret
capabilities in Samba-3 and who have asked when this functionality will be completed. The answer is maybe
someday or maybe never!
</para>

<para>
<indexterm significance="normal"><primary>domain controllers</primary></indexterm>
<indexterm significance="normal"><primary>active directory</primary></indexterm>
To be sure, Samba-3 is designed to provide most of the functionality that Microsoft Windows NT4-style
domain controllers have. Samba-3 does not have all the capabilities of Windows NT4, but it does have
a number of features that Windows NT4 domain controllers do not have. In short, Samba-3 is not NT4 and it
is not Windows Server 200x: it is not an Active Directory server. We hope this is plain and simple
enough for all to understand.
</para>

</sect1>

<sect1>
<title>Domain and Network Logon Configuration</title>

<para>
<indexterm significance="normal"><primary>domain logon</primary></indexterm>
The subject of network or domain logons is discussed here because it forms
an integral part of the essential functionality that is provided by a domain controller.
</para>

<sect2>
<title>Domain Network Logon Service</title>

<para>
<indexterm significance="normal"><primary>domain logon</primary></indexterm>
All domain controllers must run the netlogon service (<emphasis>domain logons</emphasis>
in Samba). One domain controller must be configured with <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="DOMAINMASTER" xlink:href="smb.conf.5.html#DOMAINMASTER">domain master = Yes</link>
(the PDC); on all BDCs set the parameter <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="DOMAINMASTER" xlink:href="smb.conf.5.html#DOMAINMASTER">domain master = No</link>.
</para>

<sect3>
<title>Example Configuration</title>

<example id="PDC-config">
<title>smb.conf for being a PDC</title>
<simplelist>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>domain logons</indexterm><parameter>domain logons = Yes</parameter></member>
<member><indexterm>domain master</indexterm><parameter>domain master = (Yes on PDC, No on BDCs)</parameter></member>

<member> </member><member><parameter>[netlogon]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = Network Logon Service</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /var/lib/samba/netlogon</parameter></member>
<member><indexterm>guest ok</indexterm><parameter>guest ok = Yes</parameter></member>
<member><indexterm>browseable</indexterm><parameter>browseable = No</parameter></member>
</simplelist>
</example>

</sect3>
<sect3>
<title>The Special Case of MS Windows XP Home Edition</title>

<para>
<indexterm significance="normal"><primary>Windows XP Home edition</primary></indexterm>
To be completely clear: If you want MS Windows XP Home Edition to integrate with your
MS Windows NT4 or Active Directory domain security, understand it cannot be done.
The only option is to purchase the upgrade from MS Windows XP Home Edition to
MS Windows XP Professional.
</para>

<note><para>
MS Windows XP Home Edition does not have the ability to join any type of domain
security facility. Unlike MS Windows 9x/Me, MS Windows XP Home Edition also completely
lacks the ability to log onto a network.
</para></note>

<para>
Now that this has been said, please do not ask the mailing list or email any of the
Samba Team members with your questions asking how to make this work. It can't be done.
If it can be done, then to do so would violate your software license agreement with
Microsoft, and we recommend that you do not do that.
</para>

</sect3>

<sect3>
<title>The Special Case of Windows 9x/Me</title>

<para>
<indexterm significance="normal"><primary>domain</primary></indexterm>
<indexterm significance="normal"><primary>workgroup</primary></indexterm>
<indexterm significance="normal"><primary>authentication</primary></indexterm>
<indexterm significance="normal"><primary>browsing</primary></indexterm>
<indexterm significance="normal"><primary>rights</primary></indexterm>
A domain and a workgroup are exactly the same in terms of network
browsing. The difference is that a distributable authentication
database is associated with a domain, for secure login access to a
network. Also, different access rights can be granted to users if they
successfully authenticate against a domain logon server. Samba-3 does this
now in the same way as MS Windows NT/200x.
</para>

<para>
<indexterm significance="normal"><primary>browsing</primary></indexterm>
The SMB client logging on to a domain has an expectation that every other
server in the domain should accept the same authentication information.
Network browsing functionality of domains and workgroups is identical and
is explained in this documentation under the browsing discussions.
It should be noted that browsing is totally orthogonal to logon support.
</para>

<para>
<indexterm significance="normal"><primary>single-logon</primary></indexterm>
<indexterm significance="normal"><primary>domain logons</primary></indexterm>
<indexterm significance="normal"><primary>network logon</primary></indexterm>
Issues related to the single-logon network model are discussed in this
section. Samba supports domain logons, network logon scripts, and user
profiles for MS Windows for Workgroups and MS Windows 9x/Me clients,
which are the focus of this section.
</para>

<para>
<indexterm significance="normal"><primary>broadcast request</primary></indexterm>
When an SMB client in a domain wishes to log on, it broadcasts requests for a logon server. The first one to
reply gets the job and validates its password using whatever mechanism the Samba administrator has installed.
It is possible (but ill advised) to create a domain where the user database is not shared between servers;
that is, they are effectively workgroup servers advertising themselves as participating in a domain. This
demonstrates how authentication is quite different from but closely involved with domains.
</para>

<para>
Using these features, you can make your clients verify their logon via
the Samba server, make clients run a batch file when they log on to
the network and download their preferences, desktop, and start menu.
</para>

<para><emphasis>
MS Windows XP Home edition is not able to join a domain and does not permit the use of domain logons.
</emphasis></para>

<para>
Before launching into the configuration instructions, it is worthwhile to look at how a Windows 9x/Me client
performs a logon:
</para>

<orderedlist inheritnum="ignore" continuation="restarts">
<listitem>
	<para>
	<indexterm significance="normal"><primary>DOMAIN&lt;1C&gt;</primary></indexterm>
	<indexterm significance="normal"><primary>logon server</primary></indexterm>
	The client broadcasts (to the IP broadcast address of the subnet it is in)
	a NetLogon request. This is sent to the NetBIOS name DOMAIN&lt;1C&gt; at the
	NetBIOS layer. The client chooses the first response it receives, which
	contains the NetBIOS name of the logon server to use in the format of 
	<filename moreinfo="none">\\SERVER</filename>. The <literal moreinfo="none">1C</literal> name is the name
	type that is registered by domain controllers (SMB/CIFS servers that provide
	the netlogon service).
	</para>
</listitem>

<listitem>
	<para>
	<indexterm significance="normal"><primary>IPC$</primary></indexterm>
	<indexterm significance="normal"><primary>SMBsessetupX</primary></indexterm>
	<indexterm significance="normal"><primary>SMBtconX</primary></indexterm>
	The client connects to that server, logs on (does an SMBsessetupX) and
	then connects to the IPC$ share (using an SMBtconX).
	</para>
</listitem>

<listitem>
	<para>
	<indexterm significance="normal"><primary>NetWkstaUserLogon</primary></indexterm>
	The client does a NetWkstaUserLogon request, which retrieves the name
	of the user's logon script. 
	</para>
</listitem>

<listitem>
	<para>
	The client then connects to the NetLogon share and searches for said script.	
	If it is found and can be read, it is retrieved and executed by the client.
	After this, the client disconnects from the NetLogon share.
	</para>
</listitem>

<listitem>
	<para>
	<indexterm significance="normal"><primary>NetUserGetInfo</primary></indexterm>
	<indexterm significance="normal"><primary>profile</primary></indexterm>
	The client sends a NetUserGetInfo request to the server to retrieve
	the user's home share, which is used to search for profiles. Since the
	response to the NetUserGetInfo request does not contain much more than	
	the user's home share, profiles for Windows 9x clients must reside in the user
	home directory.
	</para>
</listitem>

<listitem>
	<para>
	<indexterm significance="normal"><primary>profiles</primary></indexterm>
	The client connects to the user's home share and searches for the 
	user's profile. As it turns out, you can specify the user's home share as
	a share name and path. For example, <filename moreinfo="none">\\server\fred\.winprofile</filename>.
	If the profiles are found, they are implemented.
	</para>
</listitem>

<listitem>
	<para>
	<indexterm significance="normal"><primary>CONFIG.POL</primary></indexterm>
	The client then disconnects from the user's home share and reconnects to
	the NetLogon share and looks for <filename moreinfo="none">CONFIG.POL</filename>, the policies file. If this is
	found, it is read and implemented.
	</para>
</listitem>
</orderedlist>

<para>
The main difference between a PDC and a Windows 9x/Me logon server configuration is:
</para>

<itemizedlist>
<listitem><para>
	<indexterm significance="normal"><primary>password</primary><secondary>plaintext</secondary></indexterm>
	<indexterm significance="normal"><primary>plaintext password</primary></indexterm>
	Password encryption is not required for a Windows 9x/Me logon server. But note
	that beginning with MS Windows 98 the default setting is that plaintext
	password support is disabled. It can be re-enabled with the registry
	changes that are documented in <link linkend="PolicyMgmt">System and Account Policies</link>.
	</para></listitem>

	<listitem><para>
	<indexterm significance="normal"><primary>machine trust account</primary></indexterm>
	Windows 9x/Me clients do not require and do not use Machine Trust Accounts.
	</para></listitem>
</itemizedlist>

<para>
<indexterm significance="normal"><primary>network logon services</primary></indexterm>
A Samba PDC will act as a Windows 9x/Me logon server; after all, it does provide the
network logon services that MS Windows 9x/Me expect to find.
</para>

<note><para>
<indexterm significance="normal"><primary>sniffer</primary></indexterm>
Use of plaintext passwords is strongly discouraged. Where used they are easily detected
using a sniffer tool to examine network traffic.
</para></note>

</sect3>
</sect2>

<sect2>
<title>Security Mode and Master Browsers</title>

<para>
<indexterm significance="normal"><primary>security mode</primary></indexterm>
<indexterm significance="normal"><primary>user-mode security</primary></indexterm>
<indexterm significance="normal"><primary>share-mode security</primary></indexterm>
There are a few comments to make in order to tie up some loose ends. There has been much debate over the issue
of whether it is okay to configure Samba as a domain controller that operates with security mode other than
user-mode. The only security mode that will not work due to technical reasons is share-mode security. Domain
and server mode security are really just a variation on SMB user-level security.
</para>

<para>
<indexterm significance="normal"><primary>DOMAIN&lt;1C&gt;</primary></indexterm>
<indexterm significance="normal"><primary>DOMAIN&lt;1B&gt;</primary></indexterm>
<indexterm significance="normal"><primary>DMB</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>NetBIOS name</primary></indexterm>
<indexterm significance="normal"><primary>domain controller</primary></indexterm>
<indexterm significance="normal"><primary>election</primary></indexterm>
Actually, this issue is also closely tied to the debate on whether Samba must be the DMB for its workgroup
when operating as a domain controller.  In a pure Microsoft Windows NT domain, the PDC wins the election to be
the DMB, and then registers the DOMAIN&lt;1B&gt; NetBIOS name. This is not the name used by Windows clients
to locate the domain controller, all domain controllers register the DOMAIN&lt;1C&gt; name and Windows clients
locate a network logon server by seraching for the DOMAIN&lt;1C&gt; name.  A DMB is a Domain Master Browser
<?latex --- ?> see <link linkend="NetworkBrowsing">The Network Browsing Chapter</link>, <link linkend="DMB">Configuring WORKGROUP Browsing</link>; Microsoft PDCs expect to win the election to become the
DMB, if it loses that election it will report a continuous and rapid sequence of warning messages to its
Windows event logger complaining that it has lost the election to become a DMB.  For this reason, in networks
where a Samba server is the PDC it is wise to configure the Samba domain controller as the DMB.
</para>

<note><para>
<indexterm significance="normal"><primary>DOMAIN&lt;1D&gt;</primary></indexterm>
<indexterm significance="normal"><primary>synchronization</primary></indexterm>
<indexterm significance="normal"><primary>domain control</primary></indexterm>
<indexterm significance="normal"><primary>browse list management</primary></indexterm>
<indexterm significance="normal"><primary>network</primary><secondary>logon</secondary><tertiary>service</tertiary></indexterm>
SMB/CIFS servers that register the DOMAIN&lt;1C&gt; name do so because they provide the network logon
service. Server that register the DOMAIN&lt;1B&gt; name are DMBs <?latex --- ?> meaning that they are responsible
for browse list synchronization across all machines that have registered the DOMAIN&lt;1D&gt; name. The later
are LMBs that have the responsibility to listen to all NetBIOS name registrations that occur locally to their
own network segment. The network logon service (NETLOGON) is germane to domain control and has nothing to do
with network browsing and browse list management. The 1C and 1B/1D name services are orthogonal to each
other.
</para></note>

<para>
Now back to the issue of configuring a Samba domain controller to use a mode other than <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SECURITY" xlink:href="smb.conf.5.html#SECURITY">security = user</link>. If a Samba host is configured to use another SMB server or domain
controller in order to validate user connection requests, it is a fact that some other machine on the network
(the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PASSWORDSERVER" xlink:href="smb.conf.5.html#PASSWORDSERVER">password server</link>) knows more about the user than the Samba host. About 99 percent
of the time, this other host is a domain controller. Now to operate in domain mode security, the
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="WORKGROUP" xlink:href="smb.conf.5.html#WORKGROUP">workgroup</link> parameter must be set to the name of the Windows NT domain (which already
has a domain controller).  If the domain does not already have a domain controller, you do not yet have a
domain.
</para>

<para>
Configuring a Samba box as a domain controller for a domain that already by definition has a
PDC is asking for trouble. Therefore, you should always configure the Samba domain controller
to be the DMB for its domain and set <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SECURITY" xlink:href="smb.conf.5.html#SECURITY">security = user</link>.
This is the only officially supported mode of operation.
</para>

</sect2>

</sect1>

<sect1>
<title>Common Errors</title>

<sect2>
<title><quote>$</quote> Cannot Be Included in Machine Name</title>

<para>
<indexterm significance="normal"><primary>BSD</primary></indexterm>
<indexterm significance="normal"><primary>FreeBSD</primary></indexterm>
<indexterm significance="normal"><primary>/etc/passwd</primary></indexterm>
A machine account, typically stored in <filename moreinfo="none">/etc/passwd</filename>, takes the form of the machine
name with a <quote>$</quote> appended. Some BSD systems will not create a user with a <quote>$</quote> in the name.
Recent versions of FreeBSD have removed this limitation, but older releases are still in common use.
</para>

<para>
<indexterm significance="normal"><primary>vipw</primary></indexterm>
The problem is only in the program used to make the entry. Once made, it works perfectly.  Create a user
without the <quote>$</quote>. Then use <literal>vipw</literal> to edit the entry, adding the <quote>$</quote>.
Or create the whole entry with vipw if you like; make sure you use a unique user login ID.
</para>

<note><para>The machine account must have the exact name that the workstation has.</para></note>

<note><para>
The UNIX tool <literal>vipw</literal> is a common tool for directly editing the <filename moreinfo="none">/etc/passwd</filename> file.
The use of vipw will ensure that shadow files (where used) will remain current with the passwd file. This is
important for security reasons.
</para></note>

</sect2>

<sect2>
<title>Joining Domain Fails Because of Existing Machine Account</title>
		
<para>
<indexterm significance="normal"><primary>join domain</primary></indexterm>
<quote>I get told, `You already have a connection to the Domain....' or `Cannot join domain, the
credentials supplied conflict with an existing set...' when creating a Machine Trust Account.</quote>
</para>

<para>
This happens if you try to create a Machine Trust Account from the machine itself and already have a
connection (e.g., mapped drive) to a share (or IPC$) on the Samba PDC. The following command will remove all
network drive connections:
<screen format="linespecific">
<prompt moreinfo="none">C:\&gt; </prompt><userinput moreinfo="none">net use * /d</userinput>
</screen>
This will break all network connections.
</para>

<para>
Further, if the machine is already a <quote>member of a workgroup</quote> that is the same name as the domain
you are joining (bad idea), you will get this message. Change the workgroup name to something else <?latex --- ?>
it does not matter what <?latex --- ?> reboot, and try again.
</para>

</sect2>

<sect2>
<title>The System Cannot Log You On (C000019B)</title>

<para><quote>
I joined the domain successfully but after upgrading to a newer version of the Samba code I get the message,
<errorname>`The system cannot log you on (C000019B). Please try again or consult your system
administrator</errorname> when attempting to logon.'</quote>
</para>

<para>
<indexterm significance="normal"><primary>SID</primary></indexterm>
This occurs when the domain SID stored in the secrets.tdb database is changed. The most common cause of a
change in domain SID is when the domain name and/or the server name (NetBIOS name) is changed.  The only way
to correct the problem is to restore the original domain SID or remove the domain client from the domain and
rejoin. The domain SID may be reset using either the net or rpcclient utilities.
</para>

<para>
To reset or change the domain SID you can use the net command as follows:

<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">net getlocalsid 'OLDNAME'</userinput>
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">net setlocalsid 'SID'</userinput>
</screen>
</para>

<para>
Workstation Machine Trust Accounts work only with the domain (or network) SID. If this SID changes,
domain members (workstations) will not be able to log onto the domain. The original domain SID
can be recovered from the secrets.tdb file. The alternative is to visit each workstation to rejoin
it to the domain.
</para>

</sect2>

<sect2>
<title>The Machine Trust Account Is Not Accessible</title>

<para>
<quote>When I try to join the domain I get the message, <errorname>"The machine account 
for this computer either does not exist or is not accessible</errorname>." What's wrong?</quote>
</para>

<para>
This problem is caused by the PDC not having a suitable Machine Trust Account.  If you are using the
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="ADDMACHINESCRIPT" xlink:href="smb.conf.5.html#ADDMACHINESCRIPT">add machine script</link> method to create accounts, then this would indicate that it has not
worked. Ensure the domain admin user system is working.
</para>

<para>
Alternately, if you are creating account entries manually, then they have not been created correctly. Make
sure that you have the entry correct for the Machine Trust Account in <filename moreinfo="none">smbpasswd</filename> file on
the Samba PDC.  If you added the account using an editor rather than using the smbpasswd utility, make sure
that the account name is the machine NetBIOS name with a <quote>$</quote> appended to it (i.e.,
computer_name$). There must be an entry in both the POSIX UNIX system account backend as well as in the
SambaSAMAccount backend. The default backend for Samba-3 (i.e., the parameter <parameter moreinfo="none">passdb
backend</parameter> is not specified in the <filename moreinfo="none">smb.conf</filename> file, or if specified is set to
<literal moreinfo="none">smbpasswd</literal>, are respectively the <filename moreinfo="none">/etc/passwd</filename> and
<filename moreinfo="none">/etc/samba/smbpasswd</filename> (or <filename moreinfo="none">/usr/local/samba/lib/private/smbpasswd</filename> if
compiled using Samba Team default settings). The use of the <filename moreinfo="none">/etc/passwd</filename> can be overridden
by alternative settings in the NSS <filename moreinfo="none">/etc/nsswitch.conf</filename> file.
</para>

<para>
Some people have also reported that inconsistent subnet masks between the Samba server and the NT 
client can cause this problem.  Make sure that these are consistent for both client and server.
</para>
</sect2>

<sect2>
<title>Account Disabled</title>

<para><quote>When I attempt to log in to a Samba domain from a NT4/W200x workstation,
I get a message about my account being disabled.</quote></para>

<para>
Enable the user accounts with <userinput moreinfo="none">smbpasswd -e <replaceable>username</replaceable>
</userinput>. This is normally done as an account is created.
</para>

</sect2>

<sect2>
<title>Domain Controller Unavailable</title>
		
<para><quote>Until a few minutes after Samba has started, clients get the error `Domain Controller Unavailable'</quote></para>

<para>
A domain controller has to announce its role on the network. This usually takes a while. Be patient for up to 15 minutes,
then try again.
</para>
</sect2>

<sect2>
<title>Cannot Log onto Domain Member Workstation After Joining Domain</title>

<para>
<indexterm significance="normal"><primary>schannel</primary></indexterm>
<indexterm significance="normal"><primary>signing</primary></indexterm>
After successfully joining the domain, user logons fail with one of two messages: one to the
effect that the domain controller cannot be found; the other claims that the account does not
exist in the domain or that the password is incorrect. This may be due to incompatible
settings between the Windows client and the Samba-3 server for <emphasis>schannel</emphasis>
(secure channel) settings or <emphasis>smb signing</emphasis> settings. Check your Samba
settings for <emphasis>client schannel</emphasis>, <emphasis>server schannel</emphasis>,
<emphasis>client signing</emphasis>, <emphasis>server signing</emphasis> by executing:
<screen format="linespecific">
<literal>testparm -v | grep channel</literal> and looking for the value of these parameters.
</screen>
</para>

<para>
Also use the MMC <?latex --- ?> Local Security Settings. This tool is available from the
Control Panel. The Policy settings are found in the Local Policies/Security Options area and are prefixed by
<emphasis>Secure Channel:..., and Digitally sign...</emphasis>.
</para>

<para>
It is important that these be set consistently with the Samba-3 server settings.
</para>

</sect2>

</sect1>
</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="samba-bdc">

<chapterinfo>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
	<author>
<firstname>Volker</firstname><surname>Lendecke</surname>
<affiliation>
		<orgname>Samba Team</orgname>
		<address format="linespecific"><email>Volker.Lendecke@SerNet.DE</email></address>
</affiliation></author>
	<author>
<firstname>Guenther</firstname><surname>Deschner</surname>
<affiliation>
	<orgname>SuSE</orgname>
	<address format="linespecific"><email>gd@suse.de</email></address>
</affiliation><contrib>LDAP updates</contrib></author>
</chapterinfo>

<title>Backup Domain Control</title>

<para>
Before you continue reading this section, please make sure that you are comfortable
with configuring a Samba domain controller as described in <link linkend="samba-pdc">Domain Control</link>.
</para>

<sect1>
<title>Features and Benefits</title>

<para>
This is one of the most difficult chapters to summarize. It does not matter what we say here, for someone will
still draw conclusions and/or approach the Samba Team with expectations that are either not yet capable of
being delivered or that can be achieved far more effectively using a totally different approach. In the event
that you should have a persistent concern that is not addressed in this book, please email <ulink url="mailto:jht@samba.org">John H. Terpstra</ulink> clearly setting out your requirements and/or question, and
we will do our best to provide a solution.
</para>

<para>
<indexterm significance="normal"><primary>SAM backend</primary><secondary>LDAP</secondary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>BDC</primary></indexterm>
<indexterm significance="normal"><primary>LDAP</primary><secondary>slave</secondary></indexterm>
<indexterm significance="normal"><primary>scalability</primary></indexterm>
Samba-3 can act as a Backup Domain Controller (BDC) to another Samba Primary Domain Controller (PDC). A
Samba-3 PDC can operate with an LDAP account backend. The LDAP backend can be either a common master LDAP
server or a slave server. The use of a slave LDAP server has the benefit that when the master is down, clients
may still be able to log onto the network.  This effectively gives Samba a high degree of scalability and is
an effective solution for large organizations. If you use an LDAP slave server for a PDC, you will need to
ensure the master's continued availability <?latex --- ?> if the slave finds its master down at the wrong time,
you will have stability and operational problems.
</para>

<para>
<indexterm significance="normal"><primary>two-way</primary><secondary>propagation</secondary></indexterm>
<indexterm significance="normal"><primary>replication</primary><secondary>SAM</secondary></indexterm>
<indexterm significance="normal"><primary>non-LDAP</primary><secondary>backend</secondary></indexterm>
<indexterm significance="normal"><primary>propagate</primary></indexterm>
While it is possible to run a Samba-3 BDC with a non-LDAP backend, that backend must allow some form of
"two-way" propagation of changes from the BDC to the master.  At this time only LDAP delivers the capability
to propagate identity database changes from the BDC to the PDC. The BDC can use a slave LDAP server, while it
is preferable for the PDC to use as its primary an LDAP master server.
</para>

<para>
<indexterm significance="normal"><primary>non-LDAP</primary><secondary>backend</secondary></indexterm>
<indexterm significance="normal"><primary>SAM backend</primary><secondary>non-LDAP</secondary></indexterm>
<indexterm significance="normal"><primary>domain</primary><secondary>member</secondary><tertiary>server</tertiary></indexterm>
<indexterm significance="normal"><primary>BDC</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>trust account password</primary></indexterm>
<indexterm significance="normal"><primary>domain trust</primary></indexterm>
The use of a non-LDAP backend SAM database is particularly problematic because domain member
servers and workstations periodically change the Machine Trust Account password. The new
password is then stored only locally. This means that in the absence of a centrally stored
accounts database (such as that provided with an LDAP-based solution) if Samba-3 is running
as a BDC, the BDC instance of the domain member trust account password will not reach the
PDC (master) copy of the SAM. If the PDC SAM is then replicated to BDCs, this results in 
overwriting the SAM that contains the updated (changed) trust account password with resulting
breakage of the domain trust.
</para>

<para>
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary></indexterm>
<indexterm significance="normal"><primary>SAM backend</primary><secondary>ldapsam</secondary></indexterm>
<indexterm significance="normal"><primary>SAM backend</primary><secondary>tdbsam</secondary></indexterm>
<indexterm significance="normal"><primary>replication</primary><secondary>SAM</secondary></indexterm>
Considering the number of comments and questions raised concerning how to configure a BDC,
let's consider each possible option and look at the pros and cons for each possible solution.
<link linkend="pdc-bdc-table">The Domain Backend Account Distribution Options table below</link> lists 
possible design configurations for a PDC/BDC infrastructure.
</para>

<table frame="all" id="pdc-bdc-table"><title>Domain Backend Account Distribution Options</title>
<tgroup cols="3">
        <colspec align="center" colwidth="1*"/>
        <colspec align="center" colwidth="1*"/>
        <colspec align="left" colwidth="3*"/>

        <thead>
        <row><entry>PDC Backend</entry><entry>BDC Backend</entry><entry>Notes/Discussion</entry></row>
        </thead>
        <tbody>
        <row>
        <entry><para>Master LDAP Server</para></entry>
        <entry><para>Slave LDAP Server</para></entry>
        <entry><para>The optimal solution that provides high integrity. The SAM will be
		replicated to a common master LDAP server.</para></entry>
        </row>
        <row>
        <entry><para>Single Central LDAP Server</para></entry>
        <entry><para>Single Central LDAP Server</para></entry>
	<entry><para>
	A workable solution without failover ability. This is a usable solution, but not optimal. 
	</para></entry>
        </row>
        <row>
        <entry><para>tdbsam</para></entry>
        <entry><para>tdbsam + <literal>net rpc vampire</literal></para></entry>
        <entry><para>
	Does not work with Samba-3.0; Samba does not implement the
        server-side protocols required.
	</para></entry>
        </row>
        <row>
        <entry><para>tdbsam</para></entry>
        <entry><para>tdbsam + <literal>rsync</literal></para></entry>
        <entry><para>
	Do not use this configuration.
	Does not work because the TDB files are live and data may not
        have been flushed to disk.  Furthermore, this will cause
        domain trust breakdown.
	</para></entry>
        </row>
        <row>
        <entry><para>smbpasswd file</para></entry>
        <entry><para>smbpasswd file</para></entry>
        <entry><para>
	Do not use this configuration.
	Not an elegant solution due to the delays in synchronization
        and also suffers
        from the issue of domain trust breakdown.
	</para></entry>
        </row>
        </tbody>
</tgroup>
</table>

</sect1>

<sect1>
<title>Essential Background Information</title>

<para>
<indexterm significance="normal"><primary>domain controller</primary></indexterm>
<indexterm significance="normal"><primary>logon requests</primary></indexterm>
<indexterm significance="normal"><primary>LanMan</primary></indexterm>
<indexterm significance="normal"><primary>Netlogon</primary></indexterm>
A domain controller is a machine that is able to answer logon requests from network
workstations. Microsoft LanManager and IBM LanServer were two early products that
provided this capability. The technology has become known as the LanMan Netlogon service.
</para>

<para>
<indexterm significance="normal">network<primary/><secondary>logon</secondary><tertiary>service</tertiary></indexterm>
<indexterm significance="normal"><primary>Windows NT3.10</primary></indexterm>
When MS Windows NT3.10 was first released, it supported a new style of Domain Control
and with it a new form of the network logon service that has extended functionality.
This service became known as the NT NetLogon Service. The nature of this service has
changed with the evolution of MS Windows NT and today provides a complex array of
services that are implemented over an intricate spectrum of technologies.
</para>

<sect2>
<title>MS Windows NT4-style Domain Control</title>

<para>
<indexterm significance="normal"><primary>domain controller</primary></indexterm>
<indexterm significance="normal"><primary>authentication server</primary></indexterm>
<indexterm significance="normal"><primary>username</primary></indexterm>
<indexterm significance="normal"><primary>password</primary></indexterm>
<indexterm significance="normal"><primary>SAM</primary></indexterm>
<indexterm significance="normal"><primary>Security Account Manager</primary><see>SAM</see></indexterm>
<indexterm significance="normal"><primary>domain control database</primary><see>SAM</see></indexterm>
Whenever a user logs into a Windows NT4/200x/XP Professional workstation,
the workstation connects to a domain controller (authentication server) to validate that
the username and password the user entered are valid. If the information entered
does not match account information that has been stored in the domain
control database (the SAM, or Security Account Manager database), a set of error
codes is returned to the workstation that has made the authentication request.
</para>

<para>
<indexterm significance="normal"><primary>account information</primary></indexterm>
<indexterm significance="normal"><primary>machine accounts database</primary></indexterm>
<indexterm significance="normal"><primary>profile</primary></indexterm>
<indexterm significance="normal"><primary>network access profile</primary></indexterm>
<indexterm significance="normal"><primary>desktop profile</primary></indexterm>
When the username/password pair has been validated, the domain controller
(authentication server) will respond with full enumeration of the account information
that has been stored regarding that user in the user and machine accounts database
for that domain. This information contains a complete network access profile for
the user but excludes any information that is particular to the user's desktop profile,
or for that matter it excludes all desktop profiles for groups that the user may
belong to. It does include password time limits, password uniqueness controls,
network access time limits, account validity information, machine names from which the
user may access the network, and much more. All this information was stored in the SAM
in all versions of MS Windows NT (3.10, 3.50, 3.51, 4.0).
</para>

<para>
<indexterm significance="normal"><primary>replication</primary><secondary>SAM</secondary></indexterm>
<indexterm significance="normal"><primary>%SystemRoot%\System32\config</primary></indexterm>
<indexterm significance="normal"><primary>C:\WinNT\System32\config</primary></indexterm>
<indexterm significance="normal"><primary>BDC</primary></indexterm>
<indexterm significance="normal"><primary>SAM</primary></indexterm>
The account information (user and machine) on domain controllers is stored in two files,
one containing the security information and the other the SAM. These are stored in files
by the same name in the <filename moreinfo="none">%SystemRoot%\System32\config</filename> directory. 
This normally translates to the path <filename moreinfo="none">C:\WinNT\System32\config</filename>. These
are the files that are involved in replication of the SAM database where BDCs are present
on the network.
</para>

<para>
There are two situations in which it is desirable to install BDCs:
</para>

<itemizedlist>
	<listitem><para>
	<indexterm significance="normal"><primary>PDC</primary></indexterm>
	<indexterm significance="normal"><primary>BDC</primary></indexterm>
	On the local network that the PDC is on, if there are many
	workstations and/or where the PDC is generally very busy. In this case the BDCs
	will pick up network logon requests and help to add robustness to network services.
	</para></listitem>

	<listitem><para>
	<indexterm significance="normal"><primary>network</primary><secondary>wide-area</secondary></indexterm>
	At each remote site, to reduce wide-area network traffic and to add stability to
	remote network operations. The design of the network, and the strategic placement of
	BDCs, together with an implementation that localizes as much of network to client
	interchange as possible, will help to minimize wide-area network bandwidth needs
	(and thus costs).
	</para></listitem>
</itemizedlist>

<para>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>SAM</primary></indexterm>
<indexterm significance="normal"><primary>user account database</primary></indexterm>
<indexterm significance="normal"><primary>BDC</primary></indexterm>
<indexterm significance="normal"><primary>trigger</primary></indexterm>
The interoperation of a PDC and its BDCs in a true Windows NT4 environment is worth
mentioning here. The PDC contains the master copy of the SAM. In the event that an
administrator makes a change to the user account database while physically present
on the local network that has the PDC, the change will likely be made directly to
the PDC instance of the master copy of the SAM. In the event that this update may
be performed in a branch office, the change will likely be stored in a delta file
on the local BDC. The BDC will then send a trigger to the PDC to commence the process
of SAM synchronization. The PDC will then request the delta from the BDC and apply
it to the master SAM. The PDC will then contact all the BDCs in the domain and
trigger them to obtain the update and then apply that to their own copy of the SAM.
</para>

<para>
<indexterm significance="normal"><primary>SAM</primary><secondary>replication</secondary></indexterm>
<indexterm significance="normal"><primary>SAM</primary><secondary>delta file</secondary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>BDC</primary></indexterm>
Samba-3 cannot participate in true SAM replication and is therefore not able to
employ precisely the same protocols used by MS Windows NT4. A Samba-3 BDC will
not create SAM update delta files. It will not interoperate with a PDC (NT4 or Samba)
to synchronize the SAM from delta files that are held by BDCs.
</para>

<para>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>BDC</primary></indexterm>
Samba-3 cannot function as a BDC to an MS Windows NT4 PDC, and Samba-3 cannot
function correctly as a PDC to an MS Windows NT4 BDC. Both Samba-3 and MS Windows
NT4 can function as a BDC to its own type of PDC.
</para>

<para>
<indexterm significance="normal"><primary>SAM</primary></indexterm>
<indexterm significance="normal"><primary>BDC</primary></indexterm>
<indexterm significance="normal"><primary>domain security</primary></indexterm>
The BDC is said to hold a <emphasis>read-only</emphasis> of the SAM from which
it is able to process network logon requests and authenticate users. The BDC can
continue to provide this service, particularly while, for example, the wide-area
network link to the PDC is down. A BDC plays a very important role in both the
maintenance of domain security as well as in network integrity.
</para>

<para>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>promoted</primary></indexterm>
<indexterm significance="normal"><primary>demoted</primary></indexterm>
<indexterm significance="normal"><primary>reconfiguration</primary></indexterm>
In the event that the NT4 PDC should need to be taken out of service, or if it dies, one of the NT4 BDCs can
be promoted to a PDC. If this happens while the original NT4 PDC is online, it is automatically demoted to an
NT4 BDC. This is an important aspect of domain controller management. The tool that is used to effect a
promotion or a demotion is the Server Manager for Domains. It should be noted that Samba-3 BDCs cannot be
promoted in this manner because reconfiguration of Samba requires changes to the <filename moreinfo="none">smb.conf</filename> file. It is easy
enough to manuall change the <filename moreinfo="none">smb.conf</filename> file and then restart relevant Samba network services.
</para>

<sect3>
<title>Example PDC Configuration</title>

<para>
<indexterm significance="normal"><primary>domain logon</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
Beginning with Version 2.2, Samba officially supports domain logons for all current Windows clients, including
Windows NT4, 2003, and XP Professional. For Samba to be enabled as a PDC, some parameters in the
<parameter>[global]</parameter> section of the <filename moreinfo="none">smb.conf</filename> have to be set.  Refer to <link linkend="minimalPDC">the Minimal smb.conf for a PDC in Use with a BDC <?latex --- ?> LDAP Server on PDC
section</link> for an example of the minimum required settings.
</para>

<example id="minimalPDC">
<title>Minimal smb.conf for a PDC in Use with a BDC <?latex --- ?> LDAP Server on PDC</title>
<simplelist>
<member><indexterm>workgroup</indexterm><parameter>workgroup = MIDEARTH</parameter></member>
<member><indexterm>passdb backend</indexterm><parameter>passdb backend = ldapsam://localhost:389</parameter></member>
<member><indexterm>domain master</indexterm><parameter>domain master = yes</parameter></member>
<member><indexterm>domain logons</indexterm><parameter>domain logons = yes</parameter></member>
<member><indexterm>ldap suffix</indexterm><parameter>ldap suffix = dc=quenya,dc=org</parameter></member>
<member><indexterm>ldap user suffix</indexterm><parameter>ldap user suffix = ou=Users</parameter></member>
<member><indexterm>ldap group suffix</indexterm><parameter>ldap group suffix = ou=Groups</parameter></member>
<member><indexterm>ldap machine suffix</indexterm><parameter>ldap machine suffix = ou=Computers</parameter></member>
<member><indexterm>ldap idmap suffix</indexterm><parameter>ldap idmap suffix = ou=Idmap</parameter></member>
<member><indexterm>ldap admin dn</indexterm><parameter>ldap admin dn = cn=sambadmin,dc=quenya,dc=org</parameter></member>
</simplelist>
</example>

<para>
<indexterm significance="normal"><primary>profile path</primary></indexterm>
<indexterm significance="normal"><primary>home drive</primary></indexterm>
Several other things like a <parameter>[homes]</parameter> and a <parameter>[netlogon]</parameter> share
also need to be set along with settings for the profile path, the user's home drive, and so on. This is not
covered in this chapter; for more information please refer to <link linkend="samba-pdc">Domain Control</link>.
Refer to <link linkend="samba-pdc">the Domain Control chapter</link> for specific recommendations for PDC
configuration. Alternately, fully documented working example network configurations using OpenLDAP and Samba
as available in the <ulink url="http://www.samba.org/samba/docs/Samba3-ByExample">book</ulink> <quote>Samba-3
by Example</quote> that may be obtained from local and on-line book stores.
</para>

</sect3>
</sect2>

<sect2>
<title>LDAP Configuration Notes</title>

<para>
<indexterm significance="normal"><primary>LDAP</primary><secondary>master</secondary></indexterm>
<indexterm significance="normal"><primary>LDAP</primary><secondary>slave</secondary></indexterm>
<indexterm significance="normal"><primary>BDC</primary></indexterm>
When configuring a master and a slave LDAP server, it is advisable to use the master LDAP server
for the PDC and slave LDAP servers for the BDCs. It is not essential to use slave LDAP servers; however,
many administrators will want to do so in order to provide redundant services. Of course, one or more BDCs
may use any slave LDAP server. Then again, it is entirely possible to use a single LDAP server for the
entire network.
</para>

<para>
<indexterm significance="normal"><primary>LDAP</primary><secondary>master</secondary></indexterm>
<indexterm significance="normal"><primary>LDAP</primary><secondary>server</secondary></indexterm>
<indexterm significance="normal"><primary>CN</primary></indexterm>
<indexterm significance="normal"><primary>DN</primary></indexterm>
<indexterm significance="normal"><primary>RFC2830</primary></indexterm>
When configuring a master LDAP server that will have slave LDAP servers, do not forget to configure this in
the <filename moreinfo="none">/etc/openldap/slapd.conf</filename> file. It must be noted that the DN of a server certificate
must use the CN attribute to name the server, and the CN must carry the servers' fully qualified domain name.
Additional alias names and wildcards may be present in the subjectAltName certificate extension. More details
on server certificate names are in RFC2830.
</para>

<para>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
<indexterm significance="normal"><primary>BDC</primary></indexterm>
<indexterm significance="normal"><primary>OpenLDAP</primary></indexterm>
<indexterm significance="normal"><primary>transport layer security</primary><see>TLS</see></indexterm>
<indexterm significance="normal"><primary>/etc/ssl/certs/slapd.pem</primary></indexterm>
<indexterm significance="normal"><primary>slapd.pem</primary></indexterm>
<indexterm significance="normal"><primary>Red Hat Linux</primary></indexterm>
It does not really fit within the scope of this document, but a working LDAP installation is basic to
LDAP-enabled Samba operation. When using an OpenLDAP server with Transport Layer Security (TLS), the machine
name in <filename moreinfo="none">/etc/ssl/certs/slapd.pem</filename> must be the same as in
<filename moreinfo="none">/etc/openldap/sldap.conf</filename>. The Red Hat Linux startup script creates the
<filename moreinfo="none">slapd.pem</filename> file with hostname <quote>localhost.localdomain.</quote> It is impossible to
access this LDAP server from a slave LDAP server (i.e., a Samba BDC) unless the certificate is re-created with
a correct hostname.
</para>

<para>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>OpenLDAP</primary></indexterm>
<indexterm significance="normal"><primary>machine account</primary></indexterm>
<indexterm significance="normal"><primary>credentials</primary></indexterm>
<indexterm significance="normal"><primary>replication</primary></indexterm>
<indexterm significance="normal"><primary>duplicate</primary></indexterm>
Do not install a Samba PDC so that is uses an LDAP slave server. Joining client machines to the domain
will fail in this configuration because the change to the machine account in the LDAP tree must take place on
the master LDAP server. This is not replicated rapidly enough to the slave server that the PDC queries. It
therefore gives an error message on the client machine about not being able to set up account credentials. The
machine account is created on the LDAP server, but the password fields will be empty.  Unfortunately, some
sites are unable to avoid such configurations, and these sites should review the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LDAPREPLICATIONSLEEP" xlink:href="smb.conf.5.html#LDAPREPLICATIONSLEEP">ldap replication sleep</link> parameter, intended to slow down Samba sufficiently for the replication to catch up.
This is a kludge, and one that the administrator must manually duplicate in any scripts (such as the
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="ADDMACHINESCRIPT" xlink:href="smb.conf.5.html#ADDMACHINESCRIPT">add machine script</link>) that they use.
</para>

<para>
Possible PDC/BDC plus LDAP configurations include:
</para>

<itemizedlist>
	<listitem><para>
	PDC+BDC -&gt; One Central LDAP Server.
	</para></listitem>
	<listitem><para>
	PDC -&gt; LDAP master server, BDC -&gt; LDAP slave server.
	</para></listitem>
	<listitem><para>
	PDC -&gt; LDAP master, with secondary slave LDAP server.
	</para><para>
	BDC -&gt; LDAP master, with secondary slave LDAP server.
	</para></listitem>
	<listitem><para>
	PDC -&gt; LDAP master, with secondary slave LDAP server.
	</para><para>
	BDC -&gt; LDAP slave server, with secondary master LDAP server.
	</para></listitem>
</itemizedlist>

<para>
In order to have a fallback configuration (secondary) LDAP server, you would specify
the secondary LDAP server in the <filename moreinfo="none">smb.conf</filename> file as shown in <link linkend="mulitldapcfg">the Multiple LDAP
Servers in <filename moreinfo="none">smb.conf</filename> example</link>.
</para>

<example id="mulitldapcfg">
<title>Multiple LDAP Servers in <filename moreinfo="none">smb.conf</filename></title>
<simplelist>
<member><indexterm>passdb backend</indexterm><parameter>passdb backend = ldapsam:"ldap://master.quenya.org ldap://slave.quenya.org"</parameter></member>
</simplelist>
</example>

</sect2>

<sect2>
<title>Active Directory Domain Control</title>

<para>
<indexterm significance="normal"><primary>MS Windows 2000</primary></indexterm>
<indexterm significance="normal"><primary>Active Directory</primary></indexterm>
<indexterm significance="normal"><primary>directory</primary></indexterm>
<indexterm significance="normal"><primary>replicated</primary></indexterm>
<indexterm significance="normal"><primary>BDC</primary></indexterm>
<indexterm significance="normal"><primary>domain controller</primary></indexterm>
As of the release of MS Windows 2000 and Active Directory, this information is now stored
in a directory that can be replicated and for which partial or full administrative control
can be delegated. Samba-3 is not able to be a domain controller within an Active Directory
tree, and it cannot be an Active Directory server. This means that Samba-3 also cannot
act as a BDC to an Active Directory domain controller.
</para>

</sect2>

<sect2>
<title>What Qualifies a Domain Controller on the Network?</title>

<para>
<indexterm significance="normal"><primary>DMB</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
<indexterm significance="normal"><primary>NetBIOS</primary></indexterm>
Every machine that is a domain controller for the domain MIDEARTH has to register the NetBIOS
group name MIDEARTH&lt;1C&gt; with the WINS server and/or by broadcast on the local network.
The PDC also registers the unique NetBIOS name MIDEARTH&lt;1B&gt; with the WINS server.
The name type &lt;1B&gt; name is normally reserved for the Domain Master Browser (DMB), a role
that has nothing to do with anything related to authentication, but the Microsoft domain
implementation requires the DMB to be on the same machine as the PDC.
</para>

<para>
<indexterm significance="normal"><primary>broadcast</primary></indexterm>
<indexterm significance="normal"><primary>name registration</primary></indexterm>
<indexterm significance="normal"><primary>SMB/CIFS</primary></indexterm>
Where a WINS server is not used, broadcast name registrations alone must suffice. Refer to
<link linkend="NetworkBrowsing">Network Browsing</link>,<link linkend="netdiscuss">Discussion</link>
for more information regarding TCP/IP network protocols and how SMB/CIFS names are handled.
</para>

</sect2>

<sect2>
<title>How Does a Workstation find its Domain Controller?</title>

<para>
<indexterm significance="normal"><primary>locate domain controller</primary></indexterm>
<indexterm significance="normal"><primary>NetBIOS</primary></indexterm>
There are two different mechanisms to locate a domain controller: one method is used when
NetBIOS over TCP/IP is enabled and the other when it has been disabled in the TCP/IP
network configuration.
</para>

<para>
<indexterm significance="normal"><primary>DNS</primary></indexterm>
<indexterm significance="normal"><primary>broadcast messaging</primary></indexterm>
Where NetBIOS over TCP/IP is disabled, all name resolution involves the use of DNS, broadcast
messaging over UDP, as well as Active Directory communication technologies. In this type of
environment all machines require appropriate DNS entries. More information may be found in
<link linkend="adsdnstech">DNS and Active Directory</link>.
</para>

<sect3>
<title>NetBIOS Over TCP/IP Enabled</title>
<para>
<indexterm significance="normal"><primary>Windows NT4/200x/XP</primary></indexterm>
<indexterm significance="normal"><primary>domain controller</primary></indexterm>
<indexterm significance="normal"><primary>logon requests</primary></indexterm>
<indexterm significance="normal"><primary>credentials validation</primary></indexterm>
An MS Windows NT4/200x/XP Professional workstation in the domain MIDEARTH that wants a
local user to be authenticated has to find the domain controller for MIDEARTH. It does this
by doing a NetBIOS name query for the group name MIDEARTH&lt;1C&gt;. It assumes that each
of the machines it gets back from the queries is a domain controller and can answer logon
requests. To not open security holes, both the workstation and the selected domain controller
authenticate each other. After that the workstation sends the user's credentials (name and
password) to the local domain controller for validation.
</para>

</sect3>

<sect3>
<title>NetBIOS Over TCP/IP Disabled</title>

<para>
<indexterm significance="normal"><primary>realm</primary></indexterm>
<indexterm significance="normal"><primary>logon authentication</primary></indexterm>
<indexterm significance="normal"><primary>DNS</primary></indexterm>
<indexterm significance="normal"><primary>_ldap._tcp.pdc._msdcs.quenya.org</primary></indexterm>
An MS Windows NT4/200x/XP Professional workstation in the realm <constant>quenya.org</constant>
that has a need to affect user logon authentication will locate the domain controller by 
re-querying DNS servers for the <constant>_ldap._tcp.pdc._msdcs.quenya.org</constant> record.
More information regarding this subject may be found in <link linkend="adsdnstech">DNS and Active Directory</link>.
</para>

</sect3>
</sect2>
</sect1>

<sect1>
<title>Backup Domain Controller Configuration</title>

<para>
<indexterm significance="normal"><primary>BDC</primary></indexterm>
The creation of a BDC requires some steps to prepare the Samba server before
<application moreinfo="none">smbd</application> is executed for the first time. These steps are as follows:
</para>

<itemizedlist>
	<listitem><para>
	<indexterm significance="normal"><primary>SID</primary></indexterm>
	<indexterm significance="normal"><primary>PDC</primary></indexterm>
	<indexterm significance="normal"><primary>BDC</primary></indexterm>
	<indexterm significance="normal"><primary>private/secrets.tdb</primary></indexterm>
	<indexterm significance="normal"><primary>private/MACHINE.SID</primary></indexterm>
	<indexterm significance="normal"><primary>domain SID</primary></indexterm>
	The domain SID has to be the same on the PDC and the BDC. In Samba versions pre-2.2.5, the domain SID was
	stored in the file <filename moreinfo="none">private/MACHINE.SID</filename>.  For all versions of Samba released since 2.2.5
	the domain SID is stored in the file <filename moreinfo="none">private/secrets.tdb</filename>. This file is unique to each
	server and cannot be copied from a PDC to a BDC; the BDC will generate a new SID at startup. It will overwrite
	the PDC domain SID with the newly created BDC SID.  There is a procedure that will allow the BDC to aquire the
	domain SID. This is described here.
	</para>

	<para>
	<indexterm significance="normal"><primary>domain SID</primary></indexterm>
	<indexterm significance="normal"><primary>PDC</primary></indexterm>
	<indexterm significance="normal"><primary>BDC</primary></indexterm>
	<indexterm significance="normal"><primary>secrets.tdb</primary></indexterm>
	<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>getsid</tertiary></indexterm>
	To retrieve the domain SID from the PDC or an existing BDC and store it in the
	<filename moreinfo="none">secrets.tdb</filename>, execute:
	</para>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">net rpc getsid</userinput>
</screen>
	</listitem>

	<listitem><para>
	<indexterm significance="normal"><primary>secrets.tdb</primary></indexterm>
	<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
	<indexterm significance="normal"><primary>LDAP administration password</primary></indexterm>
	Specification of the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LDAPADMINDN" xlink:href="smb.conf.5.html#LDAPADMINDN">ldap admin dn</link> is obligatory.
	This also requires the LDAP administration password to be set in the <filename moreinfo="none">secrets.tdb</filename>
	using the <literal>smbpasswd -w <replaceable>mysecret</replaceable></literal>.
	</para></listitem>

	<listitem><para>
	The <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LDAPSUFFIX" xlink:href="smb.conf.5.html#LDAPSUFFIX">ldap suffix</link> parameter and the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LDAPIDMAPSUFFIX" xlink:href="smb.conf.5.html#LDAPIDMAPSUFFIX">ldap idmap suffix</link>
	parameter must be specified in the <filename moreinfo="none">smb.conf</filename> file.
	</para></listitem>

	<listitem><para>
	<indexterm significance="normal"><primary>replication</primary><secondary>SAM</secondary></indexterm>
	<indexterm significance="normal"><primary>user database</primary></indexterm>
	<indexterm significance="normal"><primary>synchronized</primary></indexterm>
	<indexterm significance="normal"><primary>NIS</primary></indexterm>
	The UNIX user database has to be synchronized from the PDC to the
	BDC. This means that both the <filename moreinfo="none">/etc/passwd</filename> and
	<filename moreinfo="none">/etc/group</filename> have to be replicated from the PDC
	to the BDC. This can be done manually whenever changes are made. 
	Alternately, the PDC is set up as an NIS master server and the BDC as an NIS slave
	server. To set up the BDC as a mere NIS client would not be enough,
	as the BDC would not be able to access its user database in case of
	a PDC failure. NIS is by no means the only method to synchronize
	passwords. An LDAP solution would also work.
	</para>
	</listitem>

	<listitem><para>
	<indexterm significance="normal"><primary>password database</primary></indexterm>
	<indexterm significance="normal"><primary>replicated</primary></indexterm>
	<indexterm significance="normal"><primary>PDC</primary></indexterm>
	<indexterm significance="normal"><primary>BDC</primary></indexterm>
	<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
	<indexterm significance="normal"><primary>rsync</primary></indexterm>
	<indexterm significance="normal"><primary>ssh</primary></indexterm>
	<indexterm significance="normal"><primary>LDAP</primary></indexterm>
	The Samba password database must be replicated from the PDC to the BDC.
	Although it is possible to synchronize the <filename moreinfo="none">smbpasswd</filename>
	file with <literal>rsync</literal> and <literal>ssh</literal>, this method
	is broken and flawed, and is therefore not recommended. A better solution
	is to set up slave LDAP servers for each BDC and a master LDAP server for the PDC.
	The use of rsync is inherently flawed by the fact that the data will be replicated
	at timed intervals. There is no guarantee that the BDC will be operating at all
	times with correct and current machine and user account information. This means that
	this method runs the risk of users being inconvenienced by discontinuity of access
	to network services due to inconsistent security data. It must be born in mind that
	Windows workstations update (change) the machine trust account password at regular
	intervals <?latex --- ?> administrators are not normally aware that this is happening
	or when it takes place.
	</para>

	<para>
	<indexterm significance="normal"><primary>POSIX</primary></indexterm>
	<indexterm significance="normal"><primary>LDAP</primary></indexterm>
	<indexterm significance="normal"><primary>SambaSAMAccount</primary></indexterm>
	<indexterm significance="normal"><primary>synchronize</primary></indexterm>
	The use of LDAP for both the POSIX (UNIX user and group) accounts and for the
	SambaSAMAccount data automatically ensures that all account change information
	will be written to the shared directory. This eliminates the need for any special
	action to synchronize account information because LDAP will meet that requirement.
	</para></listitem>

	<listitem><para>
	<indexterm significance="normal"><primary>netlogon share</primary></indexterm>
	<indexterm significance="normal"><primary>replicate</primary></indexterm>
	<indexterm significance="normal"><primary>PDC</primary></indexterm>
	<indexterm significance="normal"><primary>BDC</primary></indexterm>
	<indexterm significance="normal"><primary>cron</primary></indexterm>
	<indexterm significance="normal"><primary>rsync</primary></indexterm>
	The netlogon share has to be replicated from the PDC to the BDC. This can be done manually whenever login
	scripts are changed, or it can be done automatically using a <literal>cron</literal> job that will replicate
	the directory structure in this share using a tool like <literal>rsync</literal>. The use of
	<literal>rsync</literal> for replication of the netlogon data is not critical to network security and is one
	that can be manually managed given that the administrator will make all changes to the netlogon share as part
	of a conscious move.
	</para></listitem>

</itemizedlist>

<sect2>
<title>Example Configuration</title>

<para>
Finally, the BDC has to be capable of being found by the workstations. This can be done by configuring the
Samba <filename moreinfo="none">smb.conf</filename> file <parameter>[global]</parameter> section as shown in <link linkend="minim-bdc">Minimal
Setup for Being a BDC</link>.
</para>

<example id="minim-bdc">
<title>Minimal Setup for Being a BDC</title>
<simplelist>
<member><indexterm>workgroup</indexterm><parameter>workgroup = MIDEARTH</parameter></member>
<member><indexterm>passdb backend</indexterm><parameter>passdb backend = ldapsam:ldap://slave-ldap.quenya.org</parameter></member>
<member><indexterm>domain master</indexterm><parameter>domain master = no</parameter></member>
<member><indexterm>domain logons</indexterm><parameter>domain logons = yes</parameter></member>
<member><indexterm>ldap suffix</indexterm><parameter>ldap suffix = dc=abmas,dc=biz</parameter></member>
<member><indexterm>ldap user suffix</indexterm><parameter>ldap user suffix = ou=Users</parameter></member>
<member><indexterm>ldap group suffix</indexterm><parameter>ldap group suffix = ou=Groups</parameter></member>
<member><indexterm>ldap machine suffix</indexterm><parameter>ldap machine suffix = ou=Computers</parameter></member>
<member><indexterm>ldap idmap suffix</indexterm><parameter>ldap idmap suffix = ou=Idmap</parameter></member>
<member><indexterm>ldap admin dn</indexterm><parameter>ldap admin dn = cn=sambadmin,dc=quenya,dc=org</parameter></member>
<member><indexterm>idmap backend</indexterm><parameter>idmap backend = ldap:ldap://master-ldap.quenya.org</parameter></member>
<member><indexterm>idmap uid</indexterm><parameter>idmap uid = 10000-20000</parameter></member>
<member><indexterm>idmap gid</indexterm><parameter>idmap gid = 10000-20000</parameter></member>
</simplelist>
</example>

<para>
Fully documented working example network configurations using OpenLDAP and Samba
as available in the <ulink url="http://www.samba.org/samba/docs/Samba3-ByExample">book</ulink> <quote>Samba-3
by Example</quote> that may be obtained from local and on-line book stores.
</para>

<para>
<indexterm significance="normal"><primary>BDC</primary></indexterm>
<indexterm significance="normal"><primary>NetBIOS</primary></indexterm>
<indexterm significance="normal"><primary>group</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
This configuration causes the BDC to register only the name MIDEARTH&lt;1C&gt; with the WINS server. This is
not a problem, as the name MIDEARTH&lt;1C&gt; is a NetBIOS group name that is meant to be registered by more
than one machine. The parameter <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="DOMAINMASTER" xlink:href="smb.conf.5.html#DOMAINMASTER">domain master = no</link> forces the BDC not to
register MIDEARTH&lt;1B&gt;, which is a unique NetBIOS name that is reserved for the PDC.
</para>

<para>
<indexterm significance="normal"><primary>idmap backend</primary></indexterm>
<indexterm significance="normal"><primary>winbindd</primary></indexterm>
<indexterm significance="normal"><primary>redirect</primary></indexterm>
<indexterm significance="normal"><primary>winbindd</primary></indexterm>
<indexterm significance="normal"><primary>LDAP database</primary></indexterm>
<indexterm significance="normal"><primary>UID</primary></indexterm>
<indexterm significance="normal"><primary>GID</primary></indexterm>
<indexterm significance="normal"><primary>SID</primary></indexterm>
<indexterm significance="normal"><primary>nss_ldap</primary></indexterm>
The <parameter moreinfo="none">idmap backend</parameter> will redirect the <literal>winbindd</literal> utility to use the LDAP
database to store all mappings for Windows SIDs to  UIDs and GIDs for UNIX accounts in a repository that is
shared. The BDC will however depend on local resolution of UIDs and GIDs via NSS and the
<literal>nss_ldap</literal> utility.
</para>

<note><para>
<indexterm significance="normal"><primary>Server Type</primary><secondary>Domain Member</secondary></indexterm>
<indexterm significance="normal"><primary>ID mapping</primary></indexterm>
<indexterm significance="normal"><primary>domain member server</primary></indexterm>
<indexterm significance="normal"><primary>idmap backend</primary></indexterm>
Samba-3 has introduced a new ID mapping facility. One of the features of this facility is that it
allows greater flexibility in how user and group IDs are handled in respect to NT domain user and group
SIDs. One of the new facilities provides for explicitly ensuring that UNIX/Linux UID and GID values
will be consistent on the PDC, all BDCs, and all domain member servers. The parameter that controls this
is called <parameter moreinfo="none">idmap backend</parameter>. Please refer to the man page for <filename moreinfo="none">smb.conf</filename> for more information
regarding its behavior.
</para></note>

<para>
<indexterm significance="normal"><primary>BDC</primary></indexterm>
<indexterm significance="normal"><primary>winbindd</primary></indexterm>
<indexterm significance="normal"><primary>domain member servers</primary></indexterm>
The use of the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="IDMAPBACKEND" xlink:href="smb.conf.5.html#IDMAPBACKEND">idmap backend = ldap:ldap://master.quenya.org</link>
option on a BDC only makes sense where ldapsam is used on a PDC. The purpose of an LDAP-based idmap backend is
also to allow a domain member (without its own passdb backend) to use winbindd to resolve Windows network users
and groups to common UID/GIDs. In other words, this option is generally intended for use on BDCs and on domain
member servers.
</para>

</sect2>
</sect1>

<sect1>
<title>Common Errors</title>

<para>
<indexterm significance="normal"><primary>domain control</primary></indexterm>
Domain control was a new area for Samba, but there are now many examples that we may refer to.
Updated information will be published as they become available and may be found in later Samba releases or
from the Samba Web <ulink url="http://samba.org">site</ulink>; refer in particular to the
<filename moreinfo="none">WHATSNEW.txt</filename> in the Samba release tarball. The book, <quote>Samba-3 by Example</quote>
documents well tested and proven configuration examples. You can obtain a copy of this
<ulink url="http://www.samba.org/samba/docs/Samba3-ByExample.pdf">book</ulink> for the Samba web site.
</para>

<sect2>
<title>Machine Accounts Keep Expiring</title>

<para>
<indexterm significance="normal"><primary>Machine Trust Accounts</primary></indexterm>
<indexterm significance="normal"><primary>passdb</primary></indexterm>
<indexterm significance="normal"><primary>SAM</primary></indexterm>
<indexterm significance="normal"><primary>Local Machine Trust Account</primary></indexterm>
This problem will occur when the passdb (SAM) files are copied  from a central
server but the local BDC is acting as a PDC. This results in the application of
Local Machine Trust Account password updates to the local SAM. Such updates 
are not copied back to the central server. The newer machine account password is then
overwritten when the SAM is recopied from the PDC. The result is that the domain member machine
on startup will find that its passwords do not match the one now in the database, and
since the startup security check will now fail, this machine will not allow logon attempts
to proceed and the account expiry error will be reported.
</para>

<para>
The solution is to use a more robust passdb backend, such as the ldapsam backend, setting up
a slave LDAP server for each BDC and a master LDAP server for the PDC.
</para>

</sect2>

<sect2>
<title>Can Samba Be a Backup Domain Controller to an NT4 PDC?</title>

<para>
<indexterm significance="normal"><primary>replication</primary><secondary>SAM</secondary></indexterm>
<indexterm significance="normal"><primary>SAM</primary></indexterm>
No. The native NT4 SAM replication protocols have not yet been fully implemented.
</para>

<para>
<indexterm significance="normal"><primary>BDC</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>logon requests</primary></indexterm>
Can I get the benefits of a BDC with Samba?  Yes, but only to a Samba PDC.The
main reason for implementing a BDC is availability. If the PDC is a Samba
machine, a second Samba machine can be set up to service logon requests whenever
the PDC is down.
</para>

</sect2>

<sect2>
<title>How Do I Replicate the smbpasswd File?</title>

<para>
<indexterm significance="normal"><primary>replication</primary><secondary>SAM</secondary></indexterm>
<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
<indexterm significance="normal"><primary>SAM</primary></indexterm>
Replication of the smbpasswd file is sensitive. It has to be done whenever changes
to the SAM are made. Every user's password change is done in the smbpasswd file and
has to be replicated to the BDC. So replicating the smbpasswd file very often is necessary.
</para>

<para>
<indexterm significance="normal"><primary>plaintext password</primary></indexterm>
<indexterm significance="normal"><primary>ssh</primary></indexterm>
<indexterm significance="normal"><primary>rsync</primary></indexterm>
As the smbpasswd file contains plaintext password equivalents, it must not be
sent unencrypted over the wire. The best way to set up smbpasswd replication from
the PDC to the BDC is to use the utility rsync. rsync can use ssh as a transport.
<literal>ssh</literal> itself can be set up to accept <emphasis>only</emphasis>
<literal>rsync</literal> transfer without requiring the user to type a password.
</para>

<para>
<indexterm significance="normal"><primary>machine trust accounts</primary></indexterm>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
As said a few times before, use of this method is broken and flawed. Machine trust 
accounts will go out of sync, resulting in a broken domain. This method is
<emphasis>not</emphasis> recommended. Try using LDAP instead.
</para>

</sect2>

<sect2>
<title>Can I Do This All with LDAP?</title>

<para>
<indexterm significance="normal"><primary>pdb_ldap</primary></indexterm>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
The simple answer is yes. Samba's pdb_ldap code supports binding to a replica
LDAP server and will also follow referrals and rebind to the master if it ever
needs to make a modification to the database. (Normally BDCs are read-only, so
this will not occur often).
</para>

</sect2>
</sect1>
</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="domain-member">

<chapterinfo>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
	
<author>
	<firstname>Jeremy</firstname><surname>Allison</surname>
	<affiliation>
		<orgname>Samba Team</orgname>
		<address format="linespecific"><email>jra@samba.org</email></address>
	</affiliation>
</author>
	<author>
<firstname>Gerald</firstname><surname>Carter</surname><othername>(Jerry)</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jerry@samba.org</email></address>
</affiliation></author>
	
<author>
	<firstname>Andrew</firstname><surname>Tridgell</surname>
	<affiliation>
		<orgname>Samba Team</orgname>
		<address format="linespecific"><email>tridge@samba.org</email></address>
	</affiliation>
</author>
	<author>
<firstname>Jelmer</firstname><othername>R.</othername><surname>Vernooij</surname><othername>R.</othername>
<affiliation>
	<orgname>The Samba Team</orgname>
	<address format="linespecific"><email>jelmer@samba.org</email></address>
</affiliation></author>
	<author>
<firstname>Guenther</firstname><surname>Deschner</surname>
<affiliation>
	<orgname>SuSE</orgname>
	<address format="linespecific"><email>gd@suse.de</email></address>
</affiliation><contrib>LDAP updates</contrib></author>
</chapterinfo>

<title>Domain Membership</title>

<para>
<indexterm significance="normal"><primary>domain member</primary></indexterm>
<indexterm significance="normal"><primary>machine trust account</primary></indexterm>
<indexterm significance="normal"><primary>domain security</primary></indexterm>
Domain membership is a subject of vital concern. Samba must be able to
participate as a member server in a Microsoft domain security context, and
Samba must be capable of providing domain machine member trust accounts;
otherwise it would not be able to offer a viable option for many users.
</para>

<para>
<indexterm significance="normal"><primary>domain membership</primary></indexterm>
<indexterm significance="normal"><primary>misinformation</primary></indexterm>
This chapter covers background information pertaining to domain membership,
the Samba configuration for it, and MS Windows client procedures for joining a
domain. Why is this necessary? Because both are areas in which there exists
within the current MS Windows networking world, and particularly in the
UNIX/Linux networking and administration world, a considerable level of
misinformation, incorrect understanding, and lack of knowledge. Hopefully
this chapter will fill the voids.
</para>

<sect1>
<title>Features and Benefits</title>

<para>
<indexterm significance="normal"><primary>domain security</primary></indexterm>
<indexterm significance="normal"><primary>single sign-on</primary></indexterm>
<indexterm significance="normal"><primary>SSO</primary></indexterm>
MS Windows workstations and servers that want to participate in domain security need to
be made domain members. Participating in domain security is often called 
<emphasis>single sign-on</emphasis>, or <acronym>SSO</acronym> for short. This
chapter describes the process that must be followed to make a workstation
(or another server <?latex --- ?> be it an <application moreinfo="none">MS Windows NT4/200x</application>
server) or a Samba server a member of an MS Windows domain security context.
</para>

<para>
<indexterm significance="normal"><primary>native member</primary></indexterm>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
<indexterm significance="normal"><primary>domain control</primary></indexterm>
<indexterm significance="normal"><primary>Server Type</primary><secondary>Domain Member</secondary></indexterm>
Samba-3 can join an MS Windows NT4-style domain as a native member server, an 
MS Windows Active Directory domain as a native member server, or a Samba domain
control network. Domain membership has many advantages:
</para>

<itemizedlist>
	<listitem><para>
	<indexterm significance="normal"><primary>SAM</primary></indexterm>
	MS Windows workstation users get the benefit of SSO.
	</para></listitem>

	<listitem><para>
	<indexterm significance="normal"><primary>access rights</primary></indexterm>
	<indexterm significance="normal"><primary>file ownership</primary></indexterm>
	<indexterm significance="normal"><primary>access controls</primary></indexterm>
	<indexterm significance="normal"><primary>SAM</primary></indexterm>
	Domain user access rights and file ownership/access controls can be set
	from the single Domain Security Account Manager (SAM) database 
	(works with domain member servers as well as with MS Windows workstations
	that are domain members).
	</para></listitem>

	<listitem><para>
	<indexterm significance="normal"><primary>domain members</primary></indexterm>
	<indexterm significance="normal"><primary>network logon</primary></indexterm>
	Only <application moreinfo="none">MS Windows NT4/200x/XP Professional</application>
	workstations that are domain members can use network logon facilities.
	</para></listitem>

	<listitem><para>
	<indexterm significance="normal"><primary>domain member</primary></indexterm>
	<indexterm significance="normal"><primary>policy files</primary></indexterm>
	<indexterm significance="normal"><primary>NTConfig.POL</primary></indexterm>
	<indexterm significance="normal"><primary>desktop profiles</primary></indexterm>
	Domain member workstations can be better controlled through the use of
	policy files (<filename moreinfo="none">NTConfig.POL</filename>) and desktop profiles.
	</para></listitem>

	<listitem><para>
	<indexterm significance="normal"><primary>logon script</primary></indexterm>
	<indexterm significance="normal"><primary>transparent access</primary></indexterm>
	<indexterm significance="normal"><primary>application servers</primary></indexterm>
	Through the use of logon scripts, users can be given transparent access to network
	applications that run off application servers.
	</para></listitem>

	<listitem><para>
	<indexterm significance="normal"><primary>user access management</primary></indexterm>
	<indexterm significance="normal"><primary>SAM</primary></indexterm>
	<indexterm significance="normal"><primary>LDAP</primary></indexterm>
	<indexterm significance="normal"><primary>ADS</primary></indexterm>
	Network administrators gain better application and user access management
	abilities because there is no need to maintain user accounts on any network
	client or server other than the central domain database 
	(either NT4/Samba SAM-style domain, NT4 domain that is backend-ed with an
	LDAP directory, or via an Active Directory infrastructure).
	</para></listitem>
</itemizedlist>

</sect1>

<sect1 id="machine-trust-accounts">
<title>MS Windows Workstation/Server Machine Trust Accounts</title>

<para>
<indexterm significance="normal"><primary>Machine Trust Accounts</primary></indexterm>
<indexterm significance="normal"><primary>authenticate</primary></indexterm>
<indexterm significance="normal"><primary>domain controller</primary></indexterm>
<indexterm significance="normal"><primary>rogue user</primary></indexterm>
A Machine Trust Account is an account that is used to authenticate a client machine (rather than a user) to
the domain controller server. In Windows terminology, this is known as a <quote>computer account.</quote> The
purpose of the machine trust account is to prevent a rogue user and domain controller from colluding to gain
access to a domain member workstation.
</para>

<para>
<indexterm significance="normal"><primary>machine trust account</primary><secondary>password</secondary></indexterm>
<indexterm significance="normal"><primary>shared secret</primary></indexterm>
<indexterm significance="normal"><primary>unauthorized</primary></indexterm>
<indexterm significance="normal"><primary>Windows NT/200x/XP Professional</primary></indexterm>
<indexterm significance="normal"><primary>Windows 9x/Me/XP Home</primary></indexterm>
The password of a Machine Trust Account acts as the shared secret for secure communication with the domain
controller. This is a security feature to prevent an unauthorized machine with the same NetBIOS name from
joining the domain, participating in domain security operations, and gaining access to domain user/group
accounts. Windows NT/200x/XP Professional clients use machine trust accounts, but Windows 9x/Me/XP Home
clients do not. Hence, a Windows 9x/Me/XP Home client is never a true member of a domain because it does not
possess a Machine Trust Account, and, thus, has no shared secret with the domain controller.
</para>

<para>
<indexterm significance="normal"><primary>Windows Registry</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
<indexterm significance="normal"><primary>Machine Trust Account</primary></indexterm>
A Windows NT4 PDC stores each Machine Trust Account in the Windows Registry.
The introduction of MS Windows 2000 saw the introduction of Active Directory,
the new repository for Machine Trust Accounts. A Samba PDC, however, stores
each Machine Trust Account in two parts,
as follows:

<itemizedlist>
	<listitem><para>
	<indexterm significance="normal"><primary>domain security account</primary></indexterm>
	<indexterm significance="normal"><primary>account information</primary></indexterm>
	<indexterm significance="normal"><primary>backend database</primary></indexterm>
	A domain security account (stored in the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PASSDBBACKEND" xlink:href="smb.conf.5.html#PASSDBBACKEND">passdb backend</link>) that has been configured in
	the <filename moreinfo="none">smb.conf</filename> file. The precise nature of the account information that is stored depends on the type of
	backend database that has been chosen.
	</para>

	<para>
	<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
	<indexterm significance="normal"><primary>UNIX login ID</primary></indexterm>
	<indexterm significance="normal"><primary>UID</primary></indexterm>
	<indexterm significance="normal"><primary>LanMan</primary></indexterm>
	<indexterm significance="normal"><primary>NT-encrypted password</primary></indexterm>
	<indexterm significance="normal"><primary>UNIX user identifier</primary><see>UID</see></indexterm>
	The older format of this data is the <filename moreinfo="none">smbpasswd</filename> database
	that contains the UNIX login ID, the UNIX user identifier (UID), and the
	LanMan and NT-encrypted passwords. There is also some other information in
	this file that we do not need to concern ourselves with here.
	</para>

	<para>
	<indexterm significance="normal"><primary>database</primary></indexterm>
	<indexterm significance="normal"><primary>ldapsam</primary></indexterm>
	<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
	<indexterm significance="normal"><primary>account controls</primary></indexterm>
	The two newer database types are called ldapsam and tdbsam. Both store considerably more data than the older
	<filename moreinfo="none">smbpasswd</filename> file did. The extra information enables new user account controls to be
	implemented.
	</para></listitem>

	<listitem><para>
	<indexterm significance="normal"><primary>UNIX account</primary></indexterm>
	<indexterm significance="normal"><primary>/etc/passwd</primary></indexterm>
	A corresponding UNIX account, typically stored in <filename moreinfo="none">/etc/passwd</filename>. Work is in progress to
	allow a simplified mode of operation that does not require UNIX user accounts, but this has not been a feature
	of the early releases of Samba-3, and is not currently planned for release either.
	</para></listitem>
</itemizedlist>
</para>

<?latex \newpage ?>
<para>
<indexterm significance="normal"><primary>Machine Trust Accounts</primary><secondary>creating</secondary></indexterm>
There are three ways to create Machine Trust Accounts:
</para>

<itemizedlist>
	<listitem><para>
	<indexterm significance="normal"><primary>manual UNIX account creation</primary></indexterm>
	Manual creation from the UNIX/Linux command line. Here, both the Samba and
	corresponding UNIX account are created by hand.
	</para></listitem>

	<listitem><para>
	<indexterm significance="normal"><primary>Server Manager</primary></indexterm>
	<indexterm significance="normal"><primary>Nexus toolkit</primary></indexterm>
	Using the MS Windows NT4 Server Manager, either from an NT4 domain member
	server or using the Nexus toolkit available from the Microsoft Web site.
	This tool can be run from any MS Windows machine as long as the user is
	logged on as the administrator account.
	</para></listitem>
	
	<listitem><para>
	<indexterm significance="normal"><primary>Machine Trust Account</primary></indexterm>
	<indexterm significance="normal"><primary>joined client</primary></indexterm>
	<quote>On-the-fly</quote> creation. The Samba Machine Trust Account is automatically
	created by Samba at the time the client is joined to the domain.
	(For security, this is the recommended method.) The corresponding UNIX
	account may be created automatically or manually. 
	</para></listitem>
</itemizedlist>

<para>
<indexterm significance="normal"><primary>enforcing</primary></indexterm>
<indexterm significance="normal"><primary>machine trust account</primary><secondary>creation</secondary></indexterm>
Neither MS Windows NT4/200x/XP Professional, nor Samba, provide any method for enforcing the method of machine
trust account creation. This is a matter of the administrator's choice.
</para>

<sect2>
<title>Manual Creation of Machine Trust Accounts</title>

<para>
<indexterm significance="normal"><primary>/etc/passwd</primary></indexterm>
<indexterm significance="normal"><primary/></indexterm>
<indexterm significance="normal"><primary>useradd</primary></indexterm>
<indexterm significance="normal"><primary>vipw</primary></indexterm>
The first step in manually creating a Machine Trust Account is to manually
create the corresponding UNIX account in <filename moreinfo="none">/etc/passwd</filename>. 
This can be done using <literal>vipw</literal> or another <quote>adduser</quote> command
that is normally used to create new UNIX accounts. The following is an example for
a Linux-based Samba server:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">/usr/sbin/useradd -g machines -d /var/lib/nobody \
   -c <replaceable>"machine nickname"</replaceable> \
   -s /bin/false <replaceable>machine_name</replaceable>$ </userinput>

<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">passwd -l <replaceable>machine_name</replaceable>$</userinput>
</screen>
</para>

<para>
<indexterm significance="normal"><primary>primary group</primary></indexterm>
<indexterm significance="normal"><primary>GID</primary></indexterm>
<indexterm significance="normal"><primary>machine accounts</primary></indexterm>
In the example above there is an existing system group <quote>machines</quote> which is used
as the primary group for all machine accounts. In the following examples the <quote>machines</quote> group
numeric GID is 100.
</para>

<para>
<indexterm significance="normal"><primary>chpass</primary></indexterm>
<indexterm significance="normal"><primary>BSD</primary></indexterm>
On *BSD systems, this can be done using the <literal>chpass</literal> utility:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">chpass -a \
'<replaceable>machine_name</replaceable>$:*:101:100::0:0:Windows <replaceable>machine_name</replaceable>:/dev/null:/sbin/nologin'</userinput>
</screen>
</para>

<para>
<indexterm significance="normal"><primary>/etc/passwd</primary></indexterm>
<indexterm significance="normal"><primary>$</primary></indexterm>
<indexterm significance="normal"><primary>null shell</primary></indexterm>
<indexterm significance="normal"><primary>home directory</primary></indexterm>
The <filename moreinfo="none">/etc/passwd</filename> entry will list the machine name 
with a <quote>$</quote> appended, and will not have a password, will have a null shell and no 
home directory. For example, a machine named <quote>doppy</quote> would have an 
<filename moreinfo="none">/etc/passwd</filename> entry like this:
<programlisting format="linespecific">
doppy$:x:505:100:<replaceable>machine_nickname</replaceable>:/dev/null:/bin/false
</programlisting>
</para>

<para>
<indexterm significance="normal"><primary>machine_nickname</primary></indexterm>
<indexterm significance="normal"><primary>machine_name</primary></indexterm>
<indexterm significance="normal"><primary>Machine Trust Account</primary></indexterm>
in which <replaceable>machine_nickname</replaceable> can be any
descriptive name for the client, such as BasementComputer.
<replaceable>machine_name</replaceable> absolutely must be the NetBIOS
name of the client to be joined to the domain. The <quote>$</quote> must be
appended to the NetBIOS name of the client or Samba will not recognize
this as a Machine Trust Account.
</para>

<para>
<indexterm significance="normal"><primary>UNIX account</primary></indexterm>
<indexterm significance="normal"><primary>Samba account</primary></indexterm>
<indexterm significance="normal"><primary>Machine Trust Account</primary><secondary>password</secondary></indexterm>
Now that the corresponding UNIX account has been created, the next step is to create 
the Samba account for the client containing the well-known initial 
Machine Trust Account password. This can be done using the 
<literal>smbpasswd</literal> command 
as shown here:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">smbpasswd -a -m <replaceable>machine_name</replaceable></userinput>
</screen>
</para>

<para>
<indexterm significance="normal"><primary>machine_name</primary></indexterm>
<indexterm significance="normal"><primary>NetBIOS name</primary></indexterm>
<indexterm significance="normal"><primary>RID</primary></indexterm>
<indexterm significance="normal"><primary>UID</primary></indexterm>
where <replaceable>machine_name</replaceable> is the machine's NetBIOS
name. The RID of the new machine account is generated from the UID of 
the corresponding UNIX account.
</para>

<warning>
<title>Join the client to the domain immediately</title>

<para>
<indexterm significance="normal"><primary>Machine Trust Account</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>Server Manager</primary></indexterm>
<indexterm significance="normal"><primary>changes password</primary></indexterm>
<indexterm significance="normal"><primary>NetBIOS name</primary></indexterm>
Manually creating a Machine Trust Account using this method is the 
equivalent of creating a Machine Trust Account on a Windows NT PDC using 
<indexterm significance="normal"><primary>Server Manager</primary></indexterm>
the <application moreinfo="none">Server Manager</application>. From the time at which the 
account is created to the time the client joins the domain and 
changes the password, your domain is vulnerable to an intruder joining 
your domain using a machine with the same NetBIOS name. A PDC inherently 
trusts members of the domain and will serve out a large degree of user 
information to such clients. You have been warned!
</para>
</warning>
</sect2>

<sect2>
<title>Managing Domain Machine Accounts using NT4 Server Manager</title>

<para>
<indexterm significance="normal"><primary>machine trust accounts</primary></indexterm>
<indexterm significance="normal"><primary>automatic account creation</primary></indexterm>
<indexterm significance="normal"><primary>Server Manager</primary></indexterm>
A working <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="ADDMACHINESCRIPT" xlink:href="smb.conf.5.html#ADDMACHINESCRIPT">add machine script</link> is essential
for machine trust accounts to be automatically created. This applies no matter whether
you use automatic account creation or the NT4 Domain Server Manager.
</para>

<para>
<indexterm significance="normal"><primary>SRVTOOLS.EXE</primary></indexterm>
<indexterm significance="normal"><primary>SrvMgr.exe</primary></indexterm>
<indexterm significance="normal"><primary>UsrMgr.exe</primary></indexterm>
<indexterm significance="normal"><primary>domain management tools</primary></indexterm>
If the machine from which you are trying to manage the domain is an 
<application moreinfo="none">MS Windows NT4 workstation or MS Windows 200x/XP Professional</application>,
the tool of choice is the package called <literal>SRVTOOLS.EXE</literal>. 
When executed in the target directory it will unpack <literal>SrvMgr.exe</literal>
and <literal>UsrMgr.exe</literal> (both are domain management tools for MS Windows NT4 workstation).
</para>

<para>
<indexterm significance="normal"><primary>Nexus.exe</primary></indexterm>
<indexterm significance="normal"><primary>Microsoft Windows 9x/Me</primary></indexterm>
If your workstation is a <application moreinfo="none">Microsoft Windows 9x/Me</application> family product,
 you should download the <literal>Nexus.exe</literal> package from the Microsoft Web site.
When executed from the target directory, it will unpack the same tools but for use on 
this platform.
</para>

<para>
Further information about these tools may be obtained from Knowledge Base articles
<ulink url="http://support.microsoft.com/default.aspx?scid=kb;en-us;173673">173673</ulink>, and
<ulink url="http://support.microsoft.com/default.aspx?scid=kb;en-us;172540">172540</ulink>
</para>

<para>
<indexterm significance="normal"><primary>srvmgr.exe</primary></indexterm>
<indexterm significance="normal"><primary>Server Manager for Domains</primary></indexterm>
Launch the <literal>srvmgr.exe</literal> (Server Manager for Domains) and follow these steps:
</para>

<procedure>
<title>Server Manager Account Machine Account Management</title>
	<step performance="required"><para>
	From the menu select <guimenu moreinfo="none">Computer</guimenu>.
	</para></step>

	<step performance="required"><para>
	Click <guimenuitem moreinfo="none">Select Domain</guimenuitem>.
	</para></step>

	<step performance="required"><para>
	Click the name of the domain you wish to administer in the
	<guilabel moreinfo="none">Select Domain</guilabel> panel and then click 
	<guibutton moreinfo="none">OK</guibutton>.
	</para></step>

	<step performance="required"><para>
	Again from the menu select <guimenu moreinfo="none">Computer</guimenu>.
	</para></step>

	<step performance="required"><para>
	Select <guimenuitem moreinfo="none">Add to Domain</guimenuitem>.
	</para></step>

	<step performance="required"><para>
	In the dialog box, click the radio button to 
	<guilabel moreinfo="none">Add NT Workstation of Server</guilabel>, then
	enter the machine name in the field provided, and click the 
	<guibutton moreinfo="none">Add</guibutton> button.
	</para></step>
</procedure>

</sect2>

<sect2>
<title>On-the-Fly Creation of Machine Trust Accounts</title>

<para>
<indexterm significance="normal"><primary>Machine Trust Account</primary><secondary>creation</secondary></indexterm>
The third (and recommended) way of creating Machine Trust Accounts is simply to allow the Samba server to
create them as needed when the client is joined to the domain.
</para>

<para>
<indexterm significance="normal"><primary>Machine Trust Account</primary><secondary>UNIX account</secondary></indexterm>
<indexterm significance="normal"><primary>UNIX account</primary></indexterm>
<indexterm significance="normal"><primary>add machine script</primary></indexterm>
Since each Samba Machine Trust Account requires a corresponding UNIX account, a method
for automatically creating the UNIX account is usually supplied; this requires configuration of the
add machine script option in <filename moreinfo="none">smb.conf</filename>. This method is not required; however, corresponding UNIX
accounts may also be created manually.
</para>


<para>
<indexterm significance="normal"><primary>useradd</primary></indexterm>
<indexterm significance="normal"><primary>Red Hat Linux</primary></indexterm>
Here is an example for a Red Hat Linux system:
<simplelist>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>add machine script</indexterm><parameter>add machine script = /usr/sbin/useradd -d /var/lib/nobody -g 100 -s /bin/false -M %u</parameter></member>
</simplelist>
</para>

</sect2>

<sect2><title>Making an MS Windows Workstation or Server a Domain Member</title>

<para>
The procedure for making an MS Windows workstation or server a member of the domain varies
with the version of Windows.
</para>

<sect3>
	<title>Windows 200x/XP Professional Client</title>

	<para>
<indexterm significance="normal"><primary>domain member</primary></indexterm>
<indexterm significance="normal"><primary>machine trust account</primary><secondary>create privilege</secondary></indexterm>
<indexterm significance="normal"><primary>privileges</primary></indexterm>
<indexterm significance="normal"><primary>root</primary></indexterm>
	When the user elects to make the client a domain member, Windows 200x prompts for
	an account and password that has privileges to create  machine accounts in the domain.
	A Samba administrator account (i.e., a Samba account that has <constant>root</constant> privileges on the
	Samba server) must be entered here; the operation will fail if an ordinary user
	account is given. 
	</para>

	<para>
<indexterm significance="normal"><primary>administrator account</primary></indexterm>
<indexterm significance="normal"><primary>/etc/passwd</primary></indexterm>
	For security reasons, the password for this administrator account should be set
	to a password that is other than that used for the root user in <filename moreinfo="none">/etc/passwd</filename>.
	</para>

	<para>
<indexterm significance="normal"><primary>account</primary></indexterm>
<indexterm significance="normal"><primary>create domain member</primary></indexterm>
<indexterm significance="normal"><primary>root</primary></indexterm>
<indexterm significance="normal"><primary>map</primary></indexterm>
	The name of the account that is used to create domain member machine trust accounts can be
	anything the network administrator may choose. If it is other than <constant>root</constant>,
	then this is easily mapped to <constant>root</constant> in the file named in the <filename moreinfo="none">smb.conf</filename> parameter
	<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="USERNAMEMAP" xlink:href="smb.conf.5.html#USERNAMEMAP">username map = /etc/samba/smbusers</link>.
	</para>

	<para>
<indexterm significance="normal"><primary>administrator account</primary></indexterm>
<indexterm significance="normal"><primary>encryption key</primary></indexterm>
<indexterm significance="normal"><primary>machine trust account</primary></indexterm>
	The session key of the Samba administrator account acts as an encryption key for setting the password of the machine trust
	account. The Machine Trust Account will be created on-the-fly, or updated if it already exists.
	</para>
</sect3>

<sect3>
	<title>Windows NT4 Client</title>

	<para>
<indexterm significance="normal"><primary>Machine Trust Account</primary></indexterm>
<indexterm significance="normal"><primary>Create a Computer Account</primary></indexterm>
<indexterm significance="normal"><primary>join the machine</primary></indexterm>
	If the Machine Trust Account was created manually, on the
	Identification Changes menu enter the domain name, but do not
	check the box <guilabel moreinfo="none">Create a Computer Account in the Domain</guilabel>.
	In this case, the existing Machine Trust Account is used to join the machine 
	to the domain.
	</para>

	<para>
<indexterm significance="normal"><primary>Machine Trust Account</primary></indexterm>
<indexterm significance="normal"><primary>on the fly</primary></indexterm>
<indexterm significance="normal"><primary>Computer Account</primary></indexterm>
<indexterm significance="normal"><primary>administrator account</primary></indexterm>
	If the Machine Trust Account is to be created on the fly, on the Identification Changes menu enter the domain
	name and check the box <guilabel moreinfo="none">Create a Computer Account in the Domain</guilabel>. In this case, joining
	the domain proceeds as above for Windows 2000 (i.e., you must supply a Samba administrator account when
	prompted).
	</para>
</sect3>

<sect3>
	<title>Samba Client</title>

	<para>
<indexterm significance="normal"><primary/></indexterm>
	Joining a Samba client to a domain is documented in <link linkend="domain-member-server">the next section</link>.
	</para>
</sect3>

</sect2>
</sect1>

<sect1 id="domain-member-server">
<title>Domain Member Server</title>

<para>
<indexterm significance="normal"><primary>domain security</primary></indexterm>
<indexterm significance="normal"><primary>security context</primary></indexterm>
<indexterm significance="normal"><primary>authentication regime</primary></indexterm>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
This mode of server operation involves the Samba machine being made a member
of a domain security context. This means by definition that all user
authentication will be done from a centrally defined authentication regime. 
The authentication regime may come from an NT3/4-style (old domain technology)
server, or it may be provided from an Active Directory server (ADS) running on
MS Windows 2000 or later.
</para>

<para>
<emphasis>
<indexterm significance="normal"><primary>authentication</primary><secondary>backend</secondary></indexterm>
<indexterm significance="normal"><primary>distributed directory</primary></indexterm>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
<indexterm significance="normal"><primary>OpenLDAP</primary></indexterm>
<indexterm significance="normal"><primary>iPlanet</primary></indexterm>
<indexterm significance="normal"><primary>Sun</primary></indexterm>
<indexterm significance="normal"><primary>Novell</primary></indexterm>
<indexterm significance="normal"><primary>e-Directory</primary></indexterm>
Of course it should be clear that the authentication backend itself could be
from any distributed directory architecture server that is supported by Samba.
This can be LDAP (from OpenLDAP), or Sun's iPlanet, or Novell e-Directory
Server, and so on.
</emphasis>
</para>

<note><para>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
<indexterm significance="normal"><primary>identity management</primary></indexterm>
<indexterm significance="normal"><primary>machine authentication</primary></indexterm>
When Samba is configured to use an LDAP or other identity management and/or
directory service, it is Samba that continues to perform user and machine
authentication. It should be noted that the LDAP server does not perform
authentication handling in place of what Samba is designed to do.
</para></note>

<para>
<indexterm significance="normal"><primary>create a domain machine account</primary></indexterm>
<indexterm significance="normal"><primary>domain member server</primary></indexterm>
<indexterm significance="normal"><primary>join the domain</primary></indexterm>
Please refer to <link linkend="samba-pdc">Domain Control</link>, for more information regarding
how to create a domain machine account for a domain member server as well as for
information on how to enable the Samba domain member machine to join the domain
and be fully trusted by it.
</para>

<sect2>
<title>Joining an NT4-type Domain with Samba-3</title>

<para><link linkend="assumptions">Assumptions</link> lists names that are used in the remainder of this chapter.</para>

<table frame="all" id="assumptions"><title>Assumptions</title>
	<tgroup cols="2">
		<colspec align="right"/>
		<colspec align="left"/>
	<tbody>
			<row>
				<entry>Samba DMS NetBIOS name:</entry><entry>SERV1</entry>
			</row>
			<row>
				<entry>Windows 200x/NT domain name:</entry><entry>MIDEARTH</entry>
			</row>
			<row>
				<entry>Domain's PDC NetBIOS name:</entry><entry>DOMPDC</entry>
			</row>
			<row>
				<entry>Domain's BDC NetBIOS names:</entry><entry>DOMBDC1 and DOMBDC2</entry>
			</row>
	</tbody>
	</tgroup>
</table>

<para>
<indexterm significance="normal"><primary/></indexterm>
First, you must edit your <filename moreinfo="none">smb.conf</filename> file to tell Samba it should now use domain security.
</para>

<para>
<indexterm significance="normal"><primary>security = user</primary></indexterm>
<indexterm significance="normal"><primary>standalone server</primary></indexterm>
<indexterm significance="normal"><primary>domain member server</primary></indexterm>
<indexterm significance="normal"><primary>domain security</primary></indexterm>
Change (or add) your <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SECURITY" xlink:href="smb.conf.5.html#SECURITY">security</link> line in the [global] section 
of your <filename moreinfo="none">smb.conf</filename> to read:
<simplelist>
<member><indexterm>security</indexterm><parameter>security = domain</parameter></member>
</simplelist>
Note that if the parameter <parameter moreinfo="none">security = user</parameter> is used, this machine would function as a
standalone server and not as a domain member server. Domain security mode causes Samba to work within the
domain security context.
</para>

<para>
Next change the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="WORKGROUP" xlink:href="smb.conf.5.html#WORKGROUP">workgroup</link> line in the <parameter>[global]</parameter>
section to read: 
<simplelist>
<member><indexterm>workgroup</indexterm><parameter>workgroup = MIDEARTH</parameter></member>
</simplelist>
This is the name of the domain we are joining.
</para>

<para>
<indexterm significance="normal"><primary>authenticate</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
You must also have the parameter <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="ENCRYPTPASSWORDS" xlink:href="smb.conf.5.html#ENCRYPTPASSWORDS">encrypt passwords</link>
set to <constant>yes</constant> in order for your users to authenticate to the NT PDC.
This is the default setting if this parameter is not specified. There is no need to specify this
parameter, but if it is specified in the <filename moreinfo="none">smb.conf</filename> file, it must be set to <constant>Yes</constant>.
</para>

<para>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>BDC</primary></indexterm>
<indexterm significance="normal"><primary>authenticate users</primary></indexterm>
<indexterm significance="normal"><primary>domain controllers</primary></indexterm>
Finally, add (or modify) a <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PASSWORDSERVER" xlink:href="smb.conf.5.html#PASSWORDSERVER">password server</link> line in the [global]
section to read: 
<simplelist>
<member><indexterm>password server</indexterm><parameter>password server = DOMPDC DOMBDC1 DOMBDC2</parameter></member>
</simplelist>
These are the PDC and BDCs Samba 
will attempt to contact in order to authenticate users. Samba will 
try to contact each of these servers in order, so you may want to 
rearrange this list in order to spread out the authentication load 
among Domain Controllers.
</para>

<para>
<indexterm significance="normal"><primary>list of domain controllers</primary></indexterm>
<indexterm significance="normal"><primary>mechanism</primary></indexterm>
<indexterm significance="normal"><primary>broadcast-based name resolution</primary></indexterm>
<indexterm significance="normal"><primary>DNS name resolution</primary></indexterm>
Alternatively, if you want smbd to determine automatically the list of domain controllers to use for
authentication, you may set this line to be:
<simplelist>
<member><indexterm>password server</indexterm><parameter>password server = *</parameter></member>
</simplelist>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
This method allows Samba to use exactly the same mechanism that NT does. The 
method either uses broadcast-based name resolution, performs a WINS database
lookup in order to find a domain controller against which to authenticate,
or locates the domain controller using DNS name resolution.
</para>

<para>
To join the domain, run this command:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>join</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">net rpc join -S DOMPDC -U<replaceable>Administrator%password</replaceable></userinput>
</screen>
</para>

<para>
<indexterm significance="normal"><primary>NetBIOS name</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>WINS lookup</primary></indexterm>
<indexterm significance="normal"><primary>NetBIOS broadcast</primary></indexterm>
If the <option>-S DOMPDC</option> argument is not given, the domain name will be obtained from <filename moreinfo="none">smb.conf</filename> and
the NetBIOS name of the PDC will be obtained either using a WINS lookup or via NetBIOS broadcast based name
look up.
</para>

<para>
<indexterm significance="normal"><primary>joining the domain</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>Administrator%password</primary></indexterm>
<indexterm significance="normal"><primary>Joined domain</primary></indexterm>
The machine is joining the domain DOM, and the PDC for that domain (the only machine
that has write access to the domain SAM database) is DOMPDC; therefore, use the <option>-S</option>
option. The <replaceable>Administrator%password</replaceable> is the login name and
password for an account that has the necessary privilege to add machines to the
domain. If this is successful, you will see the following message in your terminal window.
Where the older NT4-style domain architecture is used:
<screen format="linespecific">
<computeroutput moreinfo="none">Joined domain DOM.</computeroutput>
</screen>
</para>

<para>
<indexterm significance="normal"><primary>net</primary><secondary>ads</secondary><tertiary>join</tertiary></indexterm>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
<indexterm significance="normal"><primary>join the ADS domain</primary></indexterm>
Where Active Directory is used, the command used to join the ADS domain is:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net ads join -U<replaceable>Administrator%password</replaceable>
</screen>
And the following output is indicative of a successful outcome:
<screen format="linespecific">
<computeroutput moreinfo="none">Joined SERV1 to realm MYREALM.</computeroutput>
</screen>
</para>

<para>
Refer to the <literal>net</literal> man page and to <link linkend="NetCommand">the chapter on remote
administration</link> for further information.
</para>

<para>
<indexterm significance="normal"><primary>join the domain</primary></indexterm>
<indexterm significance="normal"><primary>create machine trust account</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
This process joins the server to the domain without separately having to create the machine
trust account on the PDC beforehand.
</para>

<para>
<indexterm significance="normal"><primary>machine account password</primary><secondary>change protocol</secondary></indexterm>
<indexterm significance="normal"><primary>random machine account password</primary></indexterm>
<indexterm significance="normal"><primary>/usr/local/samba/private/secrets.tdb</primary></indexterm>
<indexterm significance="normal"><primary>/etc/samba/secrets.tdb</primary></indexterm>
This command goes through the machine account password change protocol, then writes the new (random) machine
account password for this Samba server into a file in the same directory in which a smbpasswd file would be
normally stored. The trust account information that is needed by the DMS is written into the file
<filename moreinfo="none">/usr/local/samba/private/secrets.tdb</filename> or <filename moreinfo="none">/etc/samba/secrets.tdb</filename>.
</para>

<para>
<indexterm significance="normal"><primary>domain-level security</primary></indexterm>
<indexterm significance="normal"><primary>shadow password file</primary></indexterm>
This file is created and owned by root and is not readable by any other user. It is
the key to the domain-level security for your system and should be treated as carefully 
as a shadow password file.
</para>

<para>
<indexterm significance="normal"><primary>Samba daemons</primary></indexterm>
<indexterm significance="normal"><primary>distribution</primary></indexterm>
<indexterm significance="normal"><primary>/etc/init.d/samba</primary></indexterm>
Finally, restart your Samba daemons and get ready for clients to begin using domain
security. The way you can restart your Samba daemons depends on your distribution,
but in most cases the following will suffice:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt>/etc/init.d/samba restart
</screen>
</para>

</sect2>

<sect2>
<title>Why Is This Better Than <parameter moreinfo="none">security = server</parameter>?</title>

<para>
<indexterm significance="normal"><primary>domain security</primary></indexterm>
<indexterm significance="normal"><primary>UNIX users</primary></indexterm>
<indexterm significance="normal"><primary>authentication</primary></indexterm>
Currently, domain security in Samba does not free you from having to create local UNIX users to represent the
users attaching to your server. This means that if domain user <constant>DOM\fred</constant> attaches to your
domain security Samba server, there needs to be a local UNIX user fred to represent that user in the UNIX file
system. This is similar to the older Samba security mode <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SECURITY" xlink:href="smb.conf.5.html#SECURITY">security = server</link>, where Samba would pass through the authentication request to a Windows
NT server in the same way as a Windows 95 or Windows 98 server would.
</para>

<para>
<indexterm significance="normal"><primary>winbind</primary></indexterm>
<indexterm significance="normal"><primary>UID</primary></indexterm>
<indexterm significance="normal"><primary>GID</primary></indexterm>
Please refer to <link linkend="winbind">Winbind: Use of Domain Accounts</link>, for information on a system
to automatically assign UNIX UIDs and GIDs to Windows NT domain users and groups.
</para>

<para>
<indexterm significance="normal"><primary>domain-level</primary></indexterm>
<indexterm significance="normal"><primary>authentication</primary></indexterm>
<indexterm significance="normal"><primary>RPC</primary></indexterm>
The advantage of domain-level security is that the authentication in domain-level security is passed down the
authenticated RPC channel in exactly the same way that an NT server would do it. This means Samba servers now
participate in domain trust relationships in exactly the same way NT servers do (i.e., you can add Samba
servers into a resource domain and have the authentication passed on from a resource domain PDC to an account
domain PDC).
</para>

<para>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>BDC</primary></indexterm>
<indexterm significance="normal"><primary>connection resources</primary></indexterm>
In addition, with <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SECURITY" xlink:href="smb.conf.5.html#SECURITY">security = server</link>, every Samba daemon on a server has to
keep a connection open to the authenticating server for as long as that daemon lasts. This can drain the
connection resources on a Microsoft NT server and cause it to run out of available connections. With
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SECURITY" xlink:href="smb.conf.5.html#SECURITY">security = domain</link>, however, the Samba daemons connect to the PDC or BDC
only for as long as is necessary to authenticate the user and then drop the connection, thus conserving PDC
connection resources.
</para>

<para>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>authentication reply</primary></indexterm>
<indexterm significance="normal"><primary>SID</primary></indexterm>
<indexterm significance="normal"><primary>NT groups</primary></indexterm>
Finally, acting in the same manner as an NT server authenticating to a PDC means that as part of the
authentication reply, the Samba server gets the user identification information such as the user SID, the list
of NT groups the user belongs to, and so on.
</para>

<note>
<para>
Much of the text of this document was first published in the Web magazine 
<ulink url="http://www.linuxworld.com"><emphasis>LinuxWorld</emphasis></ulink> as the article <ulink url="http://www.linuxworld.com/linuxworld/lw-1998-10/lw-10-samba.html"/>
<emphasis>Doing the NIS/NT Samba</emphasis>.
</para>
</note>

</sect2>
</sect1>

<sect1 id="ads-member">
<title>Samba ADS Domain Membership</title>

<para>
<indexterm significance="preferred"><primary>Active Directory</primary></indexterm>
<indexterm significance="preferred"><primary>ADS</primary><see>Active Directory</see></indexterm>
<indexterm significance="normal"><primary>KDC</primary></indexterm>
<indexterm significance="normal"><primary>Kerberos</primary></indexterm>
This is a rough guide to setting up Samba-3 with Kerberos authentication against a
Windows 200x KDC. A familiarity with Kerberos is assumed.
</para> 

<sect2>
<title>Configure <filename moreinfo="none">smb.conf</filename></title>

<para>
You must use at least the following three options in <filename moreinfo="none">smb.conf</filename>:
</para>

<simplelist>
<member><indexterm>realm</indexterm><parameter>realm = your.kerberos.REALM</parameter></member>
<member><indexterm>security</indexterm><parameter>security = ADS</parameter></member>
<member># The following parameter need only be specified if present.</member>
<member># The default setting if not present is Yes.</member>
<member><indexterm>encrypt passwords</indexterm><parameter>encrypt passwords = yes</parameter></member>
</simplelist>

<para>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
<indexterm significance="normal"><primary>realm</primary></indexterm>
<indexterm significance="normal"><primary>DNS</primary></indexterm>
<indexterm significance="normal"><primary>ADS DC</primary></indexterm>
<indexterm significance="normal"><primary>password server</primary></indexterm>
In case samba cannot correctly identify the appropriate ADS server using the realm name, use the 
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PASSWORDSERVER" xlink:href="smb.conf.5.html#PASSWORDSERVER">password server</link> option in <filename moreinfo="none">smb.conf</filename>:
<simplelist>
<member><indexterm>password server</indexterm><parameter>password server = your.kerberos.server</parameter></member>
</simplelist>
The most common reason for which Samba may not be able to locate the ADS domain controller is a consequence of
sites maintaining some DNS servers on UNIX systems without regard for the DNS requirements of the ADS
infrastructure. There is no harm in specifying a preferred ADS domain controller using the <parameter moreinfo="none">password
server</parameter>.
</para>

<note><para>
<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
<indexterm significance="normal"><primary>authenticated</primary></indexterm>
You do <emphasis>not</emphasis> need an smbpasswd file, and older clients will be authenticated as 
if <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SECURITY" xlink:href="smb.conf.5.html#SECURITY">security = domain</link>, although it will not do any harm and 
allows you to have local users not in the domain.
</para></note>

</sect2>
  
<sect2>
<title>Configure <filename moreinfo="none">/etc/krb5.conf</filename></title>

<para>
<indexterm significance="normal"><primary>/etc/krb5.conf</primary></indexterm>
<indexterm significance="normal"><primary>Kerberos</primary><secondary>/etc/krb5.conf</secondary></indexterm>
<indexterm significance="normal"><primary>MIT</primary></indexterm>
<indexterm significance="normal"><primary>Heimdal</primary></indexterm>
With both MIT and Heimdal Kerberos, it is unnecessary to configure the <filename moreinfo="none">/etc/krb5.conf</filename>,
and it may be detrimental.
</para>

<para>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
<indexterm significance="normal"><primary>SRV records</primary></indexterm>
<indexterm significance="normal"><primary>DNS zon</primary></indexterm>
<indexterm significance="normal"><primary>KDC</primary></indexterm>
<indexterm significance="normal"><primary>_kerberos.REALM.NAME</primary></indexterm>
Microsoft ADS automatically create SRV records in the DNS zone 
<parameter moreinfo="none">_kerberos._tcp.REALM.NAME</parameter> for each KDC in the realm. This is part
of the installation and configuration process used to create an Active Directory domain.
A KDC is a Kerberos Key Distribution Center and forms an integral part of the Microsoft
active directory infrastructure.
</para>

<para>
<indexterm significance="normal"><primary>kinit</primary></indexterm>
<indexterm significance="normal"><primary>DES-CBC-MD5</primary></indexterm>
<indexterm significance="normal"><primary>DES-CBC-CRC</primary></indexterm>
<indexterm significance="normal"><primary>encryption types</primary></indexterm>
<indexterm significance="normal"><primary>kerberos</primary></indexterm>
<indexterm significance="normal"><primary>Windows 2000</primary></indexterm>
UNIX systems can use kinit and the DES-CBC-MD5 or DES-CBC-CRC encryption types to authenticate to the Windows
2000 KDC. For further information regarding Windows 2000 ADS kerberos interoperability please refer to the
Microsoft Windows 2000 Kerberos <ulink url="http://www.microsoft.com/windows2000/techinfo/planning/security/kerbsteps.asp">Interoperability</ulink>
guide. Another very useful document that may be referred to for general information regarding Kerberos
interoperability is <ulink url="http://www.ietf.org/rfc/rfc1510.txt?number=1510">RFC1510</ulink>. This RFC
explains much of the magic behind the operation of Kerberos.
</para>

<para>
<indexterm significance="normal"><primary>MIT</primary></indexterm>
<indexterm significance="normal"><primary>KRB5</primary></indexterm>
<indexterm significance="normal"><primary>SRV records</primary></indexterm>
<indexterm significance="normal"><primary>krb5.conf</primary></indexterm>
<indexterm significance="normal"><primary>DNS lookup</primary></indexterm>
<indexterm significance="normal"><primary>libraries</primary></indexterm>
MIT's, as well as Heimdal's, recent KRB5 libraries default to checking for SRV records, so they will 
automatically find the KDCs. In addition, <filename moreinfo="none">krb5.conf</filename> only allows specifying 
a single KDC, even there if there may be more than one. Using the DNS lookup allows the KRB5 
libraries to use whichever KDCs are available.
</para>

<para>
<indexterm significance="normal"><primary>krb5.conf</primary></indexterm>
When manually configuring <filename moreinfo="none">krb5.conf</filename>, the minimal configuration is:
<screen format="linespecific">
[libdefaults]
	default_realm = YOUR.KERBEROS.REALM

[realms]
	YOUR.KERBEROS.REALM = {
	kdc = your.kerberos.server
	}

[domain_realms]
	.kerberos.server = YOUR.KERBEROS.REALM
</screen>
</para>

<para>
<indexterm significance="normal"><primary>Heimdal</primary></indexterm>
When using Heimdal versions before 0.6, use the following configuration settings:
<screen format="linespecific">
[libdefaults]
	default_realm      = YOUR.KERBEROS.REALM
	default_etypes     = des-cbc-crc des-cbc-md5
	default_etypes_des = des-cbc-crc des-cbc-md5

[realms]
        YOUR.KERBEROS.REALM = {
        kdc = your.kerberos.server
	}

[domain_realms]
        .kerberos.server = YOUR.KERBEROS.REALM
</screen>
</para>

<para>
<indexterm significance="normal"><primary>KDC</primary></indexterm>
<indexterm significance="normal"><primary>kinit</primary></indexterm>
Test your config by doing a <userinput moreinfo="none">kinit
<replaceable>USERNAME</replaceable>@<replaceable>REALM</replaceable></userinput> and
making sure that your password is accepted by the Win2000 KDC.
</para>

<para>
<indexterm significance="normal"><primary>Heimdal</primary></indexterm>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
<indexterm significance="normal"><primary>KDC</primary></indexterm>
<indexterm significance="normal"><primary>Windows 2003</primary></indexterm>
With Heimdal versions earlier than 0.6.x you can use only newly created accounts
in ADS or accounts that have had the password changed once after migration, or
in case of <constant>Administrator</constant> after installation. At the
moment, a Windows 2003 KDC can only be used with Heimdal releases later than 0.6
(and no default etypes in krb5.conf). Unfortunately, this whole area is still
in a state of flux.
</para>

<note><para>
<indexterm significance="normal"><primary>realm</primary></indexterm>
<indexterm significance="normal"><primary>uppercase</primary></indexterm>
<indexterm significance="normal"><primary>KDC</primary></indexterm>
The realm must be in uppercase or you will get a <quote><errorname>Cannot find KDC for
requested realm while getting initial credentials</errorname></quote> error (Kerberos
is case-sensitive!).
</para></note>

<note><para>
<indexterm significance="normal"><primary>synchronize</primary></indexterm>
<indexterm significance="normal"><primary>credentials</primary></indexterm>
<indexterm significance="normal"><primary>time difference</primary></indexterm>
<indexterm significance="normal"><primary>clock skew</primary></indexterm>
Time between the two servers must be synchronized. You will get a <quote><errorname>kinit(v5): Clock skew too
great while getting initial credentials</errorname></quote> if the time difference (clock skew) is more than five minutes.
</para></note>

<para>
<indexterm significance="normal"><primary>clock skew</primary></indexterm>
<indexterm significance="normal"><primary>Kerberos</primary></indexterm>
Clock skew limits are configurable in the Kerberos protocols. The default setting is five minutes.
</para>

<para>
<indexterm significance="normal"><primary>DNS</primary></indexterm>
<indexterm significance="normal"><primary>KDC</primary></indexterm>
<indexterm significance="normal"><primary>hostname</primary></indexterm>
<indexterm significance="normal"><primary>realm</primary></indexterm>
You also must ensure that you can do a reverse DNS lookup on the IP address of your KDC. Also, the name that
this reverse lookup maps to must either be the NetBIOS name of the KDC (i.e., the hostname with no domain
attached) or it can be the NetBIOS name followed by the realm.
</para>

<para>
<indexterm significance="normal"><primary>/etc/hosts</primary></indexterm>
<indexterm significance="normal"><primary>KDC</primary></indexterm>
<indexterm significance="normal"><primary>realm</primary></indexterm>
The easiest way to ensure you get this right is to add a <filename moreinfo="none">/etc/hosts</filename> entry mapping the IP
address of your KDC to its NetBIOS name. If you do not get this correct, then you will get a <errorname>local
error</errorname> when you try to join the realm.
</para>

<para>
<indexterm significance="normal"><primary>Kerberos</primary></indexterm>
<indexterm significance="normal"><primary>Create the Computer Account</primary></indexterm>
<indexterm significance="normal"><primary>Testing Server Setup</primary></indexterm>
<indexterm significance="normal"><primary/></indexterm>
If all you want is Kerberos support in <application moreinfo="none">smbclient</application>, then you can skip directly to <link linkend="ads-test-smbclient">Testing with <application moreinfo="none">smbclient</application></link> now.  <link linkend="ads-create-machine-account">Create the Computer Account</link> and <link linkend="ads-test-server">Testing Server Setup</link> are needed only if you want Kerberos support for <application moreinfo="none">smbd</application>
and <application moreinfo="none">winbindd</application>.
</para>

</sect2>

<sect2 id="ads-create-machine-account">
<title>Create the Computer Account</title>

<para>
<indexterm significance="normal"><primary>write permission</primary></indexterm>
<indexterm significance="normal"><primary>Samba private directory</primary></indexterm>
<indexterm significance="normal"><primary>Administrator account</primary></indexterm>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
As a user who has write permission on the Samba private directory (usually root), run:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> <userinput moreinfo="none">net ads join -U Administrator%password</userinput>
</screen>
The Administrator account can be any account that has been designated in the ADS domain security settings with
permission to add machines to the ADS domain. It is, of course, a good idea to use an account other than Administrator.
On the UNIX/Linux system, this command must be executed by an account that has UID=0 (root).
</para>

<para>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
<indexterm significance="normal"><primary>machine trust account</primary></indexterm>
<indexterm significance="normal"><primary>organizational unit</primary></indexterm>
<indexterm significance="normal"><primary>ADS manager</primary></indexterm>
<indexterm significance="normal"><primary>kinit</primary></indexterm>
<indexterm significance="normal"><primary>net</primary><secondary>ads</secondary><tertiary>join</tertiary></indexterm>
When making a Windows client a member of an ADS domain within a complex organization, you
may want to create the machine trust account within a particular organizational unit. Samba-3 permits
this to be done using the following syntax:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> <userinput moreinfo="none">kinit Administrator@your.kerberos.REALM</userinput>
<prompt moreinfo="none">root# </prompt> <userinput moreinfo="none">net ads join createcomputer="organizational_unit"</userinput>
</screen>
Your ADS manager will be able to advise what should be specified for the "organizational_unit" parameter.
</para>

<para>
<indexterm significance="normal"><primary>organizational directory</primary></indexterm>
<indexterm significance="normal"><primary>machine trust account</primary></indexterm>
<indexterm significance="normal"><primary>container</primary></indexterm>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
For example, you may want to create the machine trust account in a container called <quote>Servers</quote>
under the organizational directory <quote>Computers/BusinessUnit/Department,</quote> like this:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> <userinput moreinfo="none">net ads join "Computers/BusinessUnit/Department/Servers"</userinput>
</screen>
This command will place the Samba server machine trust account in the container
<literal moreinfo="none">Computers/BusinessUnit/Department/Servers</literal>. The container should exist in the ADS directory
before executing this command.  Please note that forward slashes must be used, because backslashes are both
valid characters in an OU name and used as escapes for other characters.  If you need a backslash in an OU 
name, it may need to be quadrupled to pass through the shell escape and ldap escape.
</para>

<sect3>
<title>Possible Errors</title>

<para>
<variablelist>
	<varlistentry><term><errorname>ADS support not compiled in</errorname></term>
	<listitem><para>
	<indexterm significance="normal"><primary>config.cache</primary></indexterm>
	<indexterm significance="normal"><primary>Kerberos</primary></indexterm>
	<indexterm significance="normal"><primary>headers files</primary></indexterm>
	Samba must be reconfigured (remove config.cache) and recompiled (make clean all install) after the
	Kerberos libraries and headers files are installed.
	</para></listitem></varlistentry>

	<varlistentry><term><errorname>net ads join prompts for user name</errorname></term>
	<listitem><para>
	<indexterm significance="normal"><primary>kinit</primary></indexterm>
	<indexterm significance="normal"><primary>rights</primary></indexterm>
	You need to log in to the domain using <userinput moreinfo="none">kinit
	<replaceable>USERNAME</replaceable>@<replaceable>REALM</replaceable></userinput>.
	<replaceable>USERNAME</replaceable> must be a user who has rights to add a machine to the domain.
	</para></listitem></varlistentry>

	<varlistentry><term>Unsupported encryption/or checksum types</term>
	<listitem><para>
	<indexterm significance="normal"><primary>/etc/krb5.conf</primary></indexterm>
	<indexterm significance="normal"><primary>unsupported encryption</primary></indexterm>
	<indexterm significance="normal"><primary>Kerberos</primary></indexterm>
	Make sure that the <filename moreinfo="none">/etc/krb5.conf</filename> is correctly configured
	for the type and version of Kerberos installed on the system.
	</para></listitem></varlistentry>
</variablelist>
</para>

</sect3>

</sect2>

<sect2 id="ads-test-server">
<title>Testing Server Setup</title>

<para>
<indexterm significance="normal"><primary>successful join</primary></indexterm>
<indexterm significance="normal"><primary>computer account</primary></indexterm>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
If the join was successful, you will see a new computer account with the
NetBIOS name of your Samba server in Active Directory (in the <quote>Computers</quote>
folder under Users and Computers.
</para>

<para>
<indexterm significance="normal"><primary>Windows 2000</primary></indexterm>
<indexterm significance="normal"><primary>net</primary><secondary>use</secondary></indexterm>
<indexterm significance="normal"><primary>DES-CBC-MD5</primary></indexterm>
On a Windows 2000 client, try <userinput moreinfo="none">net use * \\server\share</userinput>. You should
be logged in with Kerberos without needing to know a password. If this fails, then run
<userinput moreinfo="none">klist tickets</userinput>. Did you get a ticket for the server? Does it have
an encryption type of DES-CBC-MD5? 
</para>

<note><para>
<indexterm significance="normal"><primary>DES-CBC-MD5</primary></indexterm>
<indexterm significance="normal"><primary>ARCFOUR-HMAC-MD5</primary></indexterm>
<indexterm significance="normal"><primary>encoding</primary></indexterm>
Samba can use both DES-CBC-MD5 encryption as well as ARCFOUR-HMAC-MD5 encoding.
</para></note>

</sect2>

<sect2 id="ads-test-smbclient">
<title>Testing with <application moreinfo="none">smbclient</application></title>

<para>
<indexterm significance="normal"><primary>smbclient</primary></indexterm>
<indexterm significance="normal"><primary>Kerberos</primary></indexterm>
<indexterm significance="normal"><primary>Kerberos authentication</primary></indexterm>
On your Samba server try to log in to a Windows 2000 server or your Samba
server using <application moreinfo="none">smbclient</application> and Kerberos. Use <application moreinfo="none">smbclient</application> as usual, but
specify the <option>-k</option> option to choose Kerberos authentication.
</para>

</sect2>

<sect2>
<title>Notes</title>

<para>
<indexterm significance="normal"><primary>administrator password</primary></indexterm>
<indexterm significance="normal"><primary>change password</primary></indexterm>
<indexterm significance="normal"><primary>encryption types</primary></indexterm>
You must change the administrator password at least once after installing a domain controller, 
to create the right encryption types.
</para>

<para>
<indexterm significance="normal"><primary>_kerberos._udp</primary></indexterm>
<indexterm significance="normal"><primary>_ldap._tcp</primary></indexterm>
<indexterm significance="normal"><primary>default DNS setup</primary></indexterm>
Windows 200x does not seem to create the <parameter moreinfo="none">_kerberos._udp</parameter> and
<parameter moreinfo="none">_ldap._tcp</parameter> in the default DNS setup. Perhaps this will be fixed later in service packs.
</para>

</sect2>
</sect1>

<sect1>
<title>Sharing User ID Mappings between Samba Domain Members</title>

<para>
<indexterm significance="normal"><primary>maps UNIX users and groups</primary></indexterm>
<indexterm significance="normal"><primary>UID</primary></indexterm>
<indexterm significance="normal"><primary>GID</primary></indexterm>
<indexterm significance="normal"><primary>SID</primary></indexterm>
Samba maps UNIX users and groups (identified by UIDs and GIDs) to Windows users and groups (identified by SIDs).
These mappings are done by the <parameter moreinfo="none">idmap</parameter> subsystem of Samba.
</para>

<para>
<indexterm significance="normal"><primary>mappings</primary></indexterm>
<indexterm significance="normal"><primary>CIFS</primary></indexterm>
<indexterm significance="normal"><primary>NFS</primary></indexterm>
In some cases it is useful to share these mappings between Samba domain members,
so <emphasis>name-&gt;id</emphasis> mapping is identical on all machines.
This may be needed in particular when sharing files over both CIFS and NFS.
</para>

<para>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
<indexterm significance="normal"><primary>ldap idmap suffix</primary></indexterm>
To use the <emphasis>LDAP</emphasis> <parameter moreinfo="none">ldap idmap suffix</parameter>, set:
</para>

<simplelist>
<member><indexterm>ldap idmap suffix</indexterm><parameter>ldap idmap suffix = ou=Idmap</parameter></member>
</simplelist>

<para>
See the <filename moreinfo="none">smb.conf</filename> man page entry for the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LDAPIDMAPSUFFIX" xlink:href="smb.conf.5.html#LDAPIDMAPSUFFIX">ldap idmap suffix</link>
parameter for further information.
</para>

<para>
<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
<indexterm significance="normal"><primary>LDAP administrative password</primary></indexterm>
<indexterm significance="normal"><primary>secrets.tdb</primary></indexterm>
Do not forget to specify also the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LDAPADMINDN" xlink:href="smb.conf.5.html#LDAPADMINDN">ldap admin dn</link>
and to make certain to set the LDAP administrative password into the <filename moreinfo="none">secrets.tdb</filename> using:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> smbpasswd -w ldap-admin-password
</screen>
In place of <literal moreinfo="none">ldap-admin-password</literal>, substitute the LDAP administration password for your
system.
</para>

</sect1>

<sect1>
<title>Common Errors</title>

<para>
<indexterm significance="normal"><primary>domain member</primary></indexterm>
<indexterm significance="normal"><primary>machine trust accounts</primary></indexterm>
In the process of adding/deleting/re-adding domain member machine trust accounts, there are
many traps for the unwary player and many <quote>little</quote> things that can go wrong.
It is particularly interesting how often subscribers on the Samba mailing list have concluded
after repeated failed attempts to add a machine account that it is necessary to <quote>reinstall</quote>
MS Windows on the machine. In truth, it is seldom necessary to reinstall because of this type
of problem. The real solution is often quite simple, and with an understanding of how MS Windows
networking functions, it is easy to overcome.
</para>

<sect2>
<title>Cannot Add Machine Back to Domain</title>

<para>
<indexterm significance="normal"><primary>machine trust account</primary></indexterm>
<indexterm significance="normal"><primary>already exists</primary></indexterm>
<quote>A Windows workstation was reinstalled. The original domain machine trust
account was deleted and added immediately. The workstation will not join the domain if I use 
the same machine name. Attempts to add the machine fail with a message that the machine already
exists on the network <?latex --- ?> I know it does not. Why is this failing?</quote>
</para>

<para>
<indexterm significance="normal"><primary>NetBIOS name cache</primary></indexterm>
<indexterm significance="normal"><primary>nbtstat</primary></indexterm>
The original name is still in the NetBIOS name cache and must expire after machine account
deletion before adding that same name as a domain member again. The best advice is to delete
the old account and then add the machine with a new name. Alternately, the name cache can be flushed and
reloaded with current data using the <literal>nbtstat</literal> command on the Windows client:
<screen format="linespecific">
<prompt moreinfo="none">C:\&gt; </prompt> nbtstat -R
</screen>
</para>

</sect2>

<sect2>
<title>Adding Machine to Domain Fails</title>

<para>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>fails</primary></indexterm>
<quote>Adding a Windows 200x or XP Professional machine to the Samba PDC Domain fails with a
message that says, <errorname>"The machine could not be added at this time, there is a network problem.
Please try again later."</errorname> Why?</quote>
</para>

<para>
<indexterm significance="normal"><primary>check logs</primary></indexterm>
You should check that there is an <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="ADDMACHINESCRIPT" xlink:href="smb.conf.5.html#ADDMACHINESCRIPT">add machine script</link> in your <filename moreinfo="none">smb.conf</filename>
file. If there is not, please add one that is appropriate for your OS platform. If a script
has been defined, you will need to debug its operation. Increase the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOGLEVEL" xlink:href="smb.conf.5.html#LOGLEVEL">log level</link>
in the <filename moreinfo="none">smb.conf</filename> file to level 10, then try to rejoin the domain. Check the logs to see which
operation is failing.
</para>

<para>
Possible causes include:
</para>

<itemizedlist>
	<listitem><para>
<indexterm significance="normal"><primary>script</primary></indexterm>
<indexterm significance="normal"><primary>path specified</primary></indexterm>
	The script does not actually exist, or could not be located in the path specified.
	</para>

	<para>
<indexterm significance="normal"><primary>UNIX system account</primary></indexterm>
<indexterm significance="normal"><primary>Samba SAM account</primary></indexterm>
	<emphasis>Corrective action:</emphasis> Fix it. Make sure when run manually
	that the script will add both the UNIX system account and the Samba SAM account.
	</para></listitem>

	<listitem><para>
<indexterm significance="normal"><primary>UNIX system account</primary></indexterm>
<indexterm significance="normal"><primary>/etc/passwd</primary></indexterm>
	The machine could not be added to the UNIX system accounts file <filename moreinfo="none">/etc/passwd</filename>.
	</para>

	<para>
<indexterm significance="normal"><primary>legal UNIX system account name</primary></indexterm>
<indexterm significance="normal"><primary>uppercase</primary></indexterm>
	<emphasis>Corrective action:</emphasis> Check that the machine name is a legal UNIX
	system account name. If the UNIX utility <literal>useradd</literal> is called,
	then make sure that the machine name you are trying to add can be added using this
	tool. <literal>Useradd</literal> on some systems will not allow any uppercase characters
	nor will it allow spaces in the name.
	</para></listitem>
</itemizedlist>

<para>
<indexterm significance="normal"><primary>backend database</primary></indexterm>
<indexterm significance="normal"><primary>UNIX system account</primary></indexterm>
<indexterm significance="normal"><primary>Samba backend database</primary></indexterm>
The <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="ADDMACHINESCRIPT" xlink:href="smb.conf.5.html#ADDMACHINESCRIPT">add machine script</link> does not create the
machine account in the Samba backend database; it is there only to create a UNIX system
account to which the Samba backend database account can be mapped.
</para>

</sect2>

<sect2>
	<title>I Can't Join a Windows 2003 PDC</title>

	<para>
<indexterm significance="normal"><primary>SMB signing</primary></indexterm>
<indexterm significance="normal"><primary>SMB</primary></indexterm>
<indexterm significance="normal"><primary>Windows 2003</primary></indexterm>
<indexterm significance="normal"><primary>SMB/CIFS</primary></indexterm>
	Windows 2003 requires SMB signing. Client-side SMB signing has been implemented in Samba-3.0.
	Set <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="CLIENTUSESPNEGO" xlink:href="smb.conf.5.html#CLIENTUSESPNEGO">client use spnego = yes</link> when communicating 
	with a Windows 2003 server. This will not interfere with other Windows clients that do not
	support the more advanced security features of Windows 2003 because the client will simply
	negotiate a protocol that both it and the server suppport. This is a well-known fall-back facility
	that is built into the SMB/CIFS protocols.
	</para>

</sect2>

</sect1>
</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="StandAloneServer">
<chapterinfo>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
</chapterinfo>
<title>Standalone Servers</title>

<para>
<indexterm significance="normal"><primary>standalone server</primary></indexterm>
<indexterm significance="normal"><primary>not domain members</primary></indexterm>
<indexterm significance="normal"><primary>minimum security control</primary></indexterm>
Standalone servers are independent of domain controllers on the network.
They are not domain members and function more like workgroup servers. In many
cases a standalone server is configured with a minimum of security control
with the intent that all data served will be readily accessible to all users.
</para>

<sect1>
<title>Features and Benefits</title>

<para>
<indexterm significance="normal"><primary>secure</primary></indexterm>
<indexterm significance="normal"><primary>insecure</primary></indexterm>
Standalone servers can be as secure or as insecure as needs dictate. They can
have simple or complex configurations. Above all, despite the hoopla about
domain security, they remain a common installation.
</para>

<para>
<indexterm significance="normal"><primary>read-only files</primary></indexterm>
<indexterm significance="normal"><primary>share-mode</primary></indexterm>
<indexterm significance="normal"><primary>read-only</primary></indexterm>
<indexterm significance="normal"><primary>standalone server</primary></indexterm>
If all that is needed is a server for read-only files, or for
printers alone, it may not make sense to effect a complex installation.
For example, a drafting office needs to store old drawings and reference
standards. Noone can write files to the server because it is legislatively
important that all documents remain unaltered. A share-mode read-only standalone
server is an ideal solution.
</para>

<para>
<indexterm significance="normal"><primary>simplicity</primary></indexterm>
<indexterm significance="normal"><primary>printers</primary></indexterm>
<indexterm significance="normal"><primary>share-mode server</primary></indexterm>
Another situation that warrants simplicity is an office that has many printers
that are queued off a single central server. Everyone needs to be able to print
to the printers, there is no need to effect any access controls, and no files will
be served from the print server. Again, a share-mode standalone server makes
a great solution.
</para>
</sect1>

<sect1>
<title>Background</title>

<para>
<indexterm significance="normal"><primary>standalone server</primary></indexterm>
<indexterm significance="normal"><primary>local authentication</primary></indexterm>
<indexterm significance="normal"><primary>access control</primary></indexterm>
The term <emphasis>standalone server</emphasis> means that it will provide local authentication and access
control for all resources that are available from it. In general this means that there will be a local user
database. In more technical terms, it means resources on the machine will be made available in either
<emphasis>share</emphasis> mode or in <emphasis>user</emphasis> mode.
</para>

<para>
<indexterm significance="normal"><primary>create user accounts</primary></indexterm>
<indexterm significance="normal"><primary>no network logon service</primary></indexterm>
<indexterm significance="normal"><primary>independent</primary></indexterm>
No special action is needed other than to create user accounts. Standalone
servers do not provide network logon services. This means that machines that
use this server do not perform a domain logon to it. Whatever logon facility
the workstations are subject to is independent of this machine. It is, however,
necessary to accommodate any network user so the logon name he or she uses will
be translated (mapped) locally on the standalone server to a locally known
user name. There are several ways this can be done.
</para>

<para>
<indexterm significance="normal"><primary>local authentication database</primary></indexterm>
<indexterm significance="normal"><primary>SMB</primary></indexterm>
<indexterm significance="normal"><primary>not domain member</primary></indexterm>
Samba tends to blur the distinction a little in defining
a standalone server. This is because the authentication database may be
local or on a remote server, even if from the SMB protocol perspective
the Samba server is not a member of a domain security context.
</para>

<para>
<indexterm significance="normal"><primary>PAM</primary></indexterm>
<indexterm significance="normal"><primary>NSS</primary></indexterm>
<indexterm significance="normal"><primary>UNIX-user database</primary></indexterm>
<indexterm significance="normal"><primary>/etc/passwd</primary></indexterm>
<indexterm significance="normal"><primary>/etc/shadow</primary></indexterm>
<indexterm significance="normal"><primary>local smbpasswd file</primary></indexterm>
<indexterm significance="normal"><primary>LDAP backend</primary></indexterm>
<indexterm significance="normal"><primary>Winbind</primary></indexterm>
Through the use of Pluggable Authentication Modules (PAM) (see <link linkend="pam">the chapter on PAM</link>)
and the name service switcher (NSS), which maintains the UNIX-user database, the source of authentication may
reside on another server. We would be inclined to call this the authentication server.  This means that the
Samba server may use the local UNIX/Linux system password database (<filename moreinfo="none">/etc/passwd</filename> or
<filename moreinfo="none">/etc/shadow</filename>), may use a local smbpasswd file, or may use an LDAP backend, or even via PAM
and Winbind another CIFS/SMB server for authentication.
</para>

</sect1>

<sect1>
<title>Example Configuration</title>

<para>
<indexterm significance="normal"><primary>inspire simplicity</primary></indexterm>
<indexterm significance="normal"><primary>complexity</primary></indexterm>
<link linkend="simplynice">The example Reference Documentation Server</link> and <link linkend="SimplePrintServer">Central Print Serving</link> are designed to inspire simplicity. It is too easy to
attempt a high level of creativity and to introduce too much complexity in server and network design.
</para>

<sect2 id="RefDocServer">
<title>Reference Documentation Server</title>

<para>
<indexterm significance="normal"><primary>read-only</primary></indexterm>
<indexterm significance="normal"><primary>reference documents</primary></indexterm>
<indexterm significance="normal"><primary>/export</primary></indexterm>
<indexterm significance="normal"><primary>/etc/passwd</primary></indexterm>
Configuration of a read-only data server that everyone can access is very simple. By default, all shares are
read-only, unless set otherwise in the <filename moreinfo="none">smb.conf</filename> file. <link linkend="simplynice">The example - Reference
Documentation Server</link> is the <filename moreinfo="none">smb.conf</filename> file that will do this. Assume that all the reference documents
are stored in the directory <filename moreinfo="none">/export</filename>, and the documents are owned by a user other than
nobody. No home directories are shared, and there are no users in the <filename moreinfo="none">/etc/passwd</filename> UNIX
system database. This is a simple system to administer.
</para>

<example id="simplynice">
<title>smb.conf for Reference Documentation Server</title>
<simplelist>
<member>#  Global parameters</member>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>workgroup</indexterm><parameter>workgroup = MIDEARTH</parameter></member>
<member><indexterm>netbios name</indexterm><parameter>netbios name = GANDALF</parameter></member>
<member><indexterm>security</indexterm><parameter>security = SHARE</parameter></member>
<member><indexterm>passdb backend</indexterm><parameter>passdb backend = guest</parameter></member>
<member><indexterm>wins server</indexterm><parameter>wins server = 192.168.1.1</parameter></member>
<member> </member><member><parameter>[data]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = Data</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /export</parameter></member>
<member><indexterm>guest only</indexterm><parameter>guest only = Yes</parameter></member>
</simplelist>
</example>

<blockquote>
<attribution>Mark Twain</attribution>
<para>
I would have spoken more briefly, if I'd had more time to prepare.
</para>
</blockquote>

<para>
<indexterm significance="normal"><primary>password backend</primary></indexterm>
<indexterm significance="normal"><primary>guest</primary></indexterm>
<indexterm significance="normal"><primary>unprivileged account names</primary></indexterm>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
In <link linkend="simplynice">this example</link>, the machine name is set to GANDALF, and the
workgroup is set to the name of the local workgroup (MIDEARTH) so the machine will appear together
with systems with which users are familiar. The only password backend required is the <quote>guest</quote>
backend to allow default unprivileged account names to be used. As there is a WINS server on this network, we
of course make use of it.
</para>

<para>
A US Air Force Colonel was renowned for saying: <quote>Better is the enemy of good enough!</quote> There are often
sound reasons for avoiding complexity as well as for avoiding a technically perfect solution. Unfortunately,
many network administrators still need to learn the art of doing just enough to keep out of trouble.
</para>

</sect2>

<sect2 id="SimplePrintServer">
<title>Central Print Serving</title>

<para>
<indexterm significance="normal"><primary>simple print server</primary></indexterm>
<indexterm significance="normal"><primary>tools</primary></indexterm>
Configuration of a simple print server is easy if you have all the right tools on your system.
</para>

<orderedlist inheritnum="ignore" continuation="restarts">
<title> Assumptions</title>
	<listitem><para>
	The print server must require no administration.
	</para></listitem>

	<listitem><para>
	The print spooling and processing system on our print server will be CUPS.
	(Please refer to <link linkend="CUPS-printing">CUPS Printing Support</link>, for more information).
	</para></listitem>

	<listitem><para>
	The print server will service only network printers. The network administrator
	will correctly configure the CUPS environment to support the printers.
	</para></listitem>

	<listitem><para>
	All workstations will use only PostScript drivers. The printer driver
	of choice is the one shipped with the Windows OS for the Apple Color LaserWriter.
	</para></listitem>
</orderedlist>

<para>
<indexterm significance="normal"><primary>print server</primary></indexterm>
<indexterm significance="normal"><primary>/var/spool/samba</primary></indexterm>
<indexterm significance="normal"><primary>anonymous</primary></indexterm>
In this example our print server will spool all incoming print jobs to
<filename moreinfo="none">/var/spool/samba</filename> until the job is ready to be submitted by
Samba to the CUPS print processor. Since all incoming connections will be as
the anonymous (guest) user, two things will be required to enable anonymous printing.
</para>

<itemizedlist>
<title>Enabling Anonymous Printing</title>
	<listitem><para>
<indexterm significance="normal"><primary>guest account</primary></indexterm>
<indexterm significance="normal"><primary>nobody</primary></indexterm>
<indexterm significance="normal"><primary>testparm</primary></indexterm>
	The UNIX/Linux system must have a <literal>guest</literal> account.
	The default for this is usually the account <literal>nobody</literal>.
	To find the correct name to use for your version of Samba, do the 
	following:
<screen format="linespecific">
<prompt moreinfo="none">$ </prompt><userinput moreinfo="none">testparm -s -v | grep "guest account"</userinput>
</screen>
<indexterm significance="normal"><primary>/etc/passwd</primary></indexterm>
	Make sure that this account exists in your system password
	database (<filename moreinfo="none">/etc/passwd</filename>).
	</para>

	<para>
<indexterm significance="normal"><primary>set a password</primary></indexterm>
<indexterm significance="normal"><primary>lock password</primary></indexterm>
<indexterm significance="normal"><primary>passwd</primary></indexterm>
	It is a good idea either to set a password on this account, or else to lock it
	from UNIX use. Assuming that the guest account is called <literal moreinfo="none">pcguest</literal>,
	it can be locked by executing:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> passwd -l pcguest
</screen>
	The exact command may vary depending on your UNIX/Linux distribution.
	</para></listitem>

	<listitem><para>
<indexterm significance="normal"><primary>directory</primary></indexterm>
<indexterm significance="normal"><primary>guest account</primary></indexterm>
<indexterm significance="normal"><primary>available</primary></indexterm>
<indexterm significance="normal"><primary>mkdir</primary></indexterm>
<indexterm significance="normal"><primary>chown</primary></indexterm>
<indexterm significance="normal"><primary>chmod</primary></indexterm>
	The directory into which Samba will spool the file must have write
	access for the guest account. The following commands will ensure that
	this directory is available for use:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">mkdir /var/spool/samba</userinput>
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">chown nobody.nobody /var/spool/samba</userinput>
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">chmod a+rwt /var/spool/samba</userinput>
</screen>
	</para></listitem>
</itemizedlist>

<para>
The contents of the <filename moreinfo="none">smb.conf</filename> file is shown in <link linkend="AnonPtrSvr">the Anonymous Printing example</link>.
</para>

<example id="AnonPtrSvr">
<title><filename moreinfo="none">smb.conf</filename> for Anonymous Printing</title>
<simplelist>
<member>#  Global parameters</member>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>workgroup</indexterm><parameter>workgroup = MIDEARTH</parameter></member>
<member><indexterm>netbios name</indexterm><parameter>netbios name = GANDALF</parameter></member>
<member><indexterm>security</indexterm><parameter>security = SHARE</parameter></member>
<member><indexterm>passdb backend</indexterm><parameter>passdb backend = guest</parameter></member>
<member><indexterm>printing</indexterm><parameter>printing = cups</parameter></member>
<member><indexterm>printcap name</indexterm><parameter>printcap name = cups</parameter></member>

<member> </member><member><parameter>[printers]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = All Printers</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /var/spool/samba</parameter></member>
<member><indexterm>printer admin</indexterm><parameter>printer admin = root</parameter></member>
<member><indexterm>guest ok</indexterm><parameter>guest ok = Yes</parameter></member>
<member><indexterm>printable</indexterm><parameter>printable = Yes</parameter></member>
<member><indexterm>use client driver</indexterm><parameter>use client driver = Yes</parameter></member>
<member><indexterm>browseable</indexterm><parameter>browseable = No</parameter></member>
</simplelist>
</example>


<note><para>
<indexterm significance="normal"><primary>MIME</primary><secondary>raw</secondary></indexterm>
<indexterm significance="normal"><primary>raw printing</primary></indexterm>
<indexterm significance="normal"><primary>/etc/mime.conv</primary></indexterm>
<indexterm significance="normal"><primary>/etc/mime.types</primary></indexterm>
<indexterm significance="normal"><primary>CUPS print filters</primary></indexterm>
On CUPS-enabled systems there is a facility to pass raw data directly to the printer without intermediate
processing via CUPS print filters. Where use of this mode of operation is desired, it is necessary to
configure a raw printing device. It is also necessary to enable the raw mime handler in the
<filename moreinfo="none">/etc/mime.conv</filename> and <filename moreinfo="none">/etc/mime.types</filename> files. Refer to <link linkend="CUPS-printing">CUPS Printing Support</link>, <link linkend="cups-raw">Explicitly Enable raw Printing
for application/octet-stream</link>.
</para></note>

<para>
<indexterm significance="normal"><primary>CUPS libarary API</primary></indexterm>
<indexterm significance="normal"><primary>no printcap file</primary></indexterm>
<indexterm significance="normal"><primary>PDF filter</primary></indexterm>
<indexterm significance="normal"><primary>printcap name</primary></indexterm>
The example in <link linkend="AnonPtrSvr">the Anonymous Printing example</link> uses CUPS for direct printing
via the CUPS libarary API. This means that all printers will be exposed to Windows users without need to
configure a printcap file. If there is necessity to expose only a sub-set of printers, or to define a special
type of printer (for example, a PDF filter) the <parameter moreinfo="none">printcap name = cups</parameter> can be replaced
with the entry <parameter moreinfo="none">printcap name = /etc/samba/myprintcap</parameter>. In this case the file specified
should contain a list of the printer names that should be exposed to Windows network users.
</para>

</sect2>

</sect1>

<sect1>
<title>Common Errors</title>

<para>
<indexterm significance="normal"><primary>greatest mistake</primary></indexterm>
<indexterm significance="normal"><primary>configuration too complex</primary></indexterm>
The greatest mistake so often made is to make a network configuration too complex.
It pays to use the simplest solution that will meet the needs of the moment.
</para>

</sect1>
</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="ClientConfig">
<chapterinfo>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
</chapterinfo>

<title>MS Windows Network Configuration Guide</title>

<sect1>
<title>Features and Benefits</title>

<para>
<indexterm significance="normal"><primary>network difficulty</primary></indexterm>
<indexterm significance="normal"><primary>network client</primary></indexterm>
<indexterm significance="normal"><primary>client client instructions</primary></indexterm>
Occasionally network administrators report difficulty getting Microsoft Windows clients to interoperate
correctly with Samba servers. It seems that some folks just cannot accept the fact that the right way
to configure an MS Windows network client is precisely as one would do when using MS Windows NT4 or 200x
servers. Yet there is repetitious need to provide detailed Windows client configuration instructions.
</para>

<para>
<indexterm significance="normal"><primary>graphically illustrated client configuration</primary></indexterm>
<indexterm significance="normal"><primary>critical aspects of configuration</primary></indexterm>
The purpose of this chapter is to graphically illustrate MS Windows client configuration for the most common
critical aspects of such configuration. An experienced network administrator will not be interested in the
details of this chapter.
</para>

</sect1>

<sect1>
<title>Technical Details</title>

<para>
<indexterm significance="normal"><primary>TCP/IP protocol configuration</primary></indexterm>
<indexterm significance="normal"><primary>network membership</primary></indexterm>
This chapter discusses TCP/IP protocol configuration as well as network membership for the platforms
that are in common use today. These are:
</para>

<itemizedlist>
	<listitem><para>
	Microsoft Windows XP Professional
	</para></listitem>
	<listitem><para>
	Windows 2000 Professional
	</para></listitem>
	<listitem><para>
	Windows Millennium edition (Me)
	</para></listitem>
</itemizedlist>

	<sect2>
	<title>TCP/IP Configuration</title>

	<para>
<indexterm significance="normal"><primary>network configuration problems</primary></indexterm>
<indexterm significance="normal"><primary>plague network users</primary></indexterm>
	The builder of a house must ensure that all construction takes place on a firm foundation.
	The same is true for the builder of a TCP/IP-based networking system. Fundamental network configuration problems
	will plague all network users until they are resolved. 
	</para>

	<para>
<indexterm significance="normal"><primary>fixed IP addresses</primary></indexterm>
<indexterm significance="normal"><primary>DHCP</primary></indexterm>
	MS Windows workstations and servers can be configured either with fixed
	IP addresses or via DHCP. The examples that follow demonstrate the use of DHCP
	and make only passing reference to those situations where fixed IP configuration
	settings can be effected.
	</para>

	<para>
<indexterm significance="normal"><primary>shortcuts</primary></indexterm>
<indexterm significance="normal"><primary>abbreviated keystrokes</primary></indexterm>
	It is possible to use shortcuts or abbreviated keystrokes to arrive at a
	particular configuration screen. The decision was made to base all examples in this
	chapter on use of the <guibutton moreinfo="none">Start</guibutton> button.
	</para>

		<sect3>
		<title>MS Windows XP Professional</title>

		<para>
<indexterm significance="normal"><primary>Windows XP TCP/IP</primary></indexterm>
		There are two paths to the Windows XP TCP/IP configuration panel. Choose the access method that you prefer:
		</para>

		<para>
		Click <guimenu moreinfo="none">Start -&gt; Control Panel -&gt; Network Connections</guimenu>.
		</para>

		<para>
		<emphasis>Alternately,</emphasis> click <guimenu moreinfo="none">Start -&gt;</guimenu>, and right-click <guimenu moreinfo="none">My Network Places</guimenu>
		then select <guimenuitem moreinfo="none">Properties</guimenuitem>.
		</para>

		<para>
<indexterm significance="normal"><primary>Windows XP Professional</primary></indexterm>
		The following procedure steps through the Windows XP Professional TCP/IP configuration process:
		</para>

		<procedure>
			<step performance="required"><para>
<indexterm significance="normal"><primary>Local Area Connection</primary></indexterm>
<indexterm significance="normal"><primary>Network Bridge</primary></indexterm>
<indexterm significance="normal"><primary>interface</primary></indexterm>
			On some installations the interface will be called <guimenu moreinfo="none">Local Area Connection</guimenu> and
			on others it will be called <guimenu moreinfo="none">Network Bridge</guimenu>. On our system it is called <guimenu moreinfo="none">Network Bridge</guimenu>.
			Right-click on <guimenu moreinfo="none">Network Bridge -&gt; Properties</guimenu>. See <link linkend="WXPP002"/>.
			<figure id="WXPP002" float="0"><title>Network Bridge Configuration.</title><mediaobject><imageobject role="html"><imagedata fileref="images/WXPP002.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/WXPP002.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/WXPP002"/></imageobject></mediaobject></figure>
			</para>
			</step>

			<step performance="required"><para>
<indexterm significance="normal"><primary>TCP/IP protocol settings</primary></indexterm>
<indexterm significance="normal"><primary>Network Bridge Configuration</primary></indexterm>
			The Network Bridge Configuration, or Local Area Connection, panel is used to set TCP/IP protocol settings.
			In <guimenuitem moreinfo="none">This connection uses the following items:</guimenuitem> box,
			click on <guimenu moreinfo="none">Internet Protocol (TCP/IP)</guimenu>, then click on <guibutton moreinfo="none">Properties</guibutton>.
			</para>

			<para>
<indexterm significance="normal"><primary>DHCP-enabled operation</primary></indexterm>
<indexterm significance="normal"><primary>IP address automatically</primary></indexterm>
			The default setting is DHCP-enabled operation
			(i.e., <quote>Obtain an IP address automatically</quote>). See <link linkend="WXPP003"/>.
			<figure id="WXPP003" float="0">
				<title>Internet Protocol (TCP/IP) Properties.</title>
				<mediaobject><imageobject role="html"><imagedata fileref="images/WXPP003.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/WXPP003.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/WXPP003"/></imageobject></mediaobject>
			</figure>
			</para>

			<para>
<indexterm significance="normal"><primary>DHCP</primary></indexterm>
<indexterm significance="normal"><primary>TCP/IP</primary></indexterm>
<indexterm significance="normal"><primary>DNS</primary></indexterm>
<indexterm significance="normal"><primary>ISC DHCP server</primary></indexterm>
			Many network administrators will want to use DHCP to configure all client TCP/IP
			protocol stack settings. (For information on how to configure the ISC DHCP server
			for Windows client support see <link linkend="DHCP">the DNS and DHCP Configuration Guide</link>,
			<link linkend="DHCP">DHCP Server</link>.
			</para>

			<para>
<indexterm significance="normal"><primary>fixed IP address</primary></indexterm>
<indexterm significance="normal"><primary>subnet mask</primary></indexterm>
<indexterm significance="normal"><primary>gateway address</primary></indexterm>
			If it is necessary to provide a fixed IP address, click on <quote>Use the following IP address</quote> and enter the
			IP Address, the subnet mask, and the default gateway address in the boxes provided.
			</para></step>

			<step performance="required"><para>
<indexterm significance="normal"><primary>Advanced TCP/IP configuration</primary></indexterm>
<indexterm significance="normal"><primary>TCP/IP configuration</primary></indexterm>
<indexterm significance="normal"><primary>IP aliases</primary></indexterm>
<indexterm significance="normal"><primary>default gateways</primary></indexterm>
			Click the <guibutton moreinfo="none">Advanced</guibutton> button to proceed with TCP/IP configuration.
			This opens a panel in which it is possible to create additional IP addresses for this interface.
			The technical name for the additional addresses is <emphasis>IP aliases</emphasis>, and additionally this
			panel permits the setting of more default gateways (routers). In most cases where DHCP is used, it will not be
			necessary to create additional settings. See <link linkend="WXPP005"/> to see the appearance of this panel.
			<figure id="WXPP005" float="0"><title>Advanced Network Settings</title><mediaobject><imageobject role="html"><imagedata fileref="images/WXPP005.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/WXPP005.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/WXPP005"/></imageobject></mediaobject></figure>
			</para>

			<para>
<indexterm significance="normal"><primary>DNS</primary></indexterm>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
<indexterm significance="normal"><primary>DHCP</primary></indexterm>
			Fixed settings may be required for DNS and WINS if these settings are not provided automatically via DHCP.
			</para></step>

			<step performance="required"><para>
<indexterm significance="normal"><primary>DNS server settings</primary></indexterm>
<indexterm significance="normal"><primary>manually configured DNS settings</primary></indexterm>
			Click the <guimenu moreinfo="none">DNS</guimenu> tab to add DNS server settings. 
			The example system uses manually configured DNS settings. When finished making changes, click the
			<guibutton moreinfo="none">OK</guibutton> to commit the settings. See <link linkend="WXPP014"/>.
			<figure id="WXPP014" float="0"> <title>DNS Configuration.</title> <mediaobject><imageobject role="html"><imagedata fileref="images/WXPP014.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/WXPP014.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/WXPP014"/></imageobject></mediaobject> </figure>
			</para></step>

			<step performance="required"><para>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
<indexterm significance="normal"><primary>manual WINS server entries</primary></indexterm>
			Click the <guibutton moreinfo="none">WINS</guibutton> tab to add manual WINS server entries. 
			This step demonstrates an example system that uses manually configured WINS settings. 
			When finished making changes, click <guibutton moreinfo="none">OK</guibutton> to commit
                        the settings. See <link linkend="WXPP009"/>.
			<figure id="WXPP009" float="0"><title>WINS Configuration</title><mediaobject><imageobject role="html"><imagedata fileref="images/WXPP009.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/WXPP009.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/WXPP009"/></imageobject></mediaobject></figure>
			</para></step>
		</procedure>

		</sect3>

		<sect3>
		<title>MS Windows 2000</title>

		<para>
<indexterm significance="normal"><primary>Windows 2000 Professional TCP/IP</primary></indexterm>
<indexterm significance="normal"><primary>TCP/IP configuration panel</primary></indexterm>
		There are two paths to the Windows 2000 Professional TCP/IP configuration panel. Choose the access method that you prefer:
		</para>

		<para>
		Click <guimenu moreinfo="none">Start -&gt; Control Panel -&gt; Network and Dial-up Connections</guimenu>.
		</para>

		<para>
		<emphasis>Alternatively,</emphasis> click <guimenu moreinfo="none">Start</guimenu>, then right-click <guimenu moreinfo="none">My Network Places</guimenu>, and
		select <guimenuitem moreinfo="none">Properties</guimenuitem>.
		</para>

		<para>
<indexterm significance="normal"><primary>Windows XP Professional TCP/IP</primary></indexterm>
		The following procedure steps through the Windows XP Professional TCP/IP configuration process:
		</para>

		<procedure>
			<step performance="required"><para>
			Right-click on <guimenu moreinfo="none">Local Area Connection</guimenu>, then click
			<guimenuitem moreinfo="none">Properties</guimenuitem>. See <link linkend="w2kp001"/>.
			<figure id="w2kp001" float="0"><title>Local Area Connection Properties.</title><mediaobject><imageobject role="html"><imagedata fileref="images/w2kp001.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/w2kp001.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/w2kp001"/></imageobject></mediaobject></figure>
			</para></step>

			<step performance="required"><para>
<indexterm significance="normal"><primary>Local Area Connection Properties</primary></indexterm>
<indexterm significance="normal"><primary>TCP/IP protocol settings</primary></indexterm>
			The Local Area Connection Properties is used to set TCP/IP protocol settings. Click on
			<guimenu moreinfo="none">Internet Protocol (TCP/IP)</guimenu> in the <guimenuitem moreinfo="none">Components checked are used by this
			connection:</guimenuitem> box, then click the <guibutton moreinfo="none">Properties</guibutton> button.
			</para></step>

			<step performance="required"><para>
<indexterm significance="normal"><primary>DHCP-enabled</primary></indexterm>
<indexterm significance="normal"><primary>IP address automatically</primary></indexterm>
			The default setting is DHCP-enabled operation
			(i.e., <quote>Obtain an IP address automatically</quote>). See <link linkend="w2kp002"/>.
			<figure id="w2kp002" float="0"><title>Internet Protocol (TCP/IP) Properties.</title><mediaobject><imageobject role="html"><imagedata fileref="images/w2kp002.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/w2kp002.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/w2kp002"/></imageobject></mediaobject></figure>
			</para>

			<para>
<indexterm significance="normal"><primary>DHCP</primary></indexterm>
<indexterm significance="normal"><primary>protocol stack settings</primary></indexterm>
			Many network administrators will want to use DHCP to configure all client TCP/IP
			protocol stack settings. (For information on how to configure the ISC DHCP server
			for Windows client support, see, <link linkend="DHCP"/>.
			</para>

			<para>
<indexterm significance="normal"><primary>fixed IP address</primary></indexterm>
<indexterm significance="normal"><primary>network clients</primary></indexterm>
			If it is necessary to provide a fixed IP address, click on <quote>Use the following IP address</quote> and enter the
			IP Address, the subnet mask, and the default gateway address in the boxes provided.
			For this example we are assuming that all network clients will be configured using DHCP.
			</para></step>

			<step performance="required"><para>
			Click the <guimenu moreinfo="none">Advanced</guimenu> button to proceed with TCP/IP configuration.
			Refer to <link linkend="w2kp003"/>.
			<figure id="w2kp003" float="0"><title>Advanced Network Settings.</title><mediaobject><imageobject role="html"><imagedata fileref="images/w2kp003.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/w2kp003.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/w2kp003"/></imageobject></mediaobject></figure>
			</para>

			<para>
<indexterm significance="normal"><primary>DNS</primary></indexterm>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
<indexterm significance="normal"><primary>DHCP</primary></indexterm>
			Fixed settings may be required for DNS and WINS if these settings are not provided automatically via DHCP.
			</para></step>

			<step performance="required"><para>
<indexterm significance="normal"><primary>DNS server settings</primary></indexterm>
<indexterm significance="normal"><primary>commit the settings</primary></indexterm>
			Click the <guimenu moreinfo="none">DNS</guimenu> tab to add DNS server settings. 
			The example system uses manually configured DNS settings. When finished making changes,
			click <guibutton moreinfo="none">OK</guibutton> to commit the settings. See <link linkend="w2kp004"/>.
			<figure id="w2kp004" float="0"><title>DNS Configuration.</title><mediaobject><imageobject role="html"><imagedata fileref="images/w2kp004.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/w2kp004.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/w2kp004"/></imageobject></mediaobject></figure>
			</para></step>

			<step performance="required"><para>
<indexterm significance="normal"><primary>manual WINS server entries</primary></indexterm>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
			Click the <guibutton moreinfo="none">WINS</guibutton> tab to add manual WINS server entries. 
			This step demonstrates an example system that uses manually configured WINS settings. 
			When finished making changes, click <guibutton moreinfo="none">OK</guibutton> to commit the settings.
			See <link linkend="w2kp005"/>.
			<figure id="w2kp005" float="0">
				<title>WINS Configuration.</title><mediaobject><imageobject role="html"><imagedata fileref="images/w2kp005.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/w2kp005.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/w2kp005"/></imageobject></mediaobject>
			</figure>
			</para></step>

		</procedure>

		</sect3>

		<sect3>
		<title>MS Windows Me</title>

		<para>
<indexterm significance="normal"><primary>Windows Millennium edition (Me) TCP/IP</primary></indexterm>
<indexterm significance="normal"><primary>Windows Millennium</primary></indexterm>
<indexterm significance="normal"><primary>TCP/IP configuration</primary></indexterm>
		There are two paths to the Windows Millennium edition (Me) TCP/IP configuration panel. Choose the access method that you prefer:
		</para>

		<para>
		Click <guimenu moreinfo="none">Start -&gt; Control Panel -&gt; Network Connections</guimenu>.
		</para>

		<para>
<indexterm significance="normal"><primary>My Network Places</primary></indexterm>
<indexterm significance="normal"><primary>Properties</primary></indexterm>
		<emphasis>Alternatively,</emphasis> click on <guimenu moreinfo="none">Start -&gt;</guimenu>, and right click on <guimenu moreinfo="none">My Network Places</guimenu>
		then select <guimenuitem moreinfo="none">Properties</guimenuitem>.
		</para>

		<para>
<indexterm significance="normal"><primary>Windows Me TCP/IP</primary></indexterm>
		The following procedure steps through the Windows Me TCP/IP configuration process:
		</para>

		<procedure>
			<step performance="required"><para>
<indexterm significance="normal"><primary>Internet Protocol TCP/IP</primary></indexterm>
			In the box labeled <guimenuitem moreinfo="none">The following network components are installed:</guimenuitem>,
			click on <guimenu moreinfo="none">Internet Protocol TCP/IP</guimenu>, then click on the <guibutton moreinfo="none">Properties</guibutton> button.
			See <link linkend="WME001"/>.
			<figure id="WME001" float="0">
				<title>The Windows Me Network Configuration Panel.</title>
				<mediaobject><imageobject role="html"><imagedata fileref="images/WME001.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/WME001.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/WME001"/></imageobject></mediaobject>
			</figure>
			</para></step>

			<step performance="required"><para>
<indexterm significance="normal"><primary>DHCP</primary></indexterm>
<indexterm significance="normal"><primary>TCP/IP</primary></indexterm>
<indexterm significance="normal"><primary>ISC DHCP server</primary></indexterm>
			Many network administrators will want to use DHCP to configure all client TCP/IP
			protocol stack settings. (For information on how to configure the ISC DHCP server
			for Windows client support see <link linkend="DHCP">the DNS and DHCP Configuration Guide</link>,
			<link linkend="DHCP">DHCP Server</link>. The default setting on Windows Me workstations is for DHCP-enabled operation
			(i.e., <guimenu moreinfo="none">Obtain IP address automatically</guimenu> is enabled). See <link linkend="WME002"/>.
			<figure id="WME002" float="0"><title>IP Address.</title><mediaobject><imageobject role="html"><imagedata fileref="images/WME002.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/WME002.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/WME002"/></imageobject></mediaobject></figure>
			</para>

			<para>
<indexterm significance="normal"><primary>Specify an IP address</primary></indexterm>
<indexterm significance="normal"><primary>subnet mask</primary></indexterm>
<indexterm significance="normal"><primary>DHCP</primary></indexterm>
			If it is necessary to provide a fixed IP address, click on <guimenuitem moreinfo="none">Specify an IP address</guimenuitem> and enter the
			IP Address and the subnet mask in the boxes provided. For this example we are assuming that all
			network clients will be configured using DHCP.
			</para></step>

			<step performance="required"><para>
<indexterm significance="normal"><primary>DNS</primary></indexterm>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
			Fixed settings may be required for DNS and WINS if these settings are not provided automatically via DHCP.
			</para></step>

			<step performance="required"><para>
<indexterm significance="normal"><primary>WINS server settings</primary></indexterm>
			If necessary, click the <guimenu moreinfo="none">DNS Configuration</guimenu> tab to add DNS server settings. 
			Click the <guibutton moreinfo="none">WINS Configuration</guibutton> tab to add WINS server settings. 
			The <guimenu moreinfo="none">Gateway</guimenu> tab allows additional gateways (router addresses) to be added to the network
			interface settings. In most cases where DHCP is used, it will not be necessary to
			create these manual settings.
			</para></step>

			<step performance="required"><para>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
<indexterm significance="normal"><primary>manually configured</primary></indexterm>
			The following example uses manually configured WINS settings. See <link linkend="WME005"/>.
			When finished making changes, click <guibutton moreinfo="none">OK</guibutton> to commit the settings.
			<figure id="WME005" float="0"><title>DNS Configuration.</title><mediaobject><imageobject role="html"><imagedata fileref="images/WME005.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/WME005.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/WME005"/></imageobject></mediaobject></figure>
			</para>

			<para>
<indexterm significance="normal"><primary>single DHCP server</primary></indexterm>
<indexterm significance="normal"><primary>multiple Windows workgroups or domains</primary></indexterm>
			This is an example of a system that uses manually configured WINS settings. One situation where
			this might apply is on a network that has a single DHCP server that provides settings for multiple
			Windows workgroups or domains. See <link linkend="WME003"/>.
			<figure id="WME003" float="0"><title>WINS Configuration.</title><mediaobject><imageobject role="html"><imagedata fileref="images/WME003.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/WME003.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/WME003"/></imageobject></mediaobject></figure>
			</para></step>
		</procedure>


		</sect3>

	</sect2>

	<sect2>
	<title>Joining a Domain: Windows 2000/XP Professional</title>

	<para>
<indexterm significance="normal"><primary>Windows NT/200x/XP Professional</primary></indexterm>
<indexterm significance="normal"><primary>domain security</primary></indexterm>
<indexterm significance="normal"><primary>domain member</primary></indexterm>
<indexterm significance="normal"><primary>domain joining</primary></indexterm>
	Microsoft Windows NT/200x/XP Professional platforms can participate in domain security.
	This section steps through the process for making a Windows 200x/XP Professional machine a
	member of a domain security environment. It should be noted that this process is identical
	when joining a domain that is controlled by Windows NT4/200x as well as a Samba PDC.
	</para>

	<procedure>
		<step performance="required"><para>
		Click <guimenu moreinfo="none">Start</guimenu>.
		</para></step>

		<step performance="required"><para>
		Right-click <guimenu moreinfo="none">My Computer</guimenu>, then select <guimenuitem moreinfo="none">Properties</guimenuitem>.
		</para></step>

		<step performance="required"><para>
<indexterm significance="normal"><primary>Control Panel</primary></indexterm>
		The opening panel is the same one that can be reached by clicking <guimenu moreinfo="none">System</guimenu> on the Control Panel.
		See <link linkend="wxpp001"/>.
		<figure id="wxpp001" float="0"><title>The General Panel.</title><mediaobject><imageobject role="html"><imagedata fileref="images/wxpp001.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/wxpp001.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/wxpp001"/></imageobject></mediaobject></figure>
		</para></step>

		<step performance="required"><para>
<indexterm significance="normal"><primary>Computer Name</primary></indexterm>
		Click the <guimenu moreinfo="none">Computer Name</guimenu> tab.
		This panel shows the <guimenuitem moreinfo="none">Computer Description</guimenuitem>, the <guimenuitem moreinfo="none">Full computer name</guimenuitem>,
		and the <guimenuitem moreinfo="none">Workgroup</guimenuitem> or <guimenuitem moreinfo="none">Domain name</guimenuitem>.
		</para>

		<para>
<indexterm significance="normal"><primary>Network ID</primary></indexterm>
<indexterm significance="normal"><primary>configuration wizard</primary></indexterm>
		Clicking the <guimenu moreinfo="none">Network ID</guimenu> button will launch the configuration wizard. Do not use this with
		Samba-3. If you wish to change the computer name or join or leave the domain, click the <guimenu moreinfo="none">Change</guimenu> button.
		See <link linkend="wxpp004"/>.
		<figure id="wxpp004" float="0"><title>The Computer Name Panel.</title><mediaobject><imageobject role="html"><imagedata fileref="images/wxpp004.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/wxpp004.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/wxpp004"/></imageobject></mediaobject></figure>
		</para></step>

		<step performance="required"><para>
		Click on <guimenu moreinfo="none">Change</guimenu>. This panel shows that our example machine (TEMPTATION) is in a workgroup called WORKGROUP.
		We will join the domain called MIDEARTH. See <link linkend="wxpp006"/>.
		<figure id="wxpp006" float="0"><title>The Computer Name Changes Panel.</title><mediaobject><imageobject role="html"><imagedata fileref="images/wxpp006.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/wxpp006.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/wxpp006"/></imageobject></mediaobject></figure>
        </para></step>

		<step performance="required"><para>
<indexterm significance="normal"><primary>domain radio button</primary></indexterm>
		Enter the name <guimenu moreinfo="none">MIDEARTH</guimenu> in the field below the domain radio button.
		</para>

        <para>
		This panel shows that our example machine (TEMPTATION) is set to join the domain called MIDEARTH. See <link linkend="wxpp007"/>.
		<figure id="wxpp007" float="0"><title>The Computer Name Changes Panel <?latex --- ?> Domain MIDEARTH.</title><mediaobject><imageobject role="html"><imagedata fileref="images/wxpp007.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/wxpp007.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/wxpp007"/></imageobject></mediaobject></figure>
        </para></step>

		<step performance="required"><para>
<indexterm significance="normal"><primary>credentials</primary></indexterm>
<indexterm significance="normal"><primary>username and password</primary></indexterm>
		Now click the <guimenu moreinfo="none">OK</guimenu> button. A dialog box should appear to allow you to provide the
		credentials (username and password) of a domain administrative account that has the rights to add machines to
		the domain.
		</para>

        <para>
<indexterm significance="normal"><primary>root</primary></indexterm>
		Enter the name <quote>root</quote> and the root password from your Samba-3 server. See <link linkend="wxpp008"/>.
		<figure id="wxpp008" float="0">
			<title>Computer Name Changes <?latex --- ?> Username and Password Panel.</title><mediaobject><imageobject role="html"><imagedata fileref="images/wxpp008.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/wxpp008.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/wxpp008"/></imageobject></mediaobject>
		</figure>
        </para></step>

		<step performance="required"><para>
		Click on <guimenu moreinfo="none">OK</guimenu>.
        </para>

		<para>
<indexterm significance="normal"><primary>Welcome</primary></indexterm>
<indexterm significance="normal"><primary>rebooted</primary></indexterm>
		The <quote>Welcome to the MIDEARTH domain.</quote> dialog box should appear. At this point the machine must be rebooted.
		Joining the domain is now complete.
        </para></step>

	</procedure>

	</sect2>

	<sect2>
	<title>Domain Logon Configuration: Windows 9x/Me</title>

	<para>
<indexterm significance="normal"><primary>Windows 9x/Me</primary></indexterm>
<indexterm significance="normal"><primary>domain logon</primary></indexterm>
<indexterm significance="normal"><primary>LanManager</primary></indexterm>
	We follow the convention used by most in saying that Windows 9x/Me machines can participate in domain logons. The truth is
	that these platforms can use only the LanManager network logon protocols.
	</para>

	<note><para>
<indexterm significance="normal"><primary>Windows XP Home edition</primary></indexterm>
<indexterm significance="normal"><primary>LanManager</primary></indexterm>
<indexterm significance="normal"><primary>network logon</primary></indexterm>
	Windows XP Home edition cannot participate in domain or LanManager network logons.
	</para></note>

	<procedure>
		<step performance="required"><para>
		Right-click on the <guimenu moreinfo="none">Network Neighborhood</guimenu> icon.
		</para></step>

		<step performance="required"><para>
		The Network Configuration Panel allows all common network settings to be changed.
                See <link linkend="WME009"/>.
		<figure id="WME009" float="0"><title>The Network Panel.</title><mediaobject><imageobject role="html"><imagedata fileref="images/WME009.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/WME009.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/WME009"/></imageobject></mediaobject></figure>
                </para>

		<para>
<indexterm significance="normal"><primary>Client for Microsoft Networks</primary></indexterm>
<indexterm significance="normal"><primary>Properties</primary></indexterm>
		Make sure that the <guimenu moreinfo="none">Client for Microsoft Networks</guimenu> driver is installed as shown.
		Click on the <guimenu moreinfo="none">Client for Microsoft Networks</guimenu> entry in <guimenu moreinfo="none">The following network
		components are installed:</guimenu> box. Then click the <guibutton moreinfo="none">Properties</guibutton> button.
		</para></step>

		<step performance="required"><para>
<indexterm significance="normal"><primary>Networks Properties</primary></indexterm>
<indexterm significance="normal"><primary>network logon</primary></indexterm>
		The Client for Microsoft Networks Properties panel is the correct location to configure network logon
		settings. See <link linkend="WME010"/>.
		<figure id="WME010" float="0"><title>Client for Microsoft Networks Properties Panel.</title><mediaobject><imageobject role="html"><imagedata fileref="images/WME010.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/WME010.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/WME010"/></imageobject></mediaobject></figure>
                </para>

		<para>
<indexterm significance="normal"><primary>Windows NT domain name</primary></indexterm>
<indexterm significance="normal"><primary>domain name</primary></indexterm>
		Enter the Windows NT domain name, check the <guimenu moreinfo="none">Log on to Windows NT domain</guimenu> box,
		and click <guimenu moreinfo="none">OK</guimenu>.
		</para></step>

		<step performance="required"><para>
<indexterm significance="normal"><primary>Identification</primary></indexterm>
<indexterm significance="normal"><primary>workgroup</primary></indexterm>
<indexterm significance="normal"><primary>computer name</primary></indexterm>
		Click on the <guimenu moreinfo="none">Identification</guimenu> button. This is the location at which the workgroup
		(domain) name and the machine name (computer name) need to be set. See <link linkend="WME013"/>.
		<figure id="WME013" float="0"><title>Identification Panel.</title><mediaobject><imageobject role="html"><imagedata fileref="images/WME013.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/WME013.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/WME013"/></imageobject></mediaobject></figure>
                </para></step>

		<step performance="required"><para>
<indexterm significance="normal"><primary>Access Control</primary></indexterm>
<indexterm significance="normal"><primary>group accounts</primary></indexterm>
<indexterm significance="normal"><primary>domain user</primary></indexterm>
<indexterm significance="normal"><primary>User-level access control</primary></indexterm>
		Now click the <guimenu moreinfo="none">Access Control</guimenu> button. If you want to be able to assign share access
		permissions using domain user and group accounts, it is necessary to enable
		<guimenu moreinfo="none">User-level access control</guimenu> as shown in this panel.  See <link linkend="WME014"/>.
		<figure id="WME014" float="0"><title>Access Control Panel.</title><mediaobject><imageobject role="html"><imagedata fileref="images/WME014.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/WME014.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/WME014"/></imageobject></mediaobject></figure>
                </para></step>

	</procedure>

	</sect2>

</sect1>

<sect1>
<title>Common Errors</title>

<para>
<indexterm significance="normal"><primary>networking systems</primary></indexterm>
<indexterm significance="normal"><primary>errors that can afflict</primary></indexterm>
The most common errors that can afflict Windows networking systems include:
</para>

<itemizedlist>
	<listitem><para>Incorrect IP address.</para></listitem>
	<listitem><para>Incorrect or inconsistent netmasks.</para></listitem>
	<listitem><para>Incorrect router address.</para></listitem>
	<listitem><para>Incorrect DNS server address.</para></listitem>
	<listitem><para>Incorrect WINS server address.</para></listitem>
	<listitem><para>Use of a Network Scope setting <?latex --- ?> watch out for this one!</para></listitem>
</itemizedlist>

<para>
<indexterm significance="normal"><primary>Windows NT/200x/XP Professional</primary></indexterm>
<indexterm significance="normal"><primary>cannot join domain</primary></indexterm>
The most common reasons for which a Windows NT/200x/XP Professional client cannot join the Samba controlled domain are:
</para>

<itemizedlist>
	<listitem><para><filename moreinfo="none">smb.conf</filename> does not have correct <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="ADDMACHINESCRIPT" xlink:href="smb.conf.5.html#ADDMACHINESCRIPT">add machine script</link> settings.</para></listitem>
	<listitem><para><quote>root</quote> account is not in password backend database.</para></listitem>
	<listitem><para>Attempt to use a user account instead of the <quote>root</quote> account to join a machine to the domain.</para></listitem>
	<listitem><para>Open connections from the workstation to the server.</para></listitem>
	<listitem><para>Firewall or filter configurations in place on either the client or the Samba server.</para></listitem>
</itemizedlist>

</sect1>

</chapter>
	<?latex \cleardoublepage ?>

</part>

<part id="optional">
<title>Advanced Configuration</title>
<partintro>
<title>Valuable Nuts and Bolts Information</title>

<para>
Samba has several features that you might want or might not want to use.
The chapters in this part each cover specific Samba features.
</para>

</partintro>

	<?latex \cleardoublepage ?>
	<chapter id="ChangeNotes">
<chapterinfo>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
	<author>
<firstname>Gerald</firstname><surname>Carter</surname><othername>(Jerry)</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jerry@samba.org</email></address>
</affiliation></author>
</chapterinfo>

<title>Important and Critical Change Notes for the Samba 3.x Series</title>
<para>
Please read this chapter carefully before update or upgrading Samba.  You should expect to find only critical
or very important information here. Comprehensive change notes and guidance information can be found in the
section <link linkend="upgrading-to-3.0">Updating and Upgrading Samba</link>.
</para>

<sect1>

<title>Important Samba-3.2.x Change Notes</title>
<para>
!!!!!!!!!!!!Add all critical update notes here!!!!!!!!!!!!!
</para>

</sect1>

<sect1>

<title>Important Samba-3.0.x Change Notes</title>
<para>
These following notes pertain in particular to Samba 3.0.23 through Samba 3.0.25c (or more recent 3.0.25
update).  Samba is a fluid and ever changing project. Changes throughout the 3.0.x series release are
documented in this documention - See <link linkend="oldupdatenotes">Upgrading from Samba-2.x to Samba-3.0.25</link>.
</para>

<para>
Sometimes it is difficult to figure out which part, or parts, of the HOWTO documentation should be updated to
reflect the impact of new or modified features. At other times it becomes clear that the documentation is in
need of being restructured.
</para>

<para>
In recent times a group of Samba users has joined the thrust to create a new <ulink url="http://wiki.samba.org/">Samba Wiki</ulink> that is slated to become the all-singing and all-dancing
new face of Samba documentation. Hopefully, the Wiki will benefit from greater community input and
thus may be kept more up to date. Until that golden dream materializes and matures it is necessary to
continue to maintain the HOWTO. This chapter will document major departures from earlier behavior until
such time as the body of this HOWTO is restructured or modified.
</para>

<para>
This chapter is new to the release of the HOWTO for Samba 3.0.23. It includes much of the notes provided
in the <filename moreinfo="none">WHATSNEW.txt</filename> file that is included with the Samba source code release tarball.
</para>

<sect2>
<title>User and Group Changes</title>

<para>
The change documented here affects unmapped user and group accounts only.
</para>

<para>
<indexterm significance="normal"><primary>user</primary></indexterm>
<indexterm significance="normal"><primary>group</primary></indexterm>
<indexterm significance="normal"><primary>Relative Identifiers</primary><see>RID</see></indexterm>
<indexterm significance="normal"><primary>net</primary><secondary>groupmap</secondary></indexterm>
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>vampire</tertiary></indexterm>
The user and group internal management routines have been rewritten to prevent overlaps of
assigned Relative Identifiers (RIDs).  In the past the has been a potential problem when
either manually mapping Unix groups with the <literal>net groupmap</literal> command or
when migrating a Windows domain to a Samba domain by executing:
<literal>net rpc vampire</literal>.
</para>

<para>
<indexterm significance="normal"><primary>SID</primary></indexterm>
<indexterm significance="normal"><primary>SAM</primary></indexterm>
<indexterm significance="normal"><primary>RID</primary></indexterm>
<indexterm significance="normal"><primary>net</primary><secondary>getlocalsid</secondary></indexterm>
Unmapped users are now assigned a SID in the <literal moreinfo="none">S-1-22-1</literal> domain and unmapped
groups are assigned a SID in the <literal moreinfo="none">S-1-22-2</literal> domain.  Previously they were
assigned a RID within the SAM on the Samba server.  For a domain controller this would have been under the
authority of the domain SID where as on a member server or standalone server, this would have
been under the authority of the local SAM (see the man page for <literal>net getlocalsid</literal>).
</para>

<para>
<indexterm significance="normal"><primary>unmapped users</primary></indexterm>
<indexterm significance="normal"><primary>unmapped groups</primary></indexterm>
<indexterm significance="normal"><primary>SID</primary></indexterm>
<indexterm significance="normal"><primary>NTFS</primary></indexterm>
<indexterm significance="normal"><primary>GID</primary></indexterm>
The result is that any unmapped users or groups on an upgraded Samba domain controller may
be assigned a new SID.  Because the SID rather than a name is stored in Windows security
descriptors, this can cause a user to no longer have access to a resource for example if a
file was copied from a Samba file server to a local Windows client NTFS partition.  Any files
stored on the Samba server itself will continue to be accessible because UNIX stores the UNIX
GID and not the SID for authorization checks.
</para>

<para>
An example helps to illustrate the change:
</para>

<para>
<indexterm significance="normal"><primary>group mapping</primary></indexterm>
<indexterm significance="normal"><primary>GID</primary></indexterm>
<indexterm significance="normal"><primary>ACL</primary></indexterm>
<indexterm significance="normal"><primary>SID</primary></indexterm>
Assume that a group named <emphasis>developers</emphasis> exists with a UNIX GID of 782. In this
case this user does not exist in Samba's group mapping table. It would be perfectly normal for
this group to be appear in an ACL editor.  Prior to Samba-3.0.23, the group SID might appear as
<literal moreinfo="none">S-1-5-21-647511796-4126122067-3123570092-2565</literal>.
</para>

<para>
<indexterm significance="normal"><primary>SID</primary></indexterm>
<indexterm significance="normal"><primary>NTFS</primary></indexterm>
<indexterm significance="normal"><primary>access</primary></indexterm>
<indexterm significance="normal"><primary>group permissions</primary></indexterm>
With the release of Samba-3.0.23, the group SID would be reported as <literal moreinfo="none">S-1-22-2-782</literal>.  Any
security descriptors associated with files stored on a Windows NTFS disk partition will not allow access based
on the group permissions if the user was not a member of the
<literal moreinfo="none">S-1-5-21-647511796-4126122067-3123570092-2565</literal>  group.  Because this group SID is
<literal moreinfo="none">S-1-22-2-782</literal> and not reported in a user's token, Windows would fail the authorization check
even though both SIDs in some respect refer to the same UNIX group.
</para>

<para>
<indexterm significance="normal"><primary>group mapping</primary></indexterm>
<indexterm significance="normal"><primary>SID</primary></indexterm>
The workaround for versions of Samba prior to 3.0.23, is to create a manual domain group mapping
entry for the group <emphasis>developers</emphasis> to point at the
<literal moreinfo="none">S-1-5-21-647511796-4126122067-3123570092-2565</literal> SID. With the release of Samba-3.0.23 this
workaround is no longer needed.
</para>
</sect2>

<sect2>
<title>Essential Group Mappings</title>
<para>
Samba 3.0.x series  releases before 3.0.23 automatically created group mappings for the essential Windows
domain groups <literal moreinfo="none">Domain Admins, Domain Users, Domain Guests</literal>. Commencing with Samba 3.0.23
these mappings need to be created by the Samba administrator. Failure to do this may result in a failure to
correctly authenticate and recoognize valid domain users. When this happens users will not be able to log onto
the Windows client.
</para>

<note><para>
Group mappings are essentail only if the Samba servers is running as a PDC/BDC. Stand-alone servers do not
require these group mappings.
</para></note>

<para>
The following mappings are required:
</para>

<table frame="all" id="TOSH-domgroups">
	<title>Essential Domain Group Mappings</title>
	<tgroup align="center" cols="3">
	<thead>
		<row><entry>Domain Group</entry><entry>RID</entry><entry>Example UNIX Group</entry></row>
	</thead>
	<tbody>
		<row><entry>Domain Admins</entry><entry>512</entry><entry>root</entry></row>
		<row><entry>Domain Users</entry><entry>513</entry><entry>users</entry></row>
		<row><entry>Domain Guests</entry><entry>514</entry><entry>nobody</entry></row>
	</tbody>
	</tgroup>
</table>

<para>
When the POSIX (UNIX) groups are stored in LDAP, it may be desirable to call these <literal moreinfo="none">domadmins, domusers,
domguests</literal> respectively.
</para>

<para>
For further information regarding group mappings see <link linkend="groupmapping">Group Mapping: MS Windows
and UNIX</link>.
</para>

</sect2>

<sect2>
<title>Passdb Changes</title>

<para>
<indexterm significance="normal"><primary>backends</primary></indexterm>
<indexterm significance="normal"><primary>GID</primary></indexterm>
<indexterm significance="normal"><primary>SQL</primary></indexterm>
<indexterm significance="normal"><primary>XML</primary></indexterm>
The <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PASSDBBACKEND" xlink:href="smb.conf.5.html#PASSDBBACKEND">passdb backend</link> parameter no long accepts multiple passdb backends in a
chained configuration.  Also be aware that the SQL and XML based passdb modules have been
removed in the Samba-3.0.23 release.  More information regarding external support for a SQL
passdb module can be found on the  <ulink url="http://pdbsql.sourceforge.net/">pdbsql</ulink> web site.
</para>

</sect2>

<sect2>
<title>Group Mapping Changes in Samba-3.0.23</title>

<para>
<indexterm significance="normal"><primary>default mapping</primary></indexterm>
<indexterm significance="normal"><primary>Domain Admins</primary></indexterm>
<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
<indexterm significance="normal"><primary>tdbsam</primary></indexterm>
<indexterm significance="normal"><primary>passdb backend</primary></indexterm>
<indexterm significance="normal"><primary>group mappings</primary></indexterm>
<indexterm significance="normal"><primary>GID</primary></indexterm>
<indexterm significance="normal"><primary>SID</primary></indexterm>
<indexterm significance="normal"><primary>IDMAP</primary></indexterm>
<indexterm significance="normal"><primary>winbindd</primary></indexterm>
<indexterm significance="normal"><primary>domain groups</primary></indexterm>
The default mapping entries for groups such as <literal moreinfo="none">Domain Admins</literal> are no longer
created when using an <literal moreinfo="none">smbpasswd</literal> file or a <literal moreinfo="none">tdbsam</literal> passdb
backend.  This means that it is necessary to explicitly execute the <literal>net groupmap add</literal>
to create group mappings, rather than use the <literal>net groupmap modify</literal> method to create the
Windows group SID to UNIX GID mappings.  This change has no effect on winbindd's IDMAP functionality
for domain groups.
</para>

</sect2>

<sect2>
<title>LDAP Changes in Samba-3.0.23</title>

<para>
<indexterm significance="normal"><primary>LDAP schema</primary></indexterm>
<indexterm significance="normal"><primary>sambaSID</primary></indexterm>
<indexterm significance="normal"><primary>OpenLDAP</primary></indexterm>
<indexterm significance="normal"><primary>slapindex</primary></indexterm>
<indexterm significance="normal"><primary>slapd.conf</primary></indexterm>
There has been a minor update the Samba LDAP schema file. A substring matching rule has been
added to the <literal moreinfo="none">sambaSID</literal> attribute definition.  For OpenLDAP servers, this
will require the addition of <literal moreinfo="none">index sambaSID sub</literal> to the
<filename moreinfo="none">slapd.conf</filename> configuration file.  It will be necessary to execute the
<literal>slapindex</literal> command after making this change. There has been no change to the
actual data storage schema.
</para>

</sect2>
</sect1>

</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="NetworkBrowsing">
<chapterinfo>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
	<author>
<firstname>Jelmer</firstname><othername>R.</othername><surname>Vernooij</surname><othername>R.</othername>
<affiliation>
	<orgname>The Samba Team</orgname>
	<address format="linespecific"><email>jelmer@samba.org</email></address>
</affiliation></author>
	<author>
		<firstname>Jonathan</firstname><surname>Johnson</surname>
		<affiliation>
			<orgname>Sutinen Consulting, Inc.</orgname>
                        <address format="linespecific"><email>jon@sutinen.com</email></address>
		</affiliation>
	</author>
	<pubdate>July 5, 1998</pubdate>
	<pubdate>Updated: September 20, 2006</pubdate>
</chapterinfo>

<title>Network Browsing</title>

<para>
<indexterm significance="normal"><primary>browsing across subnets</primary></indexterm>
<indexterm significance="normal"><primary>resolution of NetBIOS names</primary></indexterm>
<indexterm significance="normal"><primary>browse list handling</primary></indexterm>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
This chapter contains detailed information as well as a fast-track guide to
implementing browsing across subnets and/or across workgroups (or domains).
WINS is the best tool for resolution of NetBIOS names to IP addresses; however, WINS is
not involved in browse list handling except by way of name-to-address resolution.
</para>

<note><para>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
What is WINS?
</para>
<para>
WINS is a facility that provides resolution of a NetBIOS name to its IP address. WINS is like a
Dynamic-DNS service for NetBIOS networking names.
</para></note>

<note><para>
<indexterm significance="normal"><primary>Windows 2000</primary></indexterm>
<indexterm significance="normal"><primary>NetBIOS over TCP/IP</primary></indexterm>
<indexterm significance="normal"><primary>DNS</primary></indexterm>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
MS Windows 2000 and later versions can be configured to operate with no NetBIOS
over TCP/IP. Samba-3 and later versions also support this mode of operation.
When the use of NetBIOS over TCP/IP has been disabled, the primary
means for resolution of MS Windows machine names is via DNS and Active Directory.
The following information assumes that your site is running NetBIOS over TCP/IP.
</para></note>

<sect1>
<title>Features and Benefits</title>

<para>
Charles Dickens once referred to the past in these words: <quote><emphasis>It was the best of times,
it was the worst of times.</emphasis></quote> The more we look back, the more we long for what was and
hope it never returns.
</para>


<para>
<indexterm significance="normal"><primary>NetBIOS</primary></indexterm>
<indexterm significance="normal"><primary>NetBIOS networking</primary></indexterm>
<indexterm significance="normal"><primary>fickle</primary></indexterm>
For many MS Windows network administrators, that statement sums up their feelings about
NetBIOS networking precisely. For those who mastered NetBIOS networking, its fickle
nature was just par for the course. For those who never quite managed to tame its
lusty features, NetBIOS is like Paterson's Curse.
</para>

<para>
For those not familiar with botanical problems in Australia, Paterson's Curse,
<emphasis>Echium plantagineum</emphasis>, was introduced to Australia from Europe during the mid-19th
century. Since then it has spread rapidly. The high seed production, with densities of
thousands of seeds per square meter, a seed longevity of more than 7 years, and an
ability to germinate at any time of year, given the right conditions, are some of the
features that make it such a persistent weed.
</para>

<para>
<indexterm significance="normal"><primary>Network Basic Input/Output System</primary><see>NetBIOS</see></indexterm>
<indexterm significance="normal"><primary>SMB</primary></indexterm>
<indexterm significance="normal"><primary>NetBIOS</primary></indexterm>
<indexterm significance="normal"><primary>TCP/IP</primary></indexterm>
<indexterm significance="normal"><primary>Windows network clients</primary></indexterm>
In this chapter we explore vital aspects of Server Message Block (SMB) networking with
a particular focus on SMB as implemented through running NetBIOS (Network Basic
Input/Output System) over TCP/IP. Since Samba does not implement SMB or NetBIOS over
any other protocols, we need to know how to configure our network environment and simply
remember to use nothing but TCP/IP on all our MS Windows network clients.
</para>

<para>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
<indexterm significance="normal"><primary>MS WINS</primary></indexterm>
Samba provides the ability to implement a WINS (Windows Internetworking Name Server)
and implements extensions to Microsoft's implementation of WINS. These extensions
help Samba to effect stable WINS operations beyond the normal scope of MS WINS.
</para>

<para>
<indexterm significance="normal"><primary>NetBIOS over TCP/IP</primary></indexterm>
<indexterm significance="normal"><primary>NetBIOS disabled</primary></indexterm>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
WINS is exclusively a service that applies only to those systems
that run NetBIOS over TCP/IP. MS Windows 200x/XP have the capacity to operate with
support for NetBIOS disabled, in which case WINS is of no relevance. Samba supports this also.
</para>

<para>
<indexterm significance="normal"><primary>NetBIOS disabled</primary></indexterm>
<indexterm significance="normal"><primary>DNS</primary></indexterm>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
For those networks on which NetBIOS has been disabled (i.e., WINS is not required),
the use of DNS is necessary for hostname resolution.
</para>

</sect1>

<sect1>
<title>What Is Browsing?</title>

<para>
<indexterm significance="normal"><primary>browsing</primary></indexterm>
<indexterm significance="normal"><primary>Network Neighborhood</primary></indexterm>
<indexterm significance="normal"><primary>shares</primary></indexterm>
<indexterm significance="normal"><primary>printers available</primary></indexterm>
To most people, browsing means they can see the MS Windows and Samba servers
in the Network Neighborhood, and when the computer icon for a particular server is
clicked, it opens up and shows the shares and printers available on the target server.
</para>

<para>
What seems so simple is in fact a complex interaction of different technologies.
The technologies (or methods) employed in making all of this work include:
</para>

<itemizedlist>
	<listitem><para>MS Windows machines register their presence to the network.</para></listitem>
	<listitem><para>Machines announce themselves to other machines on the network.</para></listitem>
	<listitem><para>One or more machines on the network collate the local announcements.</para></listitem>
	<listitem><para>The client machine finds the machine that has the collated list of machines.</para></listitem>
	<listitem><para>The client machine is able to resolve the machine names to IP addresses.</para></listitem>
	<listitem><para>The client machine is able to connect to a target machine.</para></listitem>
</itemizedlist>

<para>
<indexterm significance="normal"><primary>browse list management</primary></indexterm>
<indexterm significance="normal"><primary>name resolution</primary></indexterm>
<indexterm significance="normal"><primary>nmbd</primary></indexterm>
The Samba application that controls browse list management and name resolution is
called <filename moreinfo="none">nmbd</filename>. The configuration parameters involved in nmbd's operation are:
</para>

<para>
Browsing options:
</para>
<itemizedlist>
	<listitem><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="OSLEVEL" xlink:href="smb.conf.5.html#OSLEVEL">os level</link></listitem>
	<listitem><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LMANNOUNCE" xlink:href="smb.conf.5.html#LMANNOUNCE">lm announce</link></listitem>
	<listitem><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LMINTERVAL" xlink:href="smb.conf.5.html#LMINTERVAL">lm interval</link></listitem>
	<listitem><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PREFERREDMASTER" xlink:href="smb.conf.5.html#PREFERREDMASTER">preferred master</link>(*)</listitem>
	<listitem><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOCALMASTER" xlink:href="smb.conf.5.html#LOCALMASTER">local master</link>(*)</listitem>
	<listitem><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="DOMAINMASTER" xlink:href="smb.conf.5.html#DOMAINMASTER">domain master</link>(*)</listitem>
	<listitem><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="BROWSELIST" xlink:href="smb.conf.5.html#BROWSELIST">browse list</link></listitem>
	<listitem><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="ENHANCEDBROWSING" xlink:href="smb.conf.5.html#ENHANCEDBROWSING">enhanced browsing</link></listitem>
</itemizedlist>

<para>
Name Resolution Method:
</para>
<itemizedlist>
	<listitem><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="NAMERESOLVEORDER" xlink:href="smb.conf.5.html#NAMERESOLVEORDER">name resolve order</link>(*)</listitem>
</itemizedlist>

<para>
WINS options:
</para>
<itemizedlist>
	<listitem><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="DNSPROXY" xlink:href="smb.conf.5.html#DNSPROXY">dns proxy</link></listitem>
	<listitem><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="WINSPROXY" xlink:href="smb.conf.5.html#WINSPROXY">wins proxy</link></listitem>
	<listitem><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="WINSSERVER" xlink:href="smb.conf.5.html#WINSSERVER">wins server</link>(*)</listitem>
	<listitem><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="WINSSUPPORT" xlink:href="smb.conf.5.html#WINSSUPPORT">wins support</link>(*)</listitem>
	<listitem><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="WINSHOOK" xlink:href="smb.conf.5.html#WINSHOOK">wins hook</link></listitem>
</itemizedlist>

<para>
Those marked with an (*) are the only options that commonly may need to be modified. Even if none of these
parameters is set, <filename moreinfo="none">nmbd</filename> will still do its job.
</para>

<para>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
<indexterm significance="normal"><primary>WINS Server</primary></indexterm>
<indexterm significance="normal"><primary>WINS Support</primary></indexterm>
<indexterm significance="normal"><primary>nmbd</primary></indexterm>
<indexterm significance="normal"><primary>mutually exclusive options</primary></indexterm>
For Samba, the WINS Server and WINS Support are mutually exclusive options. When <literal>nmbd</literal> is
started it will fail to execute if both options are set in the <filename moreinfo="none">smb.conf</filename> file. The <literal>nmbd</literal>
understands that when it spawns an instance of itself to run as a WINS server that it has to use its own WINS
server also.
</para>

</sect1>

<sect1 id="netdiscuss">
<title>Discussion</title>

<para>
<indexterm significance="normal"><primary>SMB-based messaging</primary></indexterm>
<indexterm significance="normal"><primary>NetBIOS</primary></indexterm>
<indexterm significance="normal"><primary>NetBIOS</primary></indexterm>
<indexterm significance="normal"><primary>phasing out NetBIOS</primary></indexterm>
All MS Windows networking uses SMB-based messaging.  SMB messaging may be implemented with or without NetBIOS.
MS Windows 200x supports NetBIOS over TCP/IP for backwards compatibility. Microsoft appears intent on phasing
out NetBIOS support.
</para>

<sect2>
<title>NetBIOS over TCP/IP</title>

<para>
<indexterm significance="normal"><primary>encapsulating</primary></indexterm>
<indexterm significance="normal"><primary>broadcast</primary></indexterm>
<indexterm significance="normal"><primary>unicast</primary></indexterm>
<indexterm significance="normal"><primary>UDP</primary></indexterm>
Samba implements NetBIOS, as does MS Windows NT/200x/XP, by encapsulating it over TCP/IP.
NetBIOS-based networking uses broadcast messaging to effect browse list management. When running NetBIOS over
TCP/IP, this uses UDP-based messaging.  UDP messages can be broadcast or unicast.
</para>

<para>
<indexterm significance="normal"><primary>UDP</primary></indexterm>
Normally, only unicast UDP messaging can be forwarded by routers. The <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="REMOTEANNOUNCE" xlink:href="smb.conf.5.html#REMOTEANNOUNCE">remote announce</link>
parameter to smb.conf helps to project browse announcements to remote network segments via unicast UDP.
Similarly, the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="REMOTEBROWSESYNC" xlink:href="smb.conf.5.html#REMOTEBROWSESYNC">remote browse sync</link> parameter of <filename moreinfo="none">smb.conf</filename> implements browse list
collation using unicast UDP.
</para>

<para>
The methods used by MS Windows to perform name lookup requests (name resolution) is determined by a
configuration parameter called the NetBIOS node-type. There are four basic NetBIOS node types:
</para>

<indexterm significance="normal"><primary>b-node</primary></indexterm>
<indexterm significance="normal"><primary>p-node</primary></indexterm>
<indexterm significance="normal"><primary>m-node</primary></indexterm>
<indexterm significance="normal"><primary>h-node</primary></indexterm>
<indexterm significance="normal"><primary>node-type</primary></indexterm>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
<indexterm significance="normal"><primary>broadcast</primary></indexterm>
<indexterm significance="normal"><primary>unicast</primary></indexterm>
<itemizedlist>
	<listitem><para><emphasis>b-node (type 0x01):</emphasis> The Windows client will use only
	NetBIOS broadcast requests using UDP broadcast.</para></listitem>
	<listitem><para><emphasis>p-node (type 0x02):</emphasis> The Windows client will use point-to-point
	(NetBIOS unicast) requests using UDP unicast directed to a WINS server.</para></listitem>
	<listitem><para><emphasis>m-node (type 0x04):</emphasis> The Windows client will first use
	NetBIOS broadcast requests using UDP broadcast, then it will use (NetBIOS unicast)
	requests using UDP unicast directed to a WINS server.</para></listitem>
	<listitem><para><emphasis>h-node (type 0x08):</emphasis> The Windows client will use
	(NetBIOS unicast) requests using UDP unicast directed to a WINS server, then it will use
	NetBIOS broadcast requests using UDP broadcast.</para></listitem>
</itemizedlist>

<para>
<indexterm significance="normal"><primary>h-node</primary></indexterm>
<indexterm significance="normal"><primary>hybrid</primary></indexterm>
<indexterm significance="normal"><primary>enables NetBIOS over TCP/IP</primary></indexterm>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
<indexterm significance="normal"><primary>broadcast-based</primary></indexterm>
<indexterm significance="normal"><primary>name resolution</primary></indexterm>
The default Windows network client (or server) network configuration enables NetBIOS over TCP/IP
and b-node configuration. The use of WINS makes most sense with h-node (hybrid mode) operation so that
in the event of a WINS breakdown or non-availability, the client can use broadcast-based name resolution.
</para>

<para>
<indexterm significance="normal"><primary>LMB</primary><see>Local Master Browser</see></indexterm>
<indexterm significance="normal"><primary>Local Master Browser</primary></indexterm>
<indexterm significance="normal"><primary>SMB</primary></indexterm>
<indexterm significance="normal"><primary>nmbd</primary></indexterm>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
<indexterm significance="normal"><primary>cross-segment browsing</primary></indexterm>
<indexterm significance="normal"><primary>network segment</primary></indexterm>
In those networks where Samba is the only SMB server technology, wherever possible <filename moreinfo="none">nmbd</filename>
should be configured on one machine as the WINS server. This makes it easy to manage the browsing environment.
If each network segment is configured with its own Samba WINS server, then the only way to get cross-segment
browsing to work is by using the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="REMOTEANNOUNCE" xlink:href="smb.conf.5.html#REMOTEANNOUNCE">remote announce</link> and the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="REMOTEBROWSESYNC" xlink:href="smb.conf.5.html#REMOTEBROWSESYNC">remote browse sync</link> parameters to your <filename moreinfo="none">smb.conf</filename> file.
</para>

<para>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
If only one WINS server is used for an entire multisegment network, then
the use of the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="REMOTEANNOUNCE" xlink:href="smb.conf.5.html#REMOTEANNOUNCE">remote announce</link> and the
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="REMOTEBROWSESYNC" xlink:href="smb.conf.5.html#REMOTEBROWSESYNC">remote browse sync</link> parameters should not be necessary.
</para>

<para>
<indexterm significance="normal"><primary>replication</primary><secondary>WINS</secondary></indexterm>
As of Samba-3, WINS replication is being worked on. The bulk of the code has been committed, but it still
needs maturation. This is not a supported feature of the Samba-3.0.20 release. Hopefully, this will become a
supported feature of one of the Samba-3 release series. The delay is caused by the fact that this feature has
not been of sufficient significance to inspire someone to pay a developer to complete it.
</para>

<para>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
<indexterm significance="normal"><primary>MS-WINS replication</primary></indexterm>
<indexterm significance="normal"><primary>redundancy</primary></indexterm>
<indexterm significance="normal"><primary>DNS</primary></indexterm>
<indexterm significance="normal"><primary>NetBIOSless SMB over TCP/IP</primary></indexterm>
<indexterm significance="normal"><primary>local names</primary></indexterm>
<indexterm significance="normal"><primary>subnets</primary></indexterm>
<indexterm significance="normal"><primary>multiple WINS servers</primary></indexterm>
Right now Samba WINS does not support MS-WINS replication. This means that when setting up Samba as a WINS
server, there must only be one <filename moreinfo="none">nmbd</filename> configured as a WINS server on the network. Some
sites have used multiple Samba WINS servers for redundancy (one server per subnet) and then used
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="REMOTEBROWSESYNC" xlink:href="smb.conf.5.html#REMOTEBROWSESYNC">remote browse sync</link> and <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="REMOTEANNOUNCE" xlink:href="smb.conf.5.html#REMOTEANNOUNCE">remote announce</link> to effect browse list
collation across all segments. Note that this means clients will only resolve local names and must be
configured to use DNS to resolve names on other subnets in order to resolve the IP addresses of the servers
they can see on other subnets. This setup is not recommended but is mentioned as a practical consideration
(i.e., an <quote>if all else fails</quote> scenario). NetBIOS over TCP/IP is an ugly and difficult to manage
protocol. Its replacement, NetBIOSless SMB over TCP/IP is not without its own manageability concerns. NetBIOS
based networking is a life of compromise and trade-offs. WINS stores information that cannot be stored in
DNS; consequently, DNS is a poor substitute for WINS given that when NetBIOS over TCP/IP is used, Windows
clients are designed to use WINS.
</para>

<para>
<indexterm significance="normal"><primary>broadcast messages</primary></indexterm>
<indexterm significance="normal"><primary>repeated intervals</primary></indexterm>
<indexterm significance="normal"><primary>across network segments</primary></indexterm>
Lastly, take note that browse lists are a collection of unreliable broadcast
messages that are repeated at intervals of not more than 15 minutes. This means
that it will take time to establish a browse list, and it can take up to 45
minutes to stabilize, particularly across network segments.
</para>

<para>
<indexterm significance="normal"><primary>Windows 200x/XP</primary></indexterm>
When an MS Windows 200x/XP system attempts to resolve a host name to an IP address, it follows a defined path:
</para>

<orderedlist inheritnum="ignore" continuation="restarts">
	<listitem><para>
	Checks the <filename moreinfo="none">hosts</filename> file. It is located in <filename moreinfo="none">%SystemRoot%\System32\Drivers\etc</filename>.
	</para></listitem>

	<listitem><para>
	Does a DNS lookup.
	</para></listitem>

	<listitem><para>
	Checks the NetBIOS name cache.
	</para></listitem>

	<listitem><para>
	Queries the WINS server.
	</para></listitem>

	<listitem><para>
	Does a broadcast name lookup over UDP.
	</para></listitem>

	<listitem><para>
	Looks up entries in LMHOSTS, located in <filename moreinfo="none">%SystemRoot%\System32\Drivers\etc</filename>.
	</para></listitem>
</orderedlist>

<para>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
<indexterm significance="normal"><primary>NetBIOS over TCP/IP</primary></indexterm>
<indexterm significance="normal"><primary>name lookups</primary></indexterm>
<indexterm significance="normal"><primary>DNS</primary></indexterm>
Given the nature of how the NetBIOS over TCP/IP protocol is implemented, only WINS is capable of resolving
with any reliability name lookups for service-oriented names such as TEMPTATION&lt;1C&gt; <?latex --- ?> a NetBIOS
name query that seeks to find network logon servers. DNS has no concept of service-oriented names such as
this. In fact, the Microsoft ADS implementation specifically manages a whole range of extended
service-oriented DNS entries. This type of facility is not implemented and is not supported for the NetBIOS
over TCP/IP protocol namespace.
</para>

</sect2>

<sect2>
<title>TCP/IP without NetBIOS</title>

<para>
<indexterm significance="normal"><primary>NetBIOS</primary></indexterm>
<indexterm significance="normal"><primary>NetBIOS-less</primary></indexterm>
<indexterm significance="normal"><primary>DNS</primary></indexterm>
All TCP/IP-enabled systems use various forms of hostname resolution. The primary
methods for TCP/IP hostname resolution involve either a static file (<filename moreinfo="none">/etc/hosts</filename>)
or the Domain Name System (DNS). DNS is the technology that makes
the Internet usable. DNS-based hostname resolution is supported by nearly all
TCP/IP-enabled systems. Only a few embedded TCP/IP systems do not support DNS.
</para>

<para>
<indexterm significance="normal"><primary>DNS</primary></indexterm>
<indexterm significance="normal"><primary>DDNS</primary></indexterm>
<indexterm significance="normal"><primary>ipconfig</primary></indexterm>
<indexterm significance="normal"><primary>Dynamic DNS</primary><see>DDNS</see></indexterm>
Windows 200x/XP can register its hostname with a Dynamic DNS server (DDNS). It is possible to force register with a
dynamic DNS server in Windows 200x/XP using <literal>ipconfig /registerdns</literal>.
</para>

<para>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
<indexterm significance="normal"><primary>DNS</primary></indexterm>
<indexterm significance="normal"><primary>severely impaired</primary></indexterm>
With Active Directory, a correctly functioning DNS server is absolutely essential. In the absence of a working
DNS server that has been correctly configured, MS Windows clients and servers will be unable to locate each
other, so network services consequently will be severely impaired.
</para>

<para>
<indexterm significance="normal"><primary>raw SMB over TCP/IP</primary></indexterm>
<indexterm significance="normal"><primary>No NetBIOS layer</primary></indexterm>
<indexterm significance="normal"><primary>NetBIOS</primary></indexterm>
<indexterm significance="normal"><primary>domain member server</primary></indexterm>
<indexterm significance="normal"><primary>DNS</primary></indexterm>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
Use of raw SMB over TCP/IP (No NetBIOS layer) can be done only with Active Directory domains. Samba is not an
Active Directory domain controller: ergo, it is not possible to run Samba as a domain controller and at the same
time <emphasis>not</emphasis> use NetBIOS. Where Samba is used as an Active Directory domain member server
(DMS) it is possible to configure Samba to not use NetBIOS over TCP/IP. A Samba DMS can integrate fully into
an Active Directory domain, however, if NetBIOS over TCP/IP is disabled, it is necessary to manually create
appropriate DNS entries for the Samba DMS because they will not be automatically generated either by Samba, or
by the ADS environment.
</para>

</sect2>

<sect2 id="adsdnstech">
<title>DNS and Active Directory</title>

<para>
<indexterm significance="normal"><primary>DNS</primary><secondary>Active Directory</secondary></indexterm>
<indexterm significance="normal"><primary>DDNS</primary></indexterm>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
<indexterm significance="normal"><primary>SRV records</primary></indexterm>
<indexterm significance="normal"><primary>DNS</primary><secondary>SRV records</secondary></indexterm>
Occasionally we hear from UNIX network administrators who want to use a UNIX-based DDNS server in place
of the Microsoft DNS server. While this might be desirable to some, the MS Windows 200x DNS server is
autoconfigured to work with Active Directory. It is possible to use BIND version 8 or 9, but it will almost
certainly be necessary to create service records (SRV records) so MS Active Directory clients can resolve
hostnames to locate essential network services.  The following are some of the default service records that
Active Directory requires:
</para>

<para>
<indexterm significance="normal"><primary>DDNS</primary></indexterm>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
<indexterm significance="normal"><primary>BIND9</primary></indexterm>
The use of DDNS is highly recommended with Active Directory, in which case the use of BIND9 is preferred for
its ability to adequately support the SRV (service) records that are needed for Active Directory. Of course,
when running ADS, it makes sense to use Microsoft's own DDNS server because of the natural affinity between ADS
and MS DNS.
</para>

<variablelist>
<varlistentry>
	<term>_ldap._tcp.pdc._msdcs.<emphasis>Domain</emphasis></term>
	<listitem>
	<para>
	This provides the address of the Windows NT PDC for the domain.
	</para>
	</listitem>
</varlistentry>
<varlistentry>
	<term>_ldap._tcp.pdc._msdcs.<emphasis>DomainTree</emphasis></term>
	<listitem>
        <para>
	Resolves the addresses of global catalog servers in the domain.
        </para>
	</listitem>
</varlistentry>
<varlistentry>
        <term>_ldap._tcp.<emphasis>site</emphasis>.sites.writable._msdcs.<emphasis>Domain</emphasis></term>
	<listitem>
        <para>
	Provides list of domain controllers based on sites.
        </para>
	</listitem>
</varlistentry>
<varlistentry>
        <term>_ldap._tcp.writable._msdcs.<emphasis>Domain</emphasis></term>
	<listitem>
        <para>
	Enumerates list of domain controllers that have the writable copies of the Active Directory data store.
        </para>
	</listitem>
</varlistentry>
<varlistentry>
	<term>_ldap._tcp.<emphasis>GUID</emphasis>.domains._msdcs.<emphasis>DomainTree</emphasis></term>
	<listitem>
	<para>
	Entry used by MS Windows clients to locate machines using the global unique identifier.
	</para>
	</listitem>
</varlistentry>
<varlistentry>
	<term>_ldap._tcp.<emphasis>Site</emphasis>.gc._msdcs.<emphasis>DomainTree</emphasis></term>
	<listitem>
	<para>
	Used by Microsoft Windows clients to locate the site configuration-dependent global catalog server.
	</para>
	</listitem>
</varlistentry>
</variablelist>

	<para>
	Specific entries used by Microsoft clients to locate essential services for an example domain
	called <constant>quenya.org</constant> include:
	</para>

	<itemizedlist>
		<listitem><para>
		_kerberos._udp.quenya.org <?latex --- ?> Used to contact the KDC server via UDP.
		This entry must list port 88 for each KDC.
		</para></listitem>

		<listitem><para>
		_kpasswd._udp.quenya.org <?latex --- ?> Used to locate the <constant>kpasswd</constant> server
		when a user password change must be processed. This record must list port 464 on the
		master KDC.
		</para></listitem>

		<listitem><para>
		_kerberos._tcp.quenya.org <?latex --- ?> Used to locate the KDC server via TCP.
		This entry must list port 88 for each KDC.
		</para></listitem>

		<listitem><para>
		_ldap._tcp.quenya.org <?latex --- ?> Used to locate the LDAP service on the PDC.
		This record must list port 389 for the PDC.
		</para></listitem>

		<listitem><para>
		_kpasswd._tcp.quenya.org <?latex --- ?> Used to locate the <constant>kpasswd</constant> server
		to permit user password changes to be processed. This must list port 464.
		</para></listitem>

		<listitem><para>
		_gc._tcp.quenya.org <?latex --- ?> Used to locate the global catalog server  for the
		top of the domain. This must list port 3268.
		</para></listitem>
	</itemizedlist>

	<para>
	The following records are also used by the Windows domain member client to locate vital
	services on the Windows ADS domain controllers.
	</para>

	<itemizedlist>
		<listitem><para>
		_ldap._tcp.pdc._msdcs.quenya.org
		</para></listitem>

		<listitem><para>
		_ldap.gc._msdcs.quenya.org
		</para></listitem>

		<listitem><para>
		_ldap.default-first-site-name._sites.gc._msdcs.quenya.org
		</para></listitem>

		<listitem><para>
		_ldap.{SecID}.domains._msdcs.quenya.org
		</para></listitem>

		<listitem><para>
		_ldap._tcp.dc._msdcs.quenya.org
		</para></listitem>

		<listitem><para>
		_kerberos._tcp.dc._msdcs.quenya.org
		</para></listitem>

		<listitem><para>
		_ldap.default-first-site-name._sites.dc._msdcs.quenya.org
		</para></listitem>

		<listitem><para>
		_kerberos.default-first-site-name._sites.dc._msdcs.queyna.org
		</para></listitem>

		<listitem><para>
		SecID._msdcs.quenya.org
		</para></listitem>
	</itemizedlist>

	<para>
	Presence of the correct DNS entries can be validated by executing:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> dig @frodo -t any _ldap._tcp.dc._msdcs.quenya.org

; &lt;lt;&gt;&gt; DiG 9.2.2 &lt;lt;&gt;&gt; @frodo -t any _ldap._tcp.dc._msdcs.quenya.org
;; global options:  printcmd
;; Got answer:
;; -&gt;&gt;HEADER&lt;&lt;- opcode: QUERY, status: NOERROR, id: 3072
;; flags: qr aa rd ra; QUERY: 1, ANSWER: 2, AUTHORITY: 0, ADDITIONAL: 2


;; QUESTION SECTION:
;_ldap._tcp.dc._msdcs.quenya.org. IN        ANY


;; ANSWER SECTION:
_ldap._tcp.dc._msdcs.quenya.org. 600 IN SRV 0 100 389 frodo.quenya.org.
_ldap._tcp.dc._msdcs.quenya.org. 600 IN SRV 0 100 389 noldor.quenya.org.


;; ADDITIONAL SECTION:
frodo.quenya.org.  3600  IN      A       10.1.1.16
noldor.quenya.org. 1200  IN      A       10.1.1.17


;; Query time: 0 msec
;; SERVER: frodo#53(10.1.1.16)
;; WHEN: Wed Oct  7 14:39:31 2004
;; MSG SIZE  rcvd: 171
</screen>
	</para>

</sect2>

</sect1>

<sect1>
<title>How Browsing Functions</title>

<para>
<indexterm significance="normal"><primary>register NetBIOS names</primary></indexterm>
<indexterm significance="normal"><primary>LMHOSTS</primary></indexterm>
<indexterm significance="normal"><primary>DNS</primary></indexterm>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
<indexterm significance="normal"><primary>WINS server address</primary></indexterm>
MS Windows machines register their NetBIOS names (i.e., the machine name for each service type in operation)
on startup.  The exact method by which this name registration takes place is determined by whether or not the
MS Windows client/server has been given a WINS server address, whether or not LMHOSTS lookup is enabled,
whether or not DNS for NetBIOS name resolution is enabled, and so on.
</para>

<para>
<indexterm significance="normal"><primary>WINS server</primary></indexterm>
<indexterm significance="normal"><primary>name lookups</primary></indexterm>
<indexterm significance="normal"><primary>UDP</primary></indexterm>
In the case where there is no WINS server, all name registrations as well as name lookups are done by UDP
broadcast. This isolates name resolution to the local subnet, unless LMHOSTS is used to list all names and IP
addresses. In such situations, Samba provides a means by which the Samba server name may be forcibly injected
into the browse list of a remote MS Windows network (using the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="REMOTEANNOUNCE" xlink:href="smb.conf.5.html#REMOTEANNOUNCE">remote announce</link>
parameter).
</para>

<para>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
<indexterm significance="normal"><primary>UDP unicast</primary></indexterm>
<indexterm significance="normal"><primary>name resolution across routed networks</primary></indexterm>
Where a WINS server is used, the MS Windows client will use UDP unicast to register with the WINS server. Such
packets can be routed, and thus WINS allows name resolution to function across routed networks.
</para>

<para>
<indexterm significance="normal"><primary>LMB</primary></indexterm>
<indexterm significance="normal"><primary>local master browser</primary><see>LMB</see></indexterm>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
<indexterm significance="normal"><primary>LMHOSTS</primary></indexterm>
<indexterm significance="normal"><primary>DMB</primary></indexterm>
<indexterm significance="normal"><primary>browse list</primary></indexterm>
<indexterm significance="normal"><primary>election</primary></indexterm>
<indexterm significance="normal"><primary>election criteria</primary></indexterm>
During the startup process, an election takes place to create a local master browser (LMB) if one does not
already exist. On each NetBIOS network one machine will be elected to function as the domain master browser
(DMB). This domain browsing has nothing to do with MS security Domain Control.  Instead, the DMB serves the
role of contacting each LMB (found by asking WINS or from LMHOSTS) and exchanging browse list contents. This
way every master browser will eventually obtain a complete list of all machines that are on the network. Every
11 to 15 minutes an election is held to determine which machine will be the master browser. By the nature of
the election criteria used, the machine with the highest uptime, or the most senior protocol version or other
criteria, will win the election as DMB.
</para>

<para>
<indexterm significance="normal"><primary>WINS server</primary></indexterm>
<indexterm significance="normal"><primary>DMB</primary></indexterm>
<indexterm significance="normal"><primary>NetBIOS name type</primary></indexterm>
<indexterm significance="normal"><primary>n security context</primary></indexterm>
<indexterm significance="normal"><primary>network segment</primary></indexterm>
<indexterm significance="normal"><primary>authoritive</primary></indexterm>
<indexterm significance="normal"><primary>browse list maintainers</primary></indexterm>
<indexterm significance="normal"><primary>LMB</primary></indexterm>
Where a WINS server is used, the DMB registers its IP address with the WINS server using the name of the
domain and the NetBIOS name type 1B (e.g., DOMAIN&lt;1B&gt;). All LMBs register their IP addresses with the WINS
server, also with the name of the domain and the NetBIOS name type of 1D. The 1B name is unique to one
server within the domain security context, and only one 1D name is registered for each network segment.
Machines that have registered the 1D name will be authoritive browse list maintainers for the network segment
they are on. The DMB is responsible for synchronizing the browse lists it obtains from the LMBs.
</para>

<para>
<indexterm significance="normal"><primary>name resolution</primary></indexterm>
Clients wishing to browse the network make use of this list but also depend on the availability of correct
name resolution to the respective IP address or addresses.
</para>

<para>
<indexterm significance="normal"><primary>browsing intrinsics</primary></indexterm>
Any configuration that breaks name resolution and/or browsing intrinsics will annoy users because they will
have to put up with protracted inability to use the network services.
</para>

<para>
<indexterm significance="normal"><primary>forced synchronization</primary></indexterm>
<indexterm significance="normal"><primary>LMB</primary></indexterm>
<indexterm significance="normal"><primary>bridges networks</primary></indexterm>
<indexterm significance="normal"><primary>cross-subnet browsing</primary></indexterm>
<indexterm significance="normal"><primary>DNS</primary></indexterm>
<indexterm significance="normal"><primary>/etc/hosts</primary></indexterm>
Samba supports a feature that allows forced synchronization of browse lists across routed networks using the
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="REMOTEBROWSESYNC" xlink:href="smb.conf.5.html#REMOTEBROWSESYNC">remote browse sync</link> parameter in the <filename moreinfo="none">smb.conf</filename> file. This causes Samba to contact the
LMB on a remote network and to request browse list synchronization. This effectively bridges two networks that
are separated by routers. The two remote networks may use either broadcast-based name resolution or WINS-based
name resolution, but it should be noted that the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="REMOTEBROWSESYNC" xlink:href="smb.conf.5.html#REMOTEBROWSESYNC">remote browse sync</link> parameter provides
browse list synchronization <?latex --- ?> and that is distinct from name-to-address resolution. In other words,
for cross-subnet browsing to function correctly, it is essential that a name-to-address resolution mechanism
be provided. This mechanism could be via DNS, <filename moreinfo="none">/etc/hosts</filename>, and so on.
</para>

<sect2 id="DMB">
<title>Configuring Workgroup Browsing</title>

<para>
<indexterm significance="normal"><primary>cross-subnet browsing</primary></indexterm>
<indexterm significance="normal"><primary>DMB</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>LMB</primary></indexterm>
<indexterm significance="normal"><primary>isolated workgroup</primary></indexterm>
<indexterm significance="normal"><primary>workgroup</primary></indexterm>
To configure cross-subnet browsing on a network containing machines in a workgroup, not an NT domain, you need
to set up one Samba server to be the DMB (note that this is not the same as a Primary Domain Controller,
although in an NT domain the same machine plays both roles). The role of a DMB is to collate the browse lists
from LMB on all the subnets that have a machine participating in the workgroup. Without one machine configured
as a DMB, each subnet would be an isolated workgroup unable to see any machines on another subnet. It is the
presence of a DMB that makes cross-subnet browsing possible for a workgroup.
</para>

<para>
<indexterm significance="normal"><primary>DMB</primary></indexterm>
In a workgroup environment the DMB must be a Samba server, and there must only be one DMB per workgroup name.
To set up a Samba server as a DMB, set the following option in the <parameter>[global]</parameter> section
of the <filename moreinfo="none">smb.conf</filename> file:
</para>

<para>
<simplelist>
<member><indexterm>domain master</indexterm><parameter>domain master = yes</parameter></member>
</simplelist>
</para>

<para>
<indexterm significance="normal"><primary>DMB</primary></indexterm>
<indexterm significance="normal"><primary>LMB</primary></indexterm>
The DMB should preferably be the LMB for its own subnet. In order to achieve this, set the following options
in the <parameter>[global]</parameter> section of the <filename moreinfo="none">smb.conf</filename> file as shown in <link linkend="dmbexample">Domain Master Browser smb.conf</link>
</para>

<example id="dmbexample">
<title>Domain Master Browser smb.conf</title>
<simplelist>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>domain master</indexterm><parameter>domain master = yes</parameter></member>
<member><indexterm>local master</indexterm><parameter>local master = yes</parameter></member>
<member><indexterm>preferred master</indexterm><parameter>preferred master = yes</parameter></member>
<member><indexterm>os level</indexterm><parameter>os level = 65</parameter></member>
</simplelist>
</example>

<para>
<indexterm significance="normal"><primary>DMB</primary></indexterm>
<indexterm significance="normal"><primary>WINS server</primary></indexterm>
The DMB may be the same machine as the WINS server, if necessary.
</para>

<para>
<indexterm significance="normal"><primary>subnets</primary></indexterm>
<indexterm significance="normal"><primary>LMB</primary></indexterm>
<indexterm significance="normal"><primary>rebooted</primary></indexterm>
Next, you should ensure that each of the subnets contains a machine that can act as an LMB for the workgroup.
Any MS Windows NT/200x/XP machine should be able to do this, as will Windows 9x/Me machines (although these
tend to get rebooted more often, so it is not such a good idea to use them). To make a Samba server an LMB,
set the following options in the <parameter>[global]</parameter> section of the <filename moreinfo="none">smb.conf</filename> file as shown in
<link linkend="lmbexample">Local master browser smb.conf</link>
</para>

<example id="lmbexample">
<title>Local master browser smb.conf</title>
<simplelist>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>domain master</indexterm><parameter>domain master = no</parameter></member>
<member><indexterm>local master</indexterm><parameter>local master = yes</parameter></member>
<member><indexterm>preferred master</indexterm><parameter>preferred master = yes</parameter></member>
<member><indexterm>os level</indexterm><parameter>os level = 65</parameter></member>
</simplelist>
</example>

<para>
<indexterm significance="normal"><primary>LMB</primary></indexterm>
Do not do this for more than one Samba server on each subnet, or they will war with
each other over which is to be the LMB.
</para>

<para>
<indexterm significance="normal"><primary>LMB</primary></indexterm>
<indexterm significance="normal"><primary>browser election</primary></indexterm>
The <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOCALMASTER" xlink:href="smb.conf.5.html#LOCALMASTER">local master</link> parameter allows Samba to act as a
LMB. The <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PREFERREDMASTER" xlink:href="smb.conf.5.html#PREFERREDMASTER">preferred master</link> causes <literal>nmbd</literal>
to force a browser election on startup and the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="OSLEVEL" xlink:href="smb.conf.5.html#OSLEVEL">os level</link>
parameter sets Samba high enough so it should win any browser elections.
</para>

<para>
<indexterm significance="normal"><primary>disable LMB</primary></indexterm>
If you have an NT machine on the subnet that you wish to be the LMB, you can disable Samba from
becoming an LMB by setting the following options in the <parameter>[global]</parameter> section of the
<filename moreinfo="none">smb.conf</filename> file as shown in <link linkend="nombexample">smb.conf for Not Being a Master Browser</link>.
</para>

<para>
<example id="nombexample">
<title>smb.conf for Not Being a Master Browser</title>
<simplelist>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>domain master</indexterm><parameter>domain master = no</parameter></member>
<member><indexterm>local master</indexterm><parameter>local master = no</parameter></member>
<member><indexterm>preferred master</indexterm><parameter>preferred master = no</parameter></member>
<member><indexterm>os level</indexterm><parameter>os level = 0</parameter></member>
</simplelist>
</example>
</para>

</sect2>

<sect2>
<title>Domain Browsing Configuration</title>

<para>
<indexterm significance="normal"><primary>DMB</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>registers</primary></indexterm>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
If you are adding Samba servers to a Windows NT domain, then you must not set up a Samba server as a DMB.  By
default, a Windows NT PDC for a domain is also the DMB for that domain. Network browsing may break if a Samba
server other than the PDC registers the DMB NetBIOS name (<replaceable>DOMAIN</replaceable>&lt;1B&gt;) with
WINS.
</para>

<para>
<indexterm significance="normal"><primary>Local Master Browser</primary></indexterm>
For subnets other than the one containing the Windows NT PDC, you may set up Samba servers as LMBs as
described. To make a Samba server a Local Master Browser, set the following options in the <parameter>[global]</parameter> section of the <filename moreinfo="none">smb.conf</filename> file as shown in <link linkend="remsmb">Local Master Browser
smb.conf</link>
</para>

<example id="remsmb">
<title>Local Master Browser smb.conf</title>
<simplelist>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>domain master</indexterm><parameter>domain master = no</parameter></member>
<member><indexterm>local master</indexterm><parameter>local master = yes</parameter></member>
<member><indexterm>preferred master</indexterm><parameter>preferred master = yes</parameter></member>
<member><indexterm>os level</indexterm><parameter>os level = 65</parameter></member>
</simplelist>
</example>

<para>
<indexterm significance="normal"><primary>election</primary></indexterm>
<indexterm significance="normal"><primary>LMB</primary></indexterm>
If you wish to have a Samba server fight the election with machines on the same subnet, you may set the
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="OSLEVEL" xlink:href="smb.conf.5.html#OSLEVEL">os level</link> parameter to lower levels.  By doing this you can tune the order of machines
that will become LMBs if they are running. For more details on this, refer to <link linkend="browse-force-master">Forcing Samba to Be the Master</link>.
</para>

<para>
<indexterm significance="normal"><primary>domain members</primary></indexterm>
<indexterm significance="normal"><primary>browser elections</primary></indexterm>
<indexterm significance="normal"><primary>LMB</primary></indexterm>
If you have Windows NT machines that are members of the domain on all subnets and you are sure they will
always be running, you can disable Samba from taking part in browser elections and ever becoming an LMB by
setting the following options in the <parameter>[global]</parameter> section of the <filename moreinfo="none">smb.conf</filename> file as shown
in <link linkend="xremmb"><filename moreinfo="none">smb.conf</filename> for Not Being a master browser</link>
</para>

<para>
<example id="xremmb">
<title><filename moreinfo="none">smb.conf</filename> for Not Being a master browser</title>
<parameter>[global]</parameter>
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="DOMAINMASTER" xlink:href="smb.conf.5.html#DOMAINMASTER">domain master = no</link>
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOCALMASTER" xlink:href="smb.conf.5.html#LOCALMASTER">local master = no</link>
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PREFERREDMASTER" xlink:href="smb.conf.5.html#PREFERREDMASTER">preferred master = no</link>
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="OSLEVEL" xlink:href="smb.conf.5.html#OSLEVEL">os level = 0</link>
</example>
</para>

</sect2>

<sect2 id="browse-force-master">
<title>Forcing Samba to Be the Master</title>

<para>
<indexterm significance="normal"><primary>master browser</primary></indexterm>
<indexterm significance="normal"><primary>election process</primary></indexterm>
<indexterm significance="normal"><primary>broadcasts</primary></indexterm>
<indexterm significance="normal"><primary>election packet</primary></indexterm>
<indexterm significance="normal"><primary>bias</primary></indexterm>
<indexterm significance="normal"><primary>election</primary></indexterm>
<indexterm significance="normal"><primary>precedence</primary></indexterm>
Who becomes the master browser is determined by an election process using broadcasts. Each election packet
contains a number of parameters that determine what precedence (bias) a host should have in the election. By
default Samba uses a low precedence and thus loses elections to just about every Windows network server or
client.
</para>

<para>
If you want Samba to win elections, set the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="OSLEVEL" xlink:href="smb.conf.5.html#OSLEVEL">os level</link> global option in <filename moreinfo="none">smb.conf</filename> to a
higher number. It defaults to 20. Using 34 would make it win all elections over every other system (except
other Samba systems).
</para>

<para>
An <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="OSLEVEL" xlink:href="smb.conf.5.html#OSLEVEL">os level</link> of two would make it beat Windows for Workgroups and Windows 9x/Me, but
not MS Windows NT/200x Server. An MS Windows NT/200x Server domain controller uses level 32. The maximum os
level is 255.
</para>

<para>
<indexterm significance="normal"><primary>force an election</primary></indexterm>
<indexterm significance="normal"><primary>potential master browsers</primary></indexterm>
<indexterm significance="normal"><primary>local subnet</primary></indexterm>
<indexterm significance="normal"><primary>LMB</primary></indexterm>
If you want Samba to force an election on startup, set the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PREFERREDMASTER" xlink:href="smb.conf.5.html#PREFERREDMASTER">preferred master</link> global
option in <filename moreinfo="none">smb.conf</filename> to <constant>yes</constant>.  Samba will then have a slight advantage over other
potential master browsers that are not preferred master browsers.  Use this parameter with care, because if
you have two hosts (whether they are Windows 9x/Me or NT/200x/XP or Samba) on the same local subnet both set
with <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PREFERREDMASTER" xlink:href="smb.conf.5.html#PREFERREDMASTER">preferred master</link> to <constant>yes</constant>, then periodically and continually
they will force an election in order to become the LMB.
</para>

<para>
<indexterm significance="normal"><primary>DMB</primary></indexterm>
<indexterm significance="normal"><primary>LAN</primary></indexterm>
<indexterm significance="normal"><primary>WAN</primary></indexterm>
<indexterm significance="normal"><primary>LMB</primary></indexterm>
<indexterm significance="normal"><primary>broadcast isolated subnet</primary></indexterm>
If you want Samba to be a <emphasis>DMB</emphasis>, then it is recommended that you also set <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PREFERREDMASTER" xlink:href="smb.conf.5.html#PREFERREDMASTER">preferred master</link> to <constant>yes</constant>, because Samba will not become a DMB for the whole of
your LAN or WAN if it is not also a LMB on its own broadcast isolated subnet.
</para>

<para>
<indexterm significance="normal"><primary>DMB</primary></indexterm>
<indexterm significance="normal"><primary>automatic redundancy</primary></indexterm>
<indexterm significance="normal"><primary>UDP</primary></indexterm>
<indexterm significance="normal"><primary>network bandwidth</primary></indexterm>
<indexterm significance="normal"><primary>browser elections</primary></indexterm>
It is possible to configure two Samba servers to attempt to become the DMB for a domain. The first server that
comes up will be the DMB. All other Samba servers will attempt to become the DMB every 5 minutes. They will
find that another Samba server is already the DMB and will fail. This provides automatic redundancy should the
current DMB fail. The network bandwidth overhead of browser elections is relatively small, requiring
approximately four UDP packets per machine per election. The maximum size of a UDP packet is 576 bytes.
</para>

</sect2>

<sect2>
<title>Making Samba the Domain Master</title>

<para>
<indexterm significance="normal"><primary>DMB</primary></indexterm>
<indexterm significance="normal"><primary>collating</primary></indexterm>
<indexterm significance="normal"><primary>browse lists</primary></indexterm>
<indexterm significance="normal"><primary>browsing</primary></indexterm>
The domain master browser is responsible for collating the browse lists of multiple subnets so browsing can
occur between subnets. You can make Samba act as the domain master browser by setting <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="DOMAINMASTER" xlink:href="smb.conf.5.html#DOMAINMASTER">domain master = yes</link> in <filename moreinfo="none">smb.conf</filename>. By default it will not be a domain master browser.
</para>

<para>
<indexterm significance="normal"><primary>workgroup</primary></indexterm>
<indexterm significance="normal"><primary>network browsing problems</primary></indexterm>
Do not set Samba to be the domain master for a workgroup that has the same name as an NT/200x domain.  If
Samba is configured to be the domain master for a workgroup that is present on the same network as a Windows
NT/200x domain that has the same name, network browsing problems will certainly be experienced.
</para>

<para>
When Samba is the domain master and the master browser, it will listen for master announcements (made roughly
every 12 minutes) from LMBs on other subnets and then contact them to synchronize browse lists.
</para>

<para>
<indexterm significance="normal"><primary>win election</primary></indexterm>
<indexterm significance="normal"><primary>force election</primary></indexterm>
If you want Samba to be the domain master, you should also set the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="OSLEVEL" xlink:href="smb.conf.5.html#OSLEVEL">os level</link> high
enough to make sure it wins elections, and set <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PREFERREDMASTER" xlink:href="smb.conf.5.html#PREFERREDMASTER">preferred master</link> to
<constant>yes</constant>, to get Samba to force an election on startup.
</para>

<para>
<indexterm significance="normal"><primary>WINS server</primary></indexterm>
<indexterm significance="normal"><primary>resolve NetBIOS names</primary></indexterm>
All servers (including Samba) and clients should be using a WINS server to resolve NetBIOS names. If your
clients are only using broadcasting to resolve NetBIOS names, then two things will occur:
</para>

<orderedlist inheritnum="ignore" continuation="restarts">
<listitem>
	<para>
<indexterm significance="normal"><primary>LMB</primary></indexterm>
<indexterm significance="normal"><primary>DMB</primary></indexterm>
	LMBs will be unable to find a DMB because they will be looking only on the local subnet.
	</para>
</listitem>

<listitem>
	<para>
<indexterm significance="normal"><primary>domain-wide browse list</primary></indexterm>
	If a client happens to get hold of a domain-wide browse list and a user attempts to access a
	host in that list, it will be unable to resolve the NetBIOS name of that host.
	</para>
</listitem>
</orderedlist>

<para>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
If, however, both Samba and your clients are using a WINS server, then:
</para>

<orderedlist inheritnum="ignore" continuation="restarts">
<listitem>
	<para>
	LMBs will contact the WINS server and, as long as Samba has registered that it is a DMB with the WINS
	server, the LMB will receive Samba's IP address as its DMB.
	</para>
</listitem>

<listitem>
	<para>
	When a client receives a domain-wide browse list and a user attempts to access a host in that list, it will
	contact the WINS server to resolve the NetBIOS name of that host. As long as that host has registered its
	NetBIOS name with the same WINS server, the user will be able to see that host..
	</para>
</listitem>
</orderedlist>

</sect2>

<sect2>
<title>Note about Broadcast Addresses</title>

<para>
<indexterm significance="normal"><primary>zero-based broadcast</primary></indexterm>
If your network uses a zero-based broadcast address (for example, if it ends in a 0), then you will strike
problems. Windows for Workgroups does not seem to support a zeros broadcast, and you will probably find that
browsing and name lookups will not work.
</para>
</sect2>

<sect2>
<title>Multiple Interfaces</title>

<para>
<indexterm significance="normal"><primary>multiple network interfaces</primary></indexterm>
Samba supports machines with multiple network interfaces. If you have multiple interfaces, you will
need to use the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="INTERFACES" xlink:href="smb.conf.5.html#INTERFACES">interfaces</link> option in <filename moreinfo="none">smb.conf</filename> to configure them. For example, the
machine you are working with has 4 network interfaces; <literal moreinfo="none">eth0</literal>, <literal moreinfo="none">eth1</literal>,
<literal moreinfo="none">eth2</literal>, <literal moreinfo="none">eth3</literal> and only interfaces <literal moreinfo="none">eth1</literal> and
<literal moreinfo="none">eth4</literal> should be used by Samba. In this case, the following <filename moreinfo="none">smb.conf</filename> file entries would
permit that intent:
<simplelist>
<member><indexterm>interfaces</indexterm><parameter>interfaces = eth1, eth4</parameter></member>
<member><indexterm>bind interfaces only</indexterm><parameter>bind interfaces only = Yes</parameter></member>
</simplelist>
<indexterm significance="normal"><primary>port 135</primary></indexterm>
<indexterm significance="normal"><primary>port 137</primary></indexterm>
<indexterm significance="normal"><primary>port 138</primary></indexterm>
<indexterm significance="normal"><primary>port 139</primary></indexterm>
<indexterm significance="normal"><primary>port 445</primary></indexterm>
<indexterm significance="normal"><primary>UDP</primary></indexterm>
<indexterm significance="normal"><primary>TCP</primary></indexterm>
The <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="BINDINTERFACESONLY" xlink:href="smb.conf.5.html#BINDINTERFACESONLY">bind interfaces only = Yes</link> is necessary to exclude TCP/IP session
services (ports 135, 139, and 445) over the interfaces that are not specified. Please be aware that
<literal>nmbd</literal> will listen for incoming UDP port 137 packets on the unlisted interfaces, but it will
not answer them. It will, however, send its broadcast packets over the unlisted interfaces. Total isolation of
ethernet interface requires the use of a firewall to block ports 137 and 138 (UDP), and ports 135, 139, and
445 (TCP) on all network interfaces that must not be able to access the Samba server.
</para>

</sect2>

<sect2>
<title>Use of the Remote Announce Parameter</title>
<para>
The <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="REMOTEANNOUNCE" xlink:href="smb.conf.5.html#REMOTEANNOUNCE">remote announce</link> parameter of <filename moreinfo="none">smb.conf</filename> can be used to forcibly ensure that all
the NetBIOS names on a network get announced to a remote network.  The syntax of the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="REMOTEANNOUNCE" xlink:href="smb.conf.5.html#REMOTEANNOUNCE">remote announce</link> parameter is:
<simplelist>
<member><indexterm>remote announce</indexterm><parameter>remote announce = 192.168.12.23 [172.16.21.255] ...</parameter></member>
</simplelist>
<emphasis>or</emphasis>
<simplelist>
<member><indexterm>remote announce</indexterm><parameter>remote announce = 192.168.12.23/MIDEARTH [172.16.21.255/ELVINDORF] ...</parameter></member>
</simplelist>

where:
<variablelist>
	<varlistentry><term><replaceable>192.168.12.23</replaceable> and <replaceable>172.16.21.255</replaceable></term>
		<listitem><para>
<indexterm significance="normal"><primary>LMB</primary><see>Local Master Browser</see></indexterm>
<indexterm significance="normal"><primary>Local Master Browser</primary></indexterm>
		is either the LMB IP address or the broadcast address of the remote network.
		That is, the LMB is at 192.168.1.23, or the address could be given as 172.16.21.255 where the netmask
		is assumed to be 24 bits (255.255.255.0). When the remote announcement is made to the broadcast
		address of the remote network, every host will receive our announcements. This is noisy and therefore
		undesirable but may be necessary if we do not know the IP address of the remote LMB.
		</para></listitem>
	</varlistentry>

	<varlistentry>
		<term><replaceable>WORKGROUP</replaceable></term>
		<listitem><para>is optional and can be either our own workgroup or that of the remote network. If you use the
		workgroup name of the remote network, our NetBIOS machine names will end up looking like
		they belong to that workgroup. This may cause name resolution problems and should be avoided.
		</para></listitem>
	</varlistentry>
</variablelist>
</para>

</sect2>

<sect2>
<title>Use of the Remote Browse Sync Parameter</title>

<para>
<indexterm significance="normal"><primary>LMB</primary></indexterm>
<indexterm significance="normal"><primary>synchronize</primary></indexterm>
The <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="REMOTEBROWSESYNC" xlink:href="smb.conf.5.html#REMOTEBROWSESYNC">remote browse sync</link> parameter of <filename moreinfo="none">smb.conf</filename> is used to announce to another LMB that
it must synchronize its NetBIOS name list with our Samba LMB. This works only if the Samba server that has
this option is simultaneously the LMB on its network segment.
</para>

<para>
The syntax of the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="REMOTEBROWSESYNC" xlink:href="smb.conf.5.html#REMOTEBROWSESYNC">remote browse sync</link> parameter is:

<simplelist>
<member><indexterm>remote browse sync</indexterm></member>
</simplelist>
<indexterm significance="normal"><primary>LMB</primary></indexterm>
<indexterm significance="normal"><primary>remote segment</primary></indexterm>
where <replaceable>192.168.10.40</replaceable> is either the IP address of the
remote LMB or the network broadcast address of the remote segment.
</para>

</sect2>

</sect1>

<sect1>
<title>WINS: The Windows Internetworking Name Server</title>

<para>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
<indexterm significance="normal"><primary>name_type</primary></indexterm>
<indexterm significance="normal"><primary>LanManager-compatible</primary></indexterm>
Use of WINS (either Samba WINS or MS Windows NT Server WINS) is highly
recommended. Every NetBIOS machine registers its name together with a
name_type value for each of several types of service it has available.
It registers its name directly as a unique (the type 0x03) name.
It also registers its name if it is running the LanManager-compatible
server service (used to make shares and printers available to other users)
by registering the server (the type 0x20) name.
</para>

<para>
<indexterm significance="normal"><primary>NetBIOS name length</primary></indexterm>
<indexterm significance="normal"><primary>name_type</primary></indexterm>
All NetBIOS names are up to 15 characters in length. The name_type variable
is added to the end of the name, thus creating a 16 character name. Any
name that is shorter than 15 characters is padded with spaces to the 15th
character. Thus, all NetBIOS names are 16 characters long (including the
name_type information).
</para>

<para>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
<indexterm significance="normal"><primary>registered</primary></indexterm>
<indexterm significance="normal"><primary>NetLogon service</primary></indexterm>
<indexterm significance="normal"><primary>lmhosts</primary></indexterm>
WINS can store these 16-character names as they get registered. A client
that wants to log onto the network can ask the WINS server for a list
of all names that have registered the NetLogon service name_type. This saves
broadcast traffic and greatly expedites logon processing. Since broadcast
name resolution cannot be used across network segments, this type of
information can only be provided via WINS or via a statically configured
<filename moreinfo="none">lmhosts</filename> file that must reside on all clients in the
absence of WINS.
</para>

<para>
<indexterm significance="normal"><primary>synchronization</primary></indexterm>
<indexterm significance="normal"><primary>LMB</primary></indexterm>
<indexterm significance="normal"><primary>DMB</primary></indexterm>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
<indexterm significance="normal"><primary>browse list</primary></indexterm>
WINS also forces browse list synchronization by all LMBs. LMBs must synchronize their browse list with the
DMB, and WINS helps the LMB to identify its DMB. By definition this will work only within a single workgroup.
Note that the DMB has nothing to do with what is referred to as an MS Windows NT domain. The latter is a
reference to a security environment, while the DMB refers to the master controller for browse list information
only.
</para>

<para>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
<indexterm significance="normal"><primary>TCP/IP protocol stack</primary></indexterm>
<indexterm significance="normal"><primary>WINS servers</primary></indexterm>
<indexterm significance="normal"><primary>name-to-address</primary></indexterm>
WINS will work correctly only if every client TCP/IP protocol stack
is configured to use the WINS servers. Any client that is not
configured to use the WINS server will continue to use only broadcast-based
name registration, so WINS may never get to know about it. In any case,
machines that have not registered with a WINS server will fail name-to-address
lookup attempts by other clients and will therefore cause workstation access
errors.
</para>

<para>
To configure Samba as a WINS server, just add
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="WINSSUPPORT" xlink:href="smb.conf.5.html#WINSSUPPORT">wins support = yes</link> to the <filename moreinfo="none">smb.conf</filename>
file [global] section.
</para>

<para>
To configure Samba to register with a WINS server, just add <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="WINSSERVER" xlink:href="smb.conf.5.html#WINSSERVER">wins server = 10.0.0.18</link> to your <filename moreinfo="none">smb.conf</filename> file <parameter>[global]</parameter> section.
</para>

<important><para>
Never use <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="WINSSUPPORT" xlink:href="smb.conf.5.html#WINSSUPPORT">wins support = yes</link> together with <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="WINSSERVER" xlink:href="smb.conf.5.html#WINSSERVER">wins server = 10.0.0.18</link> particularly not using its own IP address. Specifying both will cause <application moreinfo="none">nmbd</application>
to refuse to start!
</para></important>

<sect2>
<title>WINS Server Configuration</title>

<para>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
Either a Samba server or a Windows NT server machine may be set up
as a WINS server. To configure a Samba server to be a WINS server, you must
add to the <filename moreinfo="none">smb.conf</filename> file on the selected Server the following line to
the <parameter>[global]</parameter> section:
</para>

<para>
<simplelist>
<member><indexterm>wins support</indexterm><parameter>wins support = yes</parameter></member>
</simplelist>
</para>

<para>
<indexterm significance="normal"><primary>Samba 1.9.17</primary></indexterm>
Versions of Samba prior to 1.9.17 had this parameter default to
yes. If you have any older versions of Samba on your network, it is
strongly suggested you upgrade to a recent version, or at the very
least set the parameter to <quote>no</quote> on all these machines.
</para>

<para>
Machines configured with <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="WINSSUPPORT" xlink:href="smb.conf.5.html#WINSSUPPORT">wins support = yes</link> will keep a list of
all NetBIOS names registered with them, acting as a DNS for NetBIOS names.
</para>

<para>
<indexterm significance="normal"><primary>only one WINS server</primary></indexterm>
It is strongly recommended to set up only one WINS server. Do not set the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="WINSSUPPORT" xlink:href="smb.conf.5.html#WINSSUPPORT">wins support = yes</link> option on more than one Samba server on a network.
</para>

<para>
<indexterm significance="normal"><primary>replication</primary><secondary>WINS</secondary></indexterm>
<indexterm significance="normal"><primary>Windows NT/200x</primary></indexterm>
<indexterm significance="normal"><primary>WINS service</primary></indexterm>
<indexterm significance="normal"><primary>replication protocols</primary></indexterm>
<indexterm significance="normal"><primary>WINS server</primary></indexterm>
To configure Windows NT/200x Server as a WINS server, install and configure the WINS service. See the Windows
NT/200x documentation for details.  Windows NT/200x WINS servers can replicate to each other, allowing more
than one to be set up in a complex subnet environment. Because Microsoft refuses to document the replication
protocols, Samba cannot currently participate in these replications. It is possible that a Samba-to-Samba WINS
replication protocol may be defined in the future, in which case more than one Samba machine could be set up
as a WINS server.  Currently only one Samba server should have the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="WINSSUPPORT" xlink:href="smb.conf.5.html#WINSSUPPORT">wins support = yes</link> parameter set.
</para>

<para>
<indexterm significance="normal"><primary>WINS server</primary></indexterm>
<indexterm significance="normal"><primary>Primary WINS Server</primary></indexterm>
After the WINS server has been configured, you must ensure that all machines participating on the network are
configured with the address of this WINS server. If your WINS server is a Samba machine, fill in the Samba
machine IP address in the <guilabel moreinfo="none">Primary WINS Server</guilabel> field of the <guilabel moreinfo="none">Control
Panel-&gt;Network-&gt;Protocols-&gt;TCP-&gt;WINS Server</guilabel> dialogs in Windows 9x/Me or Windows NT/200x. To tell a
Samba server the IP address of the WINS server, add the following line to the <parameter>[global]</parameter> section of all <filename moreinfo="none">smb.conf</filename> files:
<simplelist>
<member><indexterm>wins server</indexterm><parameter>wins server = &lt;name or IP address&gt;</parameter></member>
</simplelist>
where &lt;name or IP address&gt; is either the DNS name of the WINS server
machine or its IP address.
</para>

<para>
This line must not be set in the <filename moreinfo="none">smb.conf</filename> file of the Samba
server acting as the WINS server itself. If you set both the
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="WINSSUPPORT" xlink:href="smb.conf.5.html#WINSSUPPORT">wins support = yes</link> option and the
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="WINSSERVER" xlink:href="smb.conf.5.html#WINSSERVER">wins server = &lt;name&gt;</link> option then
<literal>nmbd</literal> will fail to start.
</para>

<para>
<indexterm significance="normal"><primary>cross-subnet browsing</primary></indexterm>
<indexterm significance="normal"><primary>Windows 9x/Me</primary></indexterm>
<indexterm significance="normal"><primary>Windows NT/200x</primary></indexterm>
<indexterm significance="normal"><primary>not part of domain</primary></indexterm>
There are two possible scenarios for setting up cross-subnet browsing.
The first details setting up cross-subnet browsing on a network containing
Windows 9x/Me, Samba, and Windows NT/200x machines that are not configured as
part of a Windows NT domain. The second details setting up cross-subnet
browsing on networks that contain NT domains.
</para>

</sect2>

<sect2>
<title>WINS Replication</title>

<para>
<indexterm significance="normal"><primary>replication</primary><secondary>WINS</secondary></indexterm>
<indexterm significance="normal"><primary>WINS replication</primary></indexterm>
Samba-3 does not support native WINS replication. There was an approach to implement it, called
<filename moreinfo="none">wrepld</filename>, but it was never ready for action and the development is now discontinued.
</para>
<para>
Meanwhile, there is a project named <filename moreinfo="none">samba4WINS</filename>, which makes it possible to
run the Samba-4 WINS server parallel to Samba-3 since version 3.0.21. More information about
<filename moreinfo="none">samba4WINS</filename> are available at http://ftp.sernet.de/pub/samba4WINS.

</para>

</sect2>
<sect2>
<title>Static WINS Entries</title>

<para>
<indexterm significance="normal"><primary>static WINS entries</primary></indexterm>
<indexterm significance="normal"><primary>wins.dat</primary></indexterm>
<indexterm significance="normal"><primary>/usr/local/samba/var/locks</primary></indexterm>
<indexterm significance="normal"><primary>/var/run/samba</primary></indexterm>
Adding static entries to your Samba WINS server is actually fairly easy.  All you have to do is add a line to
<filename moreinfo="none">wins.dat</filename>, typically located in <filename class="directory" moreinfo="none">/usr/local/samba/var/locks</filename> or <filename moreinfo="none">/var/run/samba</filename>.
</para>

<para>
Entries in <filename moreinfo="none">wins.dat</filename> take the form of:
<programlisting format="linespecific">
"NAME#TYPE" TTL ADDRESS+ FLAGS
</programlisting>
<indexterm significance="normal"><primary>TTL</primary></indexterm>
<indexterm significance="normal"><primary>time-to-live</primary><see>TTL</see></indexterm>
where NAME is the NetBIOS name, TYPE is the NetBIOS type, TTL is the time-to-live as an absolute time in
seconds, ADDRESS+ is one or more addresses corresponding to the registration, and FLAGS are the NetBIOS flags
for the registration.
</para>

<note><para>
A change that has been made to the <filename moreinfo="none">wins.dat</filename> will not take effect until <application moreinfo="none">nmbd</application> has been
restarted. It should be noted that since the <filename moreinfo="none">wins.dat</filename> file changes dynamically, <application moreinfo="none">nmbd</application>
should be stopped before editting this file. Do not forget to restart <application moreinfo="none">nmbd</application> when this file has been editted.
</para></note>

<para>
A typical dynamic entry looks like this:
<programlisting format="linespecific">
"MADMAN#03" 1155298378 192.168.1.2 66R
</programlisting>
To make a NetBIOS name static (permanent), simply set the TTL to 0, like this:
<programlisting format="linespecific">
"MADMAN#03" 0 192.168.1.2 66R
</programlisting>
</para>

<para>
<indexterm significance="normal"><primary>NetBIOS flags</primary></indexterm>
<indexterm significance="normal"><primary>Broadcast node</primary></indexterm>
<indexterm significance="normal"><primary>Peer node</primary></indexterm>
<indexterm significance="normal"><primary>Meta node</primary></indexterm>
<indexterm significance="normal"><primary>Hybrid node</primary></indexterm>
<indexterm significance="normal"><primary>Permanent name</primary></indexterm>
<indexterm significance="normal"><primary>nameserv.h</primary></indexterm>
The NetBIOS flags may be interpreted as additive hexadecimal values: 00 - Broadcast node registration, 20 -
Peer node registration, 40 - Meta node registration, 60 - Hybrid node registration, 02 - Permanent name, 04 -
Active name, 80 - Group name. The 'R' indicates this is a registration record. Thus 66R means: Hybrid node
active and permanent NetBIOS name. These values may be found in the <filename moreinfo="none">nameserv.h</filename> header
file from the Samba source code repository. These are the values for the NB flags.
</para>

<para>
<indexterm significance="normal"><primary>WINS replication</primary></indexterm>
Though this method works with early Samba-3 versions, there is a possibility that it may change in future
versions if WINS replication is added.
</para>

</sect2>
</sect1>

<sect1>
<title>Helpful Hints</title>

<para>
The following hints should be carefully considered because they are stumbling points
for many new network administrators.
</para>

<sect2>
<title>Windows Networking Protocols</title>

<para>
<indexterm significance="normal"><primary>browsing problems</primary></indexterm>
<indexterm significance="normal"><primary>more than one protocol</primary></indexterm>
A common cause of browsing problems results from the installation of more than one protocol on an MS Windows
machine.
</para>

<warning><para>
Do not use more than one protocol on MS Windows clients.
</para></warning>

<para>
<indexterm significance="normal"><primary>LMB</primary></indexterm>
<indexterm significance="normal"><primary>DMB</primary></indexterm>
Every NetBIOS machine takes part in a process of electing the LMB (and DMB)
every 15 minutes. A set of election criteria is used to determine the order
of precedence for winning this election process. A machine running Samba or
Windows NT will be biased, so the most suitable machine will predictably
win and thus retain its role.
</para>

<para>
<indexterm significance="normal"><primary>NetBIOS network interface</primary></indexterm>
<indexterm significance="normal"><primary>TCP/IP</primary></indexterm>
<indexterm significance="normal"><primary>IPX</primary></indexterm>
<indexterm significance="normal"><primary>LMB</primary></indexterm>
<indexterm significance="normal"><primary>Windows 9x/Me</primary></indexterm>
<indexterm significance="normal"><primary>TCP/IP-only</primary></indexterm>
The election process is <emphasis>fought out, so to speak</emphasis> over every NetBIOS network interface. In
the case of a Windows 9x/Me machine that has both TCP/IP and IPX installed and has NetBIOS enabled over both
protocols, the election will be decided over both protocols. As often happens, if the Windows 9x/Me machine is
the only one with both protocols, then the LMB may be won on the NetBIOS interface over the IPX protocol.
Samba will then lose the LMB role because Windows 9x/Me will insist it knows who the LMB is. Samba will then
cease to function as an LMB, and browse list operation on all TCP/IP-only machines will therefore fail.
</para>

<para>
<indexterm significance="normal"><primary>Windows 9x/Me</primary></indexterm>
<indexterm significance="normal"><primary>extended protocol</primary></indexterm>
Windows 95, 98, 98se, and Me are referred to generically as Windows 9x/Me.  The Windows NT4, 200x, and XP use
common protocols. These are roughly referred to as the Windows NT family, but it should be recognized that
2000 and XP/2003 introduce new protocol extensions that cause them to behave differently from MS Windows NT4.
Generally, where a server does not support the newer or extended protocol, these will fall back to the NT4
protocols.
</para>

<para>
The safest rule of all to follow is: Use only one protocol!
</para>

</sect2>

<sect2>
<title>Name Resolution Order</title>

<para>
<indexterm significance="normal"><primary>NetBIOS names</primary></indexterm>
<indexterm significance="normal"><primary>name_type</primary></indexterm>
Resolution of NetBIOS names to IP addresses can take place using a number
of methods. The only ones that can provide NetBIOS name_type information
are:
</para>

<itemizedlist>
	<listitem><para>WINS <?latex --- ?> the best tool.</para></listitem>
	<listitem><para>LMHOSTS <?latex --- ?> static and hard to maintain.</para></listitem>
	<listitem><para>Broadcast <?latex --- ?> uses UDP and cannot resolve names across remote segments.</para></listitem>
</itemizedlist>

<para>
Alternative means of name resolution include:
</para>
<itemizedlist>
<listitem><para>Static <filename moreinfo="none">/etc/hosts</filename> <?latex --- ?>  hard to maintain and lacks name_type info.</para></listitem>
<listitem><para>DNS <?latex --- ?> is a good choice but lacks essential NetBIOS name_type information.</para></listitem>
</itemizedlist>

<para>
<indexterm significance="normal"><primary>restrict DNS</primary></indexterm>
<indexterm significance="normal"><primary>name resolve order</primary></indexterm>
Many sites want to restrict DNS lookups and avoid broadcast name
resolution traffic. The <parameter moreinfo="none">name resolve order</parameter> parameter is of great help here.
The syntax of the <parameter moreinfo="none">name resolve order</parameter> parameter is:
<simplelist>
<member><indexterm>name resolve order</indexterm><parameter>name resolve order = wins lmhosts bcast host</parameter></member>
</simplelist>
<emphasis>or</emphasis>
<simplelist>
<member><indexterm>name resolve order</indexterm><parameter>name resolve order = wins lmhosts  	(eliminates bcast and host)</parameter></member>
</simplelist>
The default is:
<simplelist>
<member><indexterm>name resolve order</indexterm><parameter>name resolve order = host lmhost wins bcast</parameter></member>,
</simplelist>
<indexterm significance="normal"><primary>gethostbyname() function call</primary></indexterm>
where <quote>host</quote> refers to the native methods used by the UNIX system to implement the
gethostbyname() function call. This is normally controlled by <filename moreinfo="none">/etc/host.conf</filename>,
<filename moreinfo="none">/etc/nsswitch.conf</filename> and <filename moreinfo="none">/etc/resolv.conf</filename>.
</para>
</sect2>
</sect1>

<sect1>
<title>Technical Overview of Browsing</title>

<para>
<indexterm significance="normal"><primary>SMB</primary></indexterm>
SMB networking provides a mechanism by which clients can access a list
of machines in a network called <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="BROWSELIST" xlink:href="smb.conf.5.html#BROWSELIST">browse list</link>. This list
contains machines that are ready to offer file and/or print services
to other machines within the network. It therefore does not include
machines that aren't currently able to do server tasks. The browse
list is heavily used by all SMB clients. Configuration of SMB
browsing has been problematic for some Samba users, hence this
document.
</para>

<para>
<indexterm significance="normal"><primary>NetBIOS over TCP/IP</primary></indexterm>
<indexterm significance="normal"><primary>DNS/LDAP/ADS</primary></indexterm>
<indexterm significance="normal"><primary>name resolution</primary></indexterm>
MS Windows 2000 and later versions, as with Samba-3 and later versions, can be
configured to not use NetBIOS over TCP/IP. When configured this way,
it is imperative that name resolution (using DNS/LDAP/ADS) be correctly
configured and operative. Browsing will not work if name resolution
from SMB machine names to IP addresses does not function correctly.
</para>

<para>
<indexterm significance="normal"><primary>NetBIOS</primary></indexterm>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
Where NetBIOS over TCP/IP is enabled, use of a WINS server is highly
recommended to aid the resolution of NetBIOS (SMB) names to IP addresses.
WINS allows remote segment clients to obtain NetBIOS name_type information
that cannot be provided by any other means of name resolution.
</para>

<sect2>
<title>Browsing Support in Samba</title>

<para>
<indexterm significance="normal"><primary>browsing</primary></indexterm>
<indexterm significance="normal"><primary>LMB</primary></indexterm>
<indexterm significance="normal"><primary>domain logons</primary></indexterm>
<indexterm significance="normal"><primary>scripts</primary></indexterm>
Samba facilitates browsing. The browsing is supported by <application moreinfo="none">nmbd</application>
and is also controlled by options in the <filename moreinfo="none">smb.conf</filename> file.
Samba can act as an LMB for a workgroup, and the ability
to support domain logons and scripts is now available.
</para>

<para>
<indexterm significance="normal"><primary>DMB for a workgroup</primary></indexterm>
<indexterm significance="normal"><primary>LMB</primary></indexterm>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
Samba can also act as a DMB for a workgroup. This
means that it will collate lists from LMBs into a
wide-area network server list. In order for browse clients to
resolve the names they may find in this list, it is recommended that
both Samba and your clients use a WINS server.
</para>

<para>
<indexterm significance="normal"><primary>domain master</primary></indexterm>
Do not set Samba to be the domain master for a workgroup that has the same
name as an NT Domain. On each wide-area network, you must only ever have one
DMB per workgroup, regardless of whether it is NT, Samba,
or any other type of domain master that is providing this service.
</para>

<note><para>
<indexterm significance="normal"><primary>nmbd</primary></indexterm>
<indexterm significance="normal"><primary>WINS server</primary></indexterm>
<literal>nmbd</literal> can be configured as a WINS server, but it is not
necessary to specifically use Samba as your WINS server. MS Windows
NT4, Server or Advanced Server 200x can be configured as
your WINS server. In a mixed NT/200x server and Samba environment on
a WAN, it is recommended that you use the Microsoft
WINS server capabilities. In a Samba-only environment, it is
recommended that you use one and only one Samba server as the WINS server.
</para></note>

<para>
<indexterm significance="normal"><primary>nmbd</primary></indexterm>
To get browsing to work, you need to run <literal>nmbd</literal> as usual, but must
use the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="WORKGROUP" xlink:href="smb.conf.5.html#WORKGROUP">workgroup</link> option in <filename moreinfo="none">smb.conf</filename>
to control what workgroup Samba becomes a part of.
</para>

<para>
<indexterm significance="normal"><primary>browsing another subnet</primary></indexterm>
Samba also has a useful option for a Samba server to offer itself for browsing on another subnet. It is
recommended that this option is used only for <quote>unusual</quote> purposes: announcements over the
Internet, for example. See <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="REMOTEANNOUNCE" xlink:href="smb.conf.5.html#REMOTEANNOUNCE">remote announce</link> in the <filename moreinfo="none">smb.conf</filename> man page.
</para>
</sect2>

<sect2>
<title>Problem Resolution</title>

<para>
<indexterm significance="normal"><primary>log.nmbd</primary></indexterm>
<indexterm significance="normal"><primary>browse.dat</primary></indexterm>
If something does not work, the <filename moreinfo="none">log.nmbd</filename> file will help
to track down the problem. Try a <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOGLEVEL" xlink:href="smb.conf.5.html#LOGLEVEL">log level</link> of 2 or 3 for finding
problems. Also note that the current browse list usually gets stored
in text form in a file called <filename moreinfo="none">browse.dat</filename>.
</para>

<para>
<indexterm significance="normal"><primary>\\SERVER</primary></indexterm>
<indexterm significance="normal"><primary>filemanager</primary></indexterm>
If it does not work, you should still be able to
type the server name as <filename moreinfo="none">\\SERVER</filename> in <literal>filemanager</literal>, then
press enter, and <literal>filemanager</literal> should display the list of available shares.
</para>

<para>
<indexterm significance="normal"><primary>IPC$</primary></indexterm>
<indexterm significance="normal"><primary>guest account</primary></indexterm>
Some people find browsing fails because they do not have the global
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="GUESTACCOUNT" xlink:href="smb.conf.5.html#GUESTACCOUNT">guest account</link> set to a valid account. Remember that the
IPC$ connection that lists the shares is done as guest and so you must have a valid guest account.
</para>

<note><para>
<indexterm significance="normal"><primary>IPC$</primary></indexterm>
<indexterm significance="normal"><primary>Windows Explorer</primary></indexterm>
<indexterm significance="normal"><primary>browse resources</primary></indexterm>
<indexterm significance="normal"><primary>Network Neighborhood</primary></indexterm>
<indexterm significance="normal"><primary>My Network Places</primary></indexterm>
The <literal moreinfo="none">IPC$</literal> share is used by all SMB/CIFS clients to obtain the list of resources that is
available on the server. This is the source of the list of shares and printers when browsing an SMB/CIFS
server (also Windows machines) using the Windows Explorer to browse resources through the Windows Network
Neighborhood (also called My Network Places) through to a Windows server. At this point, the client has opened
a connection to the <literal moreinfo="none">\\server\IPC4</literal> resource.  Clicking on a share will then open up a
connection to the <literal moreinfo="none">\\server\share</literal>.
</para></note>

<para>
<indexterm significance="normal"><primary>guest account</primary></indexterm>
<indexterm significance="normal"><primary>anonymous access</primary></indexterm>
<indexterm significance="normal"><primary>IPC$</primary></indexterm>
<indexterm significance="normal"><primary>browse server resources</primary></indexterm>
MS Windows 2000 and later (as with Samba) can be configured to disallow
anonymous (i.e., guest account) access to the IPC$ share. In that case, the
MS Windows 2000/XP/2003 machine acting as an SMB/CIFS client will use the
name of the currently logged-in user to query the IPC$ share. MS Windows
9x/Me clients are not able to do this and thus will not be able to browse
server resources.
</para>

<para>
<indexterm significance="normal"><primary>broadcast address</primary></indexterm>
The other big problem people have is that their broadcast address,
netmask, or IP address is wrong (specified with the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="INTERFACES" xlink:href="smb.conf.5.html#INTERFACES">interfaces</link> option
in <filename moreinfo="none">smb.conf</filename>)
</para>
</sect2>

<sect2>
<title>Cross-Subnet Browsing</title>

<para>
<indexterm significance="normal"><primary>replication</primary><secondary>browse lists</secondary></indexterm>
<indexterm significance="normal"><primary>browse across subnet</primary></indexterm>
Since the release of Samba 1.9.17 (alpha1), Samba has supported the replication of browse lists across subnet
boundaries. This section describes how to set this feature up in different settings.
</para>

<para>
<indexterm significance="normal"><primary>browse lists</primary></indexterm>
<indexterm significance="normal"><primary>broadcast traffic</primary></indexterm>
<indexterm significance="normal"><primary>UDP</primary></indexterm>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
<indexterm significance="normal"><primary>remote announce</primary></indexterm>
<indexterm significance="normal"><primary>remote browse sync</primary></indexterm>
To see browse lists that span TCP/IP subnets (i.e., networks separated by routers that do not pass broadcast
traffic), you must set up at least one WINS server. The WINS server acts as a DNS for NetBIOS names. This will
allow NetBIOS name-to-IP address translation to be completed by a direct query of the WINS server. This is
done via a directed UDP packet on port 137 to the WINS server machine. The WINS server avoids the necessity of
default NetBIOS name-to-IP address translation, which is done using UDP broadcasts from the querying machine.
This means that machines on one subnet will not be able to resolve the names of machines on another subnet
without using a WINS server. The Samba hacks, <parameter moreinfo="none">remote browse sync</parameter>, and <parameter moreinfo="none">remote
announce</parameter> are designed to get around the natural limitations that prevent UDP broadcast
propagation. The hacks are not a universal solution and they should not be used in place of WINS, they are
considered last resort methods.
</para>

<para>
<indexterm significance="normal"><primary>DHCP</primary></indexterm>
<indexterm significance="normal"><primary>browsing across subnets</primary></indexterm>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
<indexterm significance="normal"><primary>Network settings</primary></indexterm>
Remember, for browsing across subnets to work correctly, all machines, be they Windows 95, Windows NT, or
Samba servers, must have the IP address of a WINS server given to them by a DHCP server or by manual
configuration: for Windows 9x/Me and Windows NT/200x/XP, this is in the TCP/IP Properties, under Network
settings; for Samba, this is in the <filename moreinfo="none">smb.conf</filename> file.
</para>

<para>
<indexterm significance="normal"><primary>NetBIOS over TCP/IP</primary></indexterm>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
<indexterm significance="normal"><primary>DNS</primary></indexterm>
It is possible to operate Samba-3 without NetBIOS over TCP/IP. If you do this, be warned that if used outside
of MS ADS, this will forgo network browsing support. ADS permits network browsing support through DNS,
providing appropriate DNS records are inserted for all Samba servers.
</para>

<sect3>
<title>Behavior of Cross-Subnet Browsing</title>

<para>
<indexterm significance="normal"><primary>cross-subnet browsing</primary></indexterm>
<indexterm significance="normal"><primary>complicated</primary></indexterm>
Cross-subnet browsing is a complicated dance, containing multiple moving parts. It has taken Microsoft several
years to get the code that correctly achieves this, and Samba lags behind in some areas.  Samba is capable of
cross-subnet browsing when configured correctly.
</para>

<para>
Consider a network set up as in <link linkend="browsing1">Cross-Subnet Browsing Example</link>.
</para>

<figure id="browsing1" float="0">
	<title>Cross-Subnet Browsing Example.</title>
	<mediaobject><imageobject role="html"><imagedata fileref="images/browsing1.png" scale="40" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/browsing1.png" scale="40" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/browsing1"/></imageobject></mediaobject>
</figure>

<para>
<indexterm significance="normal"><primary>broadcasts</primary></indexterm>
<indexterm significance="normal"><primary>DMB</primary></indexterm>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
This consists of three subnets (1, 2, 3) connected by two routers (R1, R2), which do not pass broadcasts.
Subnet 1 has five machines on it, subnet 2 has four machines, and subnet 3 has four machines. Assume for the
moment that all machines are configured to be in the same workgroup (for simplicity's sake). Machine N1_C on
subnet 1 is configured as the DMB (i.e., it will collate the browse lists for the workgroup). Machine N2_D is
configured as a WINS server, and all the other machines are configured to register their NetBIOS names with
it.
</para>

<para>
<indexterm significance="normal"><primary>master browsers</primary></indexterm>
<indexterm significance="normal"><primary>LMB</primary></indexterm>
<indexterm significance="normal"><primary>DMB</primary></indexterm>
As these machines are booted up, elections for master browsers
take place on each of the three subnets. Assume that machine
N1_C wins on subnet 1, N2_B wins on subnet 2, and N3_D wins on
subnet 3. These machines are known as LMBs for
their particular subnet. N1_C has an advantage in winning as the
LMB on subnet 1 because it is set up as DMB.
</para>

<para>
<indexterm significance="normal"><primary>LMB</primary></indexterm>
<indexterm significance="normal"><primary>browse list</primary></indexterm>
On each of the three networks, machines that are configured to offer sharing services will broadcast that they
are offering these services. The LMB on each subnet will receive these broadcasts and keep a record of the
fact that the machine is offering a service. This list of records is the basis of the browse list. For this
case, assume that all the machines are configured to offer services, so all machines will be on the browse
list.
</para>

<para>
<indexterm significance="normal"><primary>LMB</primary></indexterm>
<indexterm significance="normal"><primary>authoritative</primary></indexterm>
<indexterm significance="normal"><primary>verifiable</primary></indexterm>
<indexterm significance="normal"><primary>trusted</primary></indexterm>
<indexterm significance="normal"><primary>non-authoritative</primary></indexterm>
For each network, the LMB on that network is
considered <emphasis>authoritative</emphasis> for all the names it receives via
local broadcast. This is because a machine seen by the LMB
via a local broadcast must be on the same network as the
Local Master Browser and thus is a <emphasis>trusted</emphasis>
and <emphasis>verifiable</emphasis> resource. Machines on other networks that
the LMBs learn about when collating their
browse lists have not been directly seen. These records are
called <emphasis>non-authoritative.</emphasis>
</para>

<para>
<indexterm significance="normal"><primary>network neighborhood</primary></indexterm>
At this point the browse lists appear as shown in <link linkend="browsubnet">Browse Subnet Example 1</link>
(these are the machines you would see in your network neighborhood if you looked in it on a particular network
right now).
</para>

<para>
<table frame="all" id="browsubnet">
	<title>Browse Subnet Example 1</title>
	<tgroup align="left" cols="3">
	<thead>
		<row><entry>Subnet</entry><entry>Browse Master</entry><entry>List</entry></row>
	</thead>

	<tbody>
		<row><entry>Subnet1</entry><entry>N1_C</entry><entry>N1_A, N1_B, N1_C, N1_D, N1_E</entry></row>
		<row><entry>Subnet2</entry><entry>N2_B</entry><entry>N2_A, N2_B, N2_C, N2_D</entry></row>
		<row><entry>Subnet3</entry><entry>N3_D</entry><entry>N3_A, N3_B, N3_C, N3_D</entry></row>
	</tbody>
	</tgroup>
</table>
</para>

<para>
At this point all the subnets are separate, and no machine is seen across any of the subnets.
</para>

<para>
<indexterm significance="normal"><primary>DMB</primary></indexterm>
<indexterm significance="normal"><primary>LMB</primary></indexterm>
<indexterm significance="normal"><primary>synchronize</primary></indexterm>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
Now examine subnet 2 in <link linkend="brsbex">Browse Subnet Example 2</link>.  As soon as N2_B has become the
LMB, it looks for a DMB with which to synchronize its browse list. It does this by querying the WINS server
(N2_D) for the IP address associated with the NetBIOS name WORKGROUP&lt;1B&gt;. This name was registered by
the DMB (N1_C) with the WINS server as soon as it was started.
</para>

<para>
<indexterm significance="normal"><primary>MasterAnnouncement</primary></indexterm>
<indexterm significance="normal"><primary>NetServerEnum2</primary></indexterm>
<indexterm significance="normal"><primary>synchronization</primary></indexterm>
<indexterm significance="normal"><primary>browse lists</primary></indexterm>
Once N2_B knows the address of the DMB, it tells the DMB that it is the LMB
for subnet 2 by sending the DMB a
<emphasis>MasterAnnouncement</emphasis> packet to UDP port 138.  It then
synchronizes with the DMB by
doing a <emphasis>NetServerEnum2</emphasis> call. This tells the DMB to
send the sender all the server names it knows
about. Once the DMB receives the <emphasis>MasterAnnouncement</emphasis> packet, it schedules a
synchronization request to the sender of that packet. After both synchronizations are complete, the browse
lists look like those in <link linkend="brsbex">Browse Subnet Example 2</link>
</para>

<table frame="all" id="brsbex">
	<title>Browse Subnet Example 2</title>
	<tgroup cols="3">
		<colspec align="left"/>
		<colspec align="left"/>
		<colspec align="justify" colwidth="1*"/>
	<thead>
		<row><entry>Subnet</entry><entry>Browse Master</entry><entry>List</entry></row>
	</thead>

	<tbody>
		<row><entry>Subnet1</entry><entry>N1_C</entry><entry>N1_A, N1_B, N1_C, N1_D, N1_E,
N2_A(*), N2_B(*), N2_C(*), N2_D(*)</entry></row>
		<row><entry>Subnet2</entry><entry>N2_B</entry><entry>N2_A, N2_B, N2_C, N2_D, N1_A(*),
N1_B(*), N1_C(*), N1_D(*), N1_E(*)</entry></row>
		<row><entry>Subnet3</entry><entry>N3_D</entry><entry>N3_A, N3_B, N3_C, N3_D</entry></row>
	</tbody>
	</tgroup>
</table>

<para>
<indexterm significance="normal"><primary>non-authoritative</primary></indexterm>
Servers with an (*) after them are non-authoritative names.
</para>

<para>
<indexterm significance="normal"><primary>Network Neighborhood</primary></indexterm>
At this point users looking in their Network Neighborhood on subnets 1 or 2 will see all the servers on both;
users on subnet 3 will still see only the servers on their own subnet.
</para>

<para>
<indexterm significance="normal"><primary>DMB</primary></indexterm>
The same sequence of events that occurred for N2_B now occurs for the LMB on subnet 3 (N3_D). When it
synchronizes browse lists with the DMB (N1_A) it gets both the server entries on subnet 1 and those on subnet
2. After N3_D has synchronized with N1_C and vica versa, the browse lists will appear as shown in <link linkend="brsex2">Browse Subnet Example 3</link>
</para>

<table frame="all" id="brsex2">
	<title>Browse Subnet Example 3</title>		
	<tgroup cols="3" align="left">
		<colspec align="left"/>
		<colspec align="left"/>
		<colspec align="justify" colwidth="1*"/>

	<thead>
		<row><entry>Subnet</entry><entry>Browse Master</entry><entry>List</entry></row>
	</thead>

	<tbody>
		<row><entry>Subnet1</entry><entry>N1_C</entry><entry>N1_A, N1_B, N1_C, N1_D, N1_E,
N2_A(*), N2_B(*), N2_C(*), N2_D(*), N3_A(*), N3_B(*), N3_C(*), N3_D(*)</entry></row>
		<row><entry>Subnet2</entry><entry>N2_B</entry><entry>N2_A, N2_B, N2_C, N2_D, N1_A(*),
N1_B(*), N1_C(*), N1_D(*), N1_E(*)</entry></row>
		<row><entry>Subnet3</entry><entry>N3_D</entry><entry>N3_A, N3_B, N3_C, N3_D, N1_A(*),
N1_B(*), N1_C(*), N1_D(*), N1_E(*), N2_A(*), N2_B(*), N2_C(*), N2_D(*)</entry></row>
	</tbody>
	</tgroup>
</table>

<para>
Servers with an (*) after them are non-authoritative names.
</para>

<para>
At this point, users looking in their Network Neighborhood on
subnets 1 or 3 will see all the servers on all subnets, while users on
subnet 2 will still see only the servers on subnets 1 and 2, but not 3.
</para>

<para>
<indexterm significance="normal"><primary>LMB</primary></indexterm>
<indexterm significance="normal"><primary>DMB</primary></indexterm>
<indexterm significance="normal"><primary>browse lists</primary></indexterm>
Finally, the LMB for subnet 2 (N2_B) will sync again
with the DMB (N1_C) and will receive the missing
server entries. Finally, as when a steady state (if no machines
are removed or shut off) has been achieved, the browse lists will appear
as shown in <link linkend="brsex3">Browse Subnet Example 4</link>.
</para>

<table frame="all" id="brsex3">
	<title>Browse Subnet Example 4</title>
	<tgroup cols="3" align="left">
		<colspec align="left"/>
		<colspec align="left"/>
		<colspec align="justify" colwidth="1*"/>

	<thead>
		<row><entry>Subnet</entry><entry>Browse Master</entry><entry>List</entry></row>
	</thead>

	<tbody>
		<row><entry>Subnet1</entry><entry>N1_C</entry><entry>N1_A, N1_B, N1_C, N1_D, N1_E,
N2_A(*), N2_B(*), N2_C(*), N2_D(*), N3_A(*), N3_B(*),
N3_C(*), N3_D(*)</entry></row>
		<row><entry>Subnet2</entry><entry>N2_B</entry><entry>N2_A, N2_B, N2_C, N2_D, N1_A(*),
N1_B(*), N1_C(*), N1_D(*), N1_E(*), N3_A(*), N3_B(*),
N3_C(*), N3_D(*)</entry></row>
		<row><entry>Subnet3</entry><entry>N3_D</entry><entry>N3_A, N3_B, N3_C, N3_D, N1_A(*),
N1_B(*), N1_C(*), N1_D(*), N1_E(*), N2_A(*), N2_B(*),
N2_C(*), N2_D(*)</entry></row>
	</tbody>
	</tgroup>
</table>

<para>
Servers with an (*) after them are non-authoritative names.
</para>

<para>
Synchronizations between the DMB and LMBs
will continue to occur, but this should remain a
steady-state operation.
</para>

<para>
If either router R1 or R2 fails, the following will occur:
</para>

<orderedlist inheritnum="ignore" continuation="restarts">
<listitem>
	<para>
<indexterm significance="normal"><primary>Network Neighborhood</primary></indexterm>
	Names of computers on each side of the inaccessible network fragments
	will be maintained for as long as 36 minutes in the Network Neighborhood
	lists.
	</para>
</listitem>

<listitem>
	<para>
	Attempts to connect to these inaccessible computers will fail, but the
	names will not be removed from the Network Neighborhood lists.
	</para>
</listitem>

<listitem>
	<para>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
<indexterm significance="normal"><primary>NetBIOS name resolution</primary></indexterm>
<indexterm significance="normal"><primary>DNS server</primary></indexterm>
	If one of the fragments is cut off from the WINS server, it will only
	be able to access servers on its local subnet using subnet-isolated
	broadcast NetBIOS name resolution. The effect is similar to that of
	losing access to a DNS server.
	</para>
</listitem>
</orderedlist>
</sect3>
</sect2>
</sect1>

<sect1>
<title>Common Errors</title>

<para>
<indexterm significance="normal"><primary>browsing problems</primary></indexterm>
<indexterm significance="normal"><primary>name resolution</primary></indexterm>
Many questions are asked on the mailing lists regarding browsing. The majority of browsing
problems originate from incorrect configuration of NetBIOS name resolution. Some are of
particular note.
</para>

<sect2>
<title>Flushing the Samba NetBIOS Name Cache</title>

<para>
How Can One Flush the Samba NetBIOS Name Cache without Restarting Samba?
</para>

<para>
<indexterm significance="normal"><primary>flush name cache</primary></indexterm>
<indexterm significance="normal"><primary>nmbd</primary></indexterm>
<indexterm significance="normal"><primary>NetBIOS name cache</primary></indexterm>
<indexterm significance="normal"><primary>rogue machine</primary></indexterm>
Samba's <literal>nmbd</literal> process controls all browse list handling. Under normal circumstances it is
safe to restart <literal>nmbd</literal>. This will effectively flush the Samba NetBIOS name cache and cause it
to be rebuilt. This does not make certain that a rogue machine name will not reappear
in the browse list. When <literal>nmbd</literal> is taken out of service, another machine on the network will
become the browse master. This new list may still have the rogue entry in it. If you really
want to clear a rogue machine from the list, every machine on the network must be
shut down and restarted after all machines are down. Failing a complete restart, the only
other thing you can do is wait until the entry times out and is then flushed from the list.
This may take a long time on some networks (perhaps months).
</para>

</sect2>

<sect2>
	<title>Server Resources Cannot Be Listed</title>

<para><quote>My Client Reports "<quote>This server is not configured to list shared resources."</quote></quote></para>


<para>
Your guest account is probably invalid for some reason. Samba uses the
guest account for browsing in <literal>smbd</literal>. Check that your guest account is
valid.
</para>

<para>Also see <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="GUESTACCOUNT" xlink:href="smb.conf.5.html#GUESTACCOUNT">guest account</link> in the <filename moreinfo="none">smb.conf</filename> man page.</para>

</sect2>

<sect2>
	<title>I Get an "<errorname>Unable to browse the network</errorname>" Error</title>

	<para>This error can have multiple causes:
<indexterm significance="normal"><primary>browsing problems</primary></indexterm>
	</para>

	<itemizedlist>
		<listitem><para>There is no LMB. Configure <application moreinfo="none">nmbd</application>
			or any other machine to serve as LMB.</para></listitem>
		<listitem><para>You cannot log onto the machine that is the LMB.
			Can you log on to it as a guest user? </para></listitem>
		<listitem><para>There is no IP connectivity to the LMB.
			Can you reach it by broadcast?</para></listitem>
</itemizedlist>
</sect2>

<sect2>
<title>Browsing of Shares and Directories is Very Slow</title>

<para><quote>
<indexterm significance="normal"><primary>slow browsing</primary></indexterm>
There are only two machines on a test network. One is a Samba server, the other a Windows XP machine.
Authentication and logons work perfectly, but when I try to explore shares on the Samba server, the
Windows XP client becomes unresponsive. Sometimes it does not respond for some minutes. Eventually,
Windows Explorer will respond and displays files and directories without problem.
</quote>
</para>

<para><quote>
<indexterm significance="normal"><primary>cmd</primary></indexterm>
But, the share is immediately available from a command shell (<literal>cmd</literal>, followed by
exploration with DOS command. Is this a Samba problem, or is it a Windows problem? How can I solve this?
</quote></para>

<para>
Here are a few possibilities:
</para>

<variablelist>
	<varlistentry>
		<term>Bad Networking Hardware</term>
		<listitem><para>
<indexterm significance="normal"><primary>bad hardware</primary></indexterm>
<indexterm significance="normal"><primary>WebClient</primary></indexterm>
<indexterm significance="normal"><primary>defective hardware</primary></indexterm>
<indexterm significance="normal"><primary>Bad networking hardware</primary></indexterm>
<indexterm significance="normal"><primary>data corruption</primary></indexterm>
		Most common defective hardware problems center around low cost or defective hubs, routers,
		network interface controllers (NICs), and bad wiring. If one piece of hardware is defective,
		the whole network may suffer. Bad networking hardware can cause data corruption. Most bad
		networking hardware problems are accompanied by an increase in apparent network traffic,
		but not all.
		</para></listitem>
	</varlistentry>

	<varlistentry>
		<term>The Windows XP WebClient</term>
		<listitem><para>
<indexterm significance="normal"><primary>network browsing problems</primary></indexterm>
		A number of sites have reported similar slow network browsing problems and found that when
		the WebClient service is turned off, the problem disappears. This is certainly something
		that should be explored because it is a simple solution <?latex --- ?> if it works.
		</para></listitem>
	</varlistentry>

	<varlistentry>
		<term>Inconsistent WINS Configuration</term>
		<listitem><para>
<indexterm significance="normal"><primary>WINS Configuration</primary></indexterm>
<indexterm significance="normal"><primary>WINS server</primary></indexterm>
		This type of problem is common when one client is configured to use a WINS server (that is
		a TCP/IP configuration setting) and there is no WINS server on the network. Alternatively,
		this will happen if there is a WINS server and Samba is not configured to use it. The use of
		WINS is highly recommended if the network is using NetBIOS over TCP/IP protocols. If use
		of NetBIOS over TCP/IP is disabled on all clients, Samba should not be configured as a WINS
		server, nor should it be configured to use one.
		</para></listitem>
	</varlistentry>

	<varlistentry>
		<term>Incorrect DNS Configuration</term>
		<listitem><para>
<indexterm significance="normal"><primary>DNS Configuration</primary></indexterm>
<indexterm significance="normal"><primary>NetBIOS over TCP/IP disabled</primary></indexterm>
		If use of NetBIOS over TCP/IP is disabled, Active Directory is in use and the DNS server
		has been incorrectly configured. For further information refer to
		<link linkend="adsdnstech">DNS and Active Directory</link>.
		</para></listitem>
	</varlistentry>
</variablelist>

</sect2>
<sect2>
<title>Invalid Cached Share References Affects Network Browsing</title>
<para>
<indexterm significance="normal"><primary>cached references</primary></indexterm>
<indexterm significance="normal"><primary>stale network links</primary></indexterm>
Cached references on your MS Windows client (workstation or server) to shares or servers that no longer exist
can cause MS Windows Explorer to appear unresponsive as it tries to connect to these shares. After a delay
(can take a long time) it times out and browsing will appear to be mostly normal again.
</para>

<para>
To eliminate the problem the stale cached references should be removed. This does not happen automatically and
requires manual intervention. This is a design feature of MS Windows and not anything that Samba can change.
To remove the stale shortcuts found in <emphasis>My Network Places</emphasis> which refer to what are now
invalid shares or servers it is necessary to edit the Windows Registry under
<literal moreinfo="none">HKCU\Software\Microsoft\Windows\CurrentVersion\Explorer\</literal>. Edit the entry
<literal moreinfo="none">MountPoints2</literal> (on Windows XP and later, or <literal moreinfo="none">MountPoints</literal> on Windows 2000
and earlier). Remove all keys named <literal moreinfo="none">\\server\share</literal> (where 'server' and 'share' refer to a
non-existent server or share).
</para>

<note><para>
Removal of stale network links needs to be done on a per-user basis. Alternately, you can delete the
shortcuts from the MS Windows Explorer in <literal moreinfo="none">My Network Places</literal> just by right-clicking them and
selecting <emphasis>Delete.</emphasis>
</para></note>

<para>
<indexterm significance="normal"><primary>slow network browsing</primary></indexterm>
Samba users have reported that these stale references negatively affect network browsing with Windows, Samba,
and Novell servers. It is suspected to be a universal problem not directly related to the Samba
server. Samba users may experience this more often due to Samba being somewhat viewed as an experimenter's
toolkit. This results from the fact that a user might go through several reconfigurations and incarnations of
their Samba server, by different names, with different shares, increasing the chances for having stale
(invalid) cached share references. Windows clients do not expire these references thus necessitating manual
removal.
</para>

<para>
It is common for <emphasis>Open</emphasis> dialog boxes (for example; in Word and Excel) to respond very
slowly, as they attempt to locate all of the cached references, even if they are not in the current directory
being accessed.
</para>
</sect2>
</sect1>
</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="passdb">
<chapterinfo>
	<author>
<firstname>Jelmer</firstname><othername>R.</othername><surname>Vernooij</surname><othername>R.</othername>
<affiliation>
	<orgname>The Samba Team</orgname>
	<address format="linespecific"><email>jelmer@samba.org</email></address>
</affiliation></author>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
	<author>
<firstname>Gerald</firstname><surname>Carter</surname><othername>(Jerry)</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jerry@samba.org</email></address>
</affiliation></author>
	
<author>
	<firstname>Jeremy</firstname><surname>Allison</surname>
	<affiliation>
		<orgname>Samba Team</orgname>
		<address format="linespecific"><email>jra@samba.org</email></address>
	</affiliation>
</author>
	<author>
<firstname>Guenther</firstname><surname>Deschner</surname>
<affiliation>
	<orgname>SuSE</orgname>
	<address format="linespecific"><email>gd@suse.de</email></address>
</affiliation><contrib>LDAP updates</contrib></author>
	<author>
		<firstname>Olivier (lem)</firstname><surname>Lemaire</surname>
		<affiliation>
			<orgname>IDEALX</orgname>
			<address format="linespecific"><email>olem@IDEALX.org</email></address>
		</affiliation>
	</author>

	<pubdate>May 24, 2003</pubdate>
</chapterinfo>
<title>Account Information Databases</title>

<para>
<indexterm significance="normal"><primary>account backends</primary></indexterm>
<indexterm significance="normal"><primary>password backends</primary></indexterm>
<indexterm significance="normal"><primary>scalability</primary></indexterm>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
Early releases of Samba-3 implemented new capability to work concurrently with multiple account backends. This
capability was removed beginning with release of Samba 3.0.23. Commencing with Samba 3.0.23 it is possible to
work with only one specified passwd backend.
</para>

<para>
<indexterm significance="normal"><primary>passdb backend</primary></indexterm>
<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
<indexterm significance="normal"><primary>tdbsam</primary></indexterm>
<indexterm significance="normal"><primary>ldapsam</primary></indexterm>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
<indexterm significance="normal"><primary>single repository</primary></indexterm>
The three passdb backends that are fully maintained (actively supported) by the Samba Team are:
<literal moreinfo="none">smbpasswd</literal> (being obsoleted), <literal moreinfo="none">tdbsam</literal> (a tdb-based binary file format),
and <literal moreinfo="none">ldapsam</literal> (LDAP directory).  Of these, only the <literal moreinfo="none">ldapsam</literal> backend
stores both POSIX (UNIX) and Samba user and group account information in a single repository. The
<literal moreinfo="none">smbpasswd</literal> and <literal moreinfo="none">tdbsam</literal> backends store only Samba user accounts.
</para>

<para>
In a strict sense, there are three supported account storage and access systems. One of these is considered
obsolete (smbpasswd). It is recommended to use the <literal moreinfo="none">tdbsam</literal> method for all simple systems. Use
<literal moreinfo="none">ldapsam</literal> for larger and more complex networks.
</para>

<para>
<indexterm significance="normal"><primary>passdb backend</primary></indexterm>
<indexterm significance="normal"><primary>account storage mechanisms</primary></indexterm>
<indexterm significance="normal"><primary>account storage system</primary></indexterm>
<indexterm significance="normal"><primary>user and trust accounts</primary></indexterm>
<indexterm significance="normal"><primary>machine trust accounts</primary></indexterm>
<indexterm significance="normal"><primary>computer accounts</primary></indexterm>
<indexterm significance="normal"><primary>interdomain trust accounts</primary></indexterm>
In a strict and literal sense, the passdb backends are account storage mechanisms (or methods) alone. The choice
of terminology can be misleading, however we are stuck with this choice of wording. This chapter documents the
nature of the account storage system with a focus on user and trust accounts. Trust accounts have two forms,
machine trust accounts (computer accounts) and interdomain trust accounts. These are all treated as user-like
entities.
</para>

<sect1>
<title>Features and Benefits</title>

<para>
Samba-3 provides for complete backward compatibility with Samba-2.2.x functionality
as follows:
<indexterm significance="normal"><primary>SAM backend</primary><secondary>smbpasswd</secondary></indexterm>
<indexterm significance="normal"><primary>SAM backend</primary><secondary>ldapsam_compat</secondary></indexterm>
<indexterm significance="normal"><primary>encrypted passwords</primary></indexterm>
</para>

<sect2>
	<title>Backward Compatibility Account Storage Systems</title>

<variablelist>
	<varlistentry><term>Plaintext</term>
		<listitem>
			<para>
<indexterm significance="normal"><primary>plaintext</primary></indexterm>
<indexterm significance="normal"><primary>plaintext authentication</primary></indexterm>
<indexterm significance="normal"><primary>/etc/passwd</primary></indexterm>
<indexterm significance="normal"><primary>/etc/shadow</primary></indexterm>
<indexterm significance="normal"><primary>PAM</primary></indexterm>
			This isn't really a backend at all, but is listed here for simplicity.  Samba can be configured to pass
			plaintext authentication requests to the traditional UNIX/Linux <filename moreinfo="none">/etc/passwd</filename> and
			<filename moreinfo="none">/etc/shadow</filename>-style subsystems.  On systems that have Pluggable Authentication Modules
			(PAM) support, all PAM modules are supported. The behavior is just as it was with Samba-2.2.x, and the
			protocol limitations imposed by MS Windows clients apply likewise. Please refer to <link linkend="passdbtech">Technical Information</link>, for more information regarding the limitations of plaintext
			password usage.
			</para>
		</listitem>
	</varlistentry>

	<varlistentry><term>smbpasswd</term>
		<listitem>
			<para>
<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
<indexterm significance="normal"><primary>LanMan passwords</primary></indexterm>
<indexterm significance="normal"><primary>NT-encrypted passwords</primary></indexterm>
<indexterm significance="normal"><primary>SAM</primary></indexterm>
			This option allows continued use of the <filename moreinfo="none">smbpasswd</filename>
			file that maintains a plain ASCII (text) layout that includes the MS Windows
			LanMan and NT-encrypted passwords as well as a field that stores some
			account information. This form of password backend does not store any of
			the MS Windows NT/200x SAM (Security Account Manager) information required to
			provide the extended controls that are needed for more comprehensive
			interoperation with MS Windows NT4/200x servers.
			</para>

			<para>
			This backend should be used only for backward compatibility with older
			versions of Samba. It may be deprecated in future releases.
			</para>
		</listitem>
	</varlistentry>

	<varlistentry><term>ldapsam_compat (Samba-2.2 LDAP Compatibility)</term>
		<listitem>
			<para>
<indexterm significance="normal"><primary>ldapsam_compat</primary></indexterm>
<indexterm significance="normal"><primary>Samba-2.2.x LDAP schema</primary></indexterm>
<indexterm significance="normal"><primary>OpenLDAP backend</primary></indexterm>
			There is a password backend option that allows continued operation with
			an existing OpenLDAP backend that uses the Samba-2.2.x LDAP schema extension.
			This option is provided primarily as a migration tool, although there is
			no reason to force migration at this time. This tool will eventually
			be deprecated.
			</para>
		</listitem>
	</varlistentry>
</variablelist>

</sect2>

<sect2>
<title>New Account Storage Systems</title>

<para>
Samba-3 introduces a number of new password backend capabilities.
<indexterm significance="normal"><primary>SAM backend</primary><secondary>tdbsam</secondary></indexterm>
<indexterm significance="normal"><primary>SAM backend</primary><secondary>ldapsam</secondary></indexterm>
</para>

<variablelist>
	<varlistentry><term>tdbsam</term>
		<listitem>
			<para>
<indexterm significance="normal"><primary>rich database backend</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>BDC</primary></indexterm>
			This backend provides a rich database backend for local servers. This
			backend is not suitable for multiple domain controllers (i.e., PDC + one
			or more BDC) installations.
			</para>

			<para>
<indexterm significance="normal"><primary>extended SAM</primary></indexterm>
<indexterm significance="normal"><primary>TDB</primary></indexterm>
<indexterm significance="normal"><primary>binary format TDB</primary></indexterm>
<indexterm significance="normal"><primary>trivial database</primary></indexterm>
<indexterm significance="normal"><primary>system access controls</primary></indexterm>
<indexterm significance="normal"><primary>MS Windows NT4/200x</primary></indexterm>
			The <emphasis>tdbsam</emphasis> password backend stores the old <emphasis>
			smbpasswd</emphasis> information plus the extended MS Windows NT/200x
			SAM information into a binary format TDB (trivial database) file.
			The inclusion of the extended information makes it possible for Samba-3
			to implement the same account and system access controls that are possible
			with MS Windows NT4/200x-based systems.
			</para>

			<para>
<indexterm significance="normal"><primary>simple operation</primary></indexterm>
<indexterm significance="normal"><primary>OpenLDAP</primary></indexterm>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
			The inclusion of the <emphasis>tdbsam</emphasis> capability is a direct
			response to user requests to allow simple site operation without the overhead
			of the complexities of running OpenLDAP. It is recommended to use this only
			for sites that have fewer than 250 users. For larger sites or implementations,
			the use of OpenLDAP or of Active Directory integration is strongly recommended.
			</para>
		</listitem>
	</varlistentry>

	<varlistentry><term>ldapsam</term>
		<listitem>
			<para>
<indexterm significance="normal"><primary>rich directory backend</primary></indexterm>
<indexterm significance="normal"><primary>distributed account</primary></indexterm>
			This provides a rich directory backend for distributed account installation.
			</para>

			<para>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
<indexterm significance="normal"><primary>OpenLDAP</primary></indexterm>
<indexterm significance="normal"><primary>Samba schema</primary></indexterm>
<indexterm significance="normal"><primary>schema file</primary></indexterm>
<indexterm significance="normal"><primary>examples/LDAP</primary></indexterm>
			Samba-3 has a new and extended LDAP implementation that requires configuration
			of OpenLDAP with a new format Samba schema. The new format schema file is
			included in the <filename class="directory" moreinfo="none">examples/LDAP</filename> directory of the Samba distribution.
			</para>

			<para>
<indexterm significance="normal"><primary>expands control abilities</primary></indexterm>
<indexterm significance="normal"><primary>profile</primary></indexterm>
<indexterm significance="normal"><primary>home directories</primary></indexterm>
<indexterm significance="normal"><primary>account access controls</primary></indexterm>
<indexterm significance="normal"><primary>greater scalability</primary></indexterm>
			The new LDAP implementation significantly expands the control abilities that
			were possible with prior versions of Samba. It is now possible to specify
			<quote>per-user</quote> profile settings, home directories, account access controls, and
			much more. Corporate sites will see that the Samba Team has listened to their
			requests both for capability and greater scalability.
			</para>
		</listitem>
	</varlistentry>

</variablelist>

</sect2>

</sect1>

<sect1 id="passdbtech">
	<title>Technical Information</title>

	<para>
<indexterm significance="normal"><primary>plaintext passwords</primary></indexterm>
<indexterm significance="normal"><primary>encrypted passwords</primary></indexterm>
	Old Windows clients send plaintext passwords over the wire. Samba can check these
	passwords by encrypting them and comparing them to the hash stored in the UNIX user database.
	</para>

	<para>
<indexterm significance="normal"><primary>encrypted passwords</primary></indexterm>
<indexterm significance="normal"><primary>LanMan</primary></indexterm>
<indexterm significance="normal"><primary>plaintext passwords</primary></indexterm>
<indexterm significance="normal"><primary>registry</primary></indexterm>
	Newer Windows clients send encrypted passwords (LanMan and NT hashes) instead of plaintext passwords over
	the wire. The newest clients will send only encrypted passwords and refuse to send plaintext passwords unless
	their registry is tweaked.
	</para>

	<para>
<indexterm significance="normal"><primary>UNIX-style encrypted passwords</primary></indexterm>
<indexterm significance="normal"><primary>converted</primary></indexterm>
	Many people ask why Samba cannot simply use the UNIX password database. Windows requires
	passwords that are encrypted in its own format.  The UNIX passwords can't be converted to
	UNIX-style encrypted passwords. Because of that, you can't use the standard UNIX user
	database, and you have to store the LanMan and NT hashes somewhere else.
	</para>

	<para>
<indexterm significance="normal"><primary>differently encrypted passwords</primary></indexterm>
<indexterm significance="normal"><primary>profile</primary></indexterm>
<indexterm significance="normal"><primary>workstations</primary></indexterm>
<indexterm significance="normal"><primary>tdbsam</primary></indexterm>
	In addition to differently encrypted passwords, Windows also stores certain data for each
	user that is not stored in a UNIX user database: for example, workstations the user may logon from,
	the location where the user's profile is stored, and so on. Samba retrieves and stores this
	information using a <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PASSDBBACKEND" xlink:href="smb.conf.5.html#PASSDBBACKEND">passdb backend</link>. Commonly available backends are LDAP,
	tdbsam, and plain text file. For more information, see the man page for <filename moreinfo="none">smb.conf</filename> regarding the
	<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PASSDBBACKEND" xlink:href="smb.conf.5.html#PASSDBBACKEND">passdb backend</link> parameter.
	</para>


	<figure id="idmap-sid2uid" float="0">
		<title>IDMAP: Resolution of SIDs to UIDs.</title>
		<mediaobject><imageobject role="html"><imagedata fileref="images/idmap-sid2uid.png" scale="40" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/idmap-sid2uid.png" scale="40" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/idmap-sid2uid"/></imageobject></mediaobject>
	</figure>

	<para>
	<indexterm significance="normal"><primary>SID</primary></indexterm>
<indexterm significance="normal"><primary>UID</primary></indexterm>
<indexterm significance="normal"><primary>SID</primary></indexterm>
	The resolution of SIDs to UIDs is fundamental to correct operation of Samba. In both cases shown, if winbindd
	is not running or cannot be contacted, then only local SID/UID resolution is possible. See <link linkend="idmap-sid2uid">resolution of SIDs to UIDs</link> and <link linkend="idmap-uid2sid">resolution of UIDs
	to SIDs</link> diagrams.
	</para>

	<figure id="idmap-uid2sid" float="0">
		<title>IDMAP: Resolution of UIDs to SIDs.</title>
		<mediaobject><imageobject role="html"><imagedata fileref="images/idmap-uid2sid.png" scale="50" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/idmap-uid2sid.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/idmap-uid2sid"/></imageobject></mediaobject>
	</figure>

	<sect2>
	<title>Important Notes About Security</title>

		<para>
<indexterm significance="normal"><primary>SMB password encryption</primary></indexterm>
<indexterm significance="normal"><primary>clear-text passwords</primary></indexterm>
<indexterm significance="normal"><primary>hashed password equivalent</primary></indexterm>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
<indexterm significance="normal"><primary>secret</primary></indexterm>
		The UNIX and SMB password encryption techniques seem similar on the surface. This
		similarity is, however, only skin deep. The UNIX scheme typically sends clear-text
		passwords over the network when logging in. This is bad. The SMB encryption scheme
		never sends the clear-text password over the network, but it does store the 16-byte
		hashed values on disk. This is also bad. Why? Because the 16 byte hashed values
		are a <quote>password equivalent.</quote> You cannot derive the user's password from them, but
		they could potentially be used in a modified client to gain access to a server.
		This would require considerable technical knowledge on behalf of the attacker but
		is perfectly possible. You should therefore treat the data stored in whatever passdb
		backend you use (smbpasswd file, LDAP) as though it contained the clear-text
		passwords of all your users. Its contents must be kept secret, and the file should
		be protected accordingly.
		</para>

		<para>
<indexterm significance="normal"><primary>password scheme</primary></indexterm>
<indexterm significance="normal"><primary>plaintext passwords</primary></indexterm>
<indexterm significance="normal"><primary>compatible</primary></indexterm>
		Ideally, we would like a password scheme that involves neither plaintext passwords
		on the network nor plaintext passwords on disk. Unfortunately, this is not available because Samba is stuck with
		having to be compatible with other SMB systems (Windows NT, Windows for Workgroups, Windows 9x/Me).
		</para>

		<para>
<indexterm significance="normal"><primary>encrypted passwords</primary></indexterm>
<indexterm significance="normal"><primary>plaintext passwords</primary></indexterm>
		Windows NT 4.0 Service Pack 3 changed the default setting so plaintext passwords
		are disabled from being sent over the wire. This mandates either the use of encrypted
		password support or editing the Windows NT registry to re-enable plaintext passwords.
		</para>

		<para>
<indexterm significance="normal"><primary>domain security</primary></indexterm>
<indexterm significance="normal"><primary>domain environment</primary></indexterm>
		The following versions of Microsoft Windows do not support full domain security protocols,
		although they may log onto a domain environment:
		</para>

		<itemizedlist>
			<listitem><para>MS DOS Network client 3.0 with the basic network redirector installed.</para></listitem>
			<listitem><para>Windows 95 with the network redirector update installed.</para></listitem>
			<listitem><para>Windows 98 [Second Edition].</para></listitem>
			<listitem><para>Windows Me.</para></listitem>
		</itemizedlist>

		<note>
		<para>
<indexterm significance="normal"><primary>Windows XP Home</primary></indexterm>
<indexterm significance="normal"><primary>domain member</primary></indexterm>
<indexterm significance="normal"><primary>domain logons</primary></indexterm>
 		MS Windows XP Home does not have facilities to become a domain member, and it cannot participate in domain logons.
		</para>
		</note>

		<para>
		The following versions of MS Windows fully support domain security protocols.
		</para>

		<itemizedlist>
			<listitem><para>Windows NT 3.5x.</para></listitem>
			<listitem><para>Windows NT 4.0.</para></listitem>
			<listitem><para>Windows 2000 Professional.</para></listitem>
			<listitem><para>Windows 200x Server/Advanced Server.</para></listitem>
			<listitem><para>Windows XP Professional.</para></listitem>
		</itemizedlist>

		<para>
<indexterm significance="normal"><primary>SMB/CIFS</primary></indexterm>
<indexterm significance="normal"><primary>authentication</primary></indexterm>
<indexterm significance="normal"><primary>challenge/response mechanis</primary></indexterm>
<indexterm significance="normal"><primary>clear-text</primary></indexterm>
<indexterm significance="normal"><primary>encrypted</primary></indexterm>
<indexterm significance="normal"><primary>negotiate</primary></indexterm>
		All current releases of Microsoft SMB/CIFS clients support authentication via the
		SMB challenge/response mechanism described here. Enabling clear-text authentication
		does not disable the ability of the client to participate in encrypted authentication.
		Instead, it allows the client to negotiate either plaintext or encrypted password
		handling.
		</para>

		<para>
<indexterm significance="normal"><primary>cached encrypted password</primary></indexterm>
<indexterm significance="normal"><primary>plaintext passwords</primary></indexterm>
<indexterm significance="normal"><primary>registry change</primary></indexterm>
<indexterm significance="normal"><primary>auto-reconnect</primary></indexterm>
<indexterm significance="normal"><primary>encrypted passwords</primary></indexterm>
		MS Windows clients will cache the encrypted password alone. Where plaintext passwords
		are re-enabled through the appropriate registry change, the plaintext password is never
		cached. This means that in the event that a network connections should become disconnected
		(broken), only the cached (encrypted) password will be sent to the resource server to
		effect an auto-reconnect. If the resource server does not support encrypted passwords, the
		auto-reconnect will fail. Use of encrypted passwords is strongly advised.
		</para>

		<sect3>
		<title>Advantages of Encrypted Passwords</title>

			<itemizedlist>
				<listitem><para>
<indexterm significance="normal"><primary>passed across the network</primary></indexterm>
<indexterm significance="normal"><primary>network sniffer</primary></indexterm>
<indexterm significance="normal"><primary>SMB server</primary></indexterm>
				Plaintext passwords are not passed across the network. Someone using a network sniffer
				cannot just record passwords going to the SMB server.
				</para></listitem>

				<listitem><para>
<indexterm significance="normal"><primary>not stored anywhere</primary></indexterm>
<indexterm significance="normal"><primary>memory</primary></indexterm>
<indexterm significance="normal"><primary>disk</primary></indexterm>
				Plaintext passwords are not stored anywhere in memory or on disk.
				</para></listitem>
 
				<listitem><para>
<indexterm significance="normal"><primary>encrypted passwords</primary></indexterm>
<indexterm significance="normal"><primary>user-level security</primary></indexterm>
<indexterm significance="normal"><primary>password prompt</primary></indexterm>
<indexterm significance="normal"><primary>SMB encryption</primary></indexterm>
				Windows NT does not like talking to a server that does not support encrypted passwords. It will refuse to
				browse the server if the server is also in user-level security mode. It will insist on prompting the user for
				the password on each connection, which is very annoying. The only thing you can do to stop this is to use SMB
				encryption.
				</para></listitem>

				<listitem><para>
<indexterm significance="normal"><primary>encrypted password</primary></indexterm>
<indexterm significance="normal"><primary>automatic reconnects</primary></indexterm>
				Encrypted password support allows automatic share (resource) reconnects.
				</para></listitem>

				<listitem><para>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>BDC</primary></indexterm>
				Encrypted passwords are essential for PDC/BDC operation.
				</para></listitem>
			</itemizedlist>
		</sect3>


		<sect3>
		<title>Advantages of Non-Encrypted Passwords</title>

			<itemizedlist>
				<listitem><para>
<indexterm significance="normal"><primary>cached in memory</primary></indexterm>
				Plaintext passwords are not kept on disk and are not cached in memory.
				</para></listitem>

				<listitem><para>
<indexterm significance="normal"><primary>Login</primary></indexterm>
<indexterm significance="normal"><primary>FTP</primary></indexterm>
				Plaintext passwords use the same password file as other UNIX services, such as Login and FTP.
				</para></listitem>

				<listitem><para>
<indexterm significance="normal"><primary>Telnet</primary></indexterm>
<indexterm significance="normal"><primary>FTP</primary></indexterm>
				Use of other services (such as Telnet and FTP) that send plaintext passwords over
				the network makes sending them for SMB not such a big deal.
				</para></listitem>
			</itemizedlist>
		</sect3>
	</sect2>

	<sect2>
	<title>Mapping User Identifiers between MS Windows and UNIX</title>

	<para>
<indexterm significance="normal"><primary>UID</primary></indexterm>
<indexterm significance="normal"><primary>SID</primary></indexterm>
<indexterm significance="normal"><primary>mapping</primary></indexterm>
	Every operation in UNIX/Linux requires a user identifier (UID), just as in
	MS Windows NT4/200x this requires a security identifier (SID). Samba provides
	two means for mapping an MS Windows user to a UNIX/Linux UID.
	</para>

	<para>
<indexterm significance="normal"><primary>Samba SAM</primary></indexterm>
<indexterm significance="normal"><primary>SAM</primary></indexterm>
<indexterm significance="normal"><primary>UID</primary></indexterm>
<indexterm significance="normal"><primary>account information database</primary></indexterm>
<indexterm significance="normal"><primary>local user account</primary></indexterm>
	First, all Samba SAM database accounts require a UNIX/Linux UID that the account will map to. As users are
	added to the account information database, Samba will call the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="ADDUSERSCRIPT" xlink:href="smb.conf.5.html#ADDUSERSCRIPT">add user script</link>
	interface to add the account to the Samba host OS. In essence all accounts in the local SAM require a local
	user account.
	</para>

	<para>
	<indexterm significance="normal"><primary>idmap uid</primary></indexterm>
	<indexterm significance="normal"><primary>idmap gid</primary></indexterm>
	<indexterm significance="normal"><primary>UID</primary></indexterm>
	<indexterm significance="normal"><primary>SAM</primary></indexterm>
	<indexterm significance="normal"><primary>foreign domain</primary></indexterm>
	<indexterm significance="normal"><primary>non-member Windows client</primary></indexterm>
	<indexterm significance="normal"><primary>SID</primary></indexterm>
	The second way to map Windows SID to UNIX UID is via the <emphasis>idmap uid</emphasis> and
	<emphasis>idmap gid</emphasis> parameters in <filename moreinfo="none">smb.conf</filename>.  Please refer to the man page for information about
	these parameters.  These parameters are essential when mapping users from a remote (non-member Windows client
	or a member of a foreign domain) SAM server.
	</para>

	</sect2>

	<sect2 id="idmapbackend">
	<title>Mapping Common UIDs/GIDs on Distributed Machines</title>

	<para>
<indexterm significance="normal"><primary>UID</primary></indexterm>
<indexterm significance="normal"><primary>GID</primary></indexterm>
<indexterm significance="normal"><primary>BDC</primary></indexterm>
<indexterm significance="normal"><primary>domain member servers</primary></indexterm>
<indexterm significance="normal"><primary>NFS</primary></indexterm>
<indexterm significance="normal"><primary>rsync</primary></indexterm>
	Samba-3 has a special facility that makes it possible to maintain identical UIDs and GIDs
	on all servers in a distributed network. A distributed network is one where there exists
	a PDC, one or more BDCs, and/or one or more domain member servers. Why is this important?
	This is important if files are being shared over more than one protocol (e.g., NFS) and where
	users are copying files across UNIX/Linux systems using tools such as <literal>rsync</literal>.
	</para>

	<para>
<indexterm significance="normal"><primary>LDAP-based</primary></indexterm>
<indexterm significance="normal"><primary>idmap backend</primary></indexterm>
<indexterm significance="normal"><primary>UID</primary></indexterm>
<indexterm significance="normal"><primary>GID</primary></indexterm>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
<indexterm significance="normal"><primary>SAM backend</primary></indexterm>
<indexterm significance="normal"><primary>LDAP idmap Backend</primary></indexterm>
	<indexterm significance="normal"><primary>idmap backend</primary></indexterm>
	The special facility is enabled using a parameter called <parameter moreinfo="none">idmap backend</parameter>.
	The default setting for this parameter is an empty string. Technically it is possible to use
	an LDAP-based idmap backend for UIDs and GIDs, but it makes most sense when this is done for
	network configurations that also use LDAP for the SAM backend.
	<link linkend="idmapbackendexample">Example Configuration with the LDAP idmap Backend</link>
	shows that configuration.
	</para>

<indexterm significance="normal"><primary>SAM backend</primary><secondary>ldapsam</secondary></indexterm>
<example id="idmapbackendexample">
<title>Example Configuration with the LDAP idmap Backend</title>
<simplelist>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>idmap backend</indexterm><parameter>idmap backend = ldap:ldap://ldap-server.quenya.org:636</parameter></member>
<member># Alternatively, this could be specified as:</member>
<member><indexterm>idmap backend</indexterm><parameter>idmap backend = ldap:ldaps://ldap-server.quenya.org</parameter></member>
</simplelist>
</example>

	<para>
<indexterm significance="normal"><primary>LDAP backends</primary></indexterm>
<indexterm significance="normal"><primary>PADL Software</primary></indexterm>
	A network administrator who wants to make significant use of LDAP backends will sooner or later be
	exposed to the excellent work done by PADL Software. PADL <ulink url="http://www.padl.com"/> have
	produced and released to open source an array of tools that might be of interest. These tools include:
	</para>

	<itemizedlist>
		<listitem>
		<para>
<indexterm significance="normal"><primary>nss_ldap</primary></indexterm>
<indexterm significance="normal"><primary>NSS</primary></indexterm>
<indexterm significance="normal"><primary>AIX</primary></indexterm>
<indexterm significance="normal"><primary>Linux</primary></indexterm>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
<indexterm significance="normal"><primary>Solaris</primary></indexterm>
<indexterm significance="normal"><primary>UID</primary></indexterm>
<indexterm significance="normal"><primary>GID</primary></indexterm>
		<emphasis>nss_ldap:</emphasis> An LDAP name service switch (NSS) module to provide native
		name service support for AIX, Linux, Solaris, and other operating systems. This tool
		can be used for centralized storage and retrieval of UIDs and GIDs.
		</para>
		</listitem>

		<listitem>
		<para>
<indexterm significance="normal"><primary>pam_ldap</primary></indexterm>
<indexterm significance="normal"><primary>PAM</primary></indexterm>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
<indexterm significance="normal"><primary>access authentication</primary></indexterm>
		<emphasis>pam_ldap:</emphasis> A PAM module that provides LDAP integration for UNIX/Linux
		system access authentication.
		</para>
		</listitem>

		<listitem>
		<para>
<indexterm significance="normal"><primary>idmap_ad</primary></indexterm>
<indexterm significance="normal"><primary>IDMAP backend</primary></indexterm>
<indexterm significance="normal"><primary>RFC 2307</primary></indexterm>
<indexterm significance="normal"><primary>PADL</primary></indexterm>
		<emphasis>idmap_ad:</emphasis> An IDMAP backend that supports the Microsoft Services for
		UNIX RFC 2307 schema available from the PADL Web
		<ulink url="http://www.padl.com/download/xad_oss_plugins.tar.gz">site</ulink>.
		</para>
		</listitem>
	</itemizedlist>

	</sect2>

	<sect2>
	<title>Comments Regarding LDAP</title>

	<para>
<indexterm significance="normal"><primary>LDAP</primary><secondary>directories</secondary></indexterm>
<indexterm significance="normal"><primary>architecture</primary></indexterm>
<indexterm significance="normal"><primary>FIM</primary></indexterm>
<indexterm significance="normal"><primary>SSO</primary></indexterm>
	There is much excitement and interest in LDAP directories in the information technology world
	today. The LDAP architecture was designed to be highly scalable. It was also designed for
	use across a huge number of potential areas of application encompassing a wide range of operating
	systems and platforms. LDAP technologies are at the heart of the current generations of Federated
	Identity Management (FIM) solutions that can underlie a corporate Single Sign-On (SSO) environment.
	</para>

	<para>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
<indexterm significance="normal"><primary>eDirectory</primary></indexterm>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
<indexterm significance="normal"><primary>authentication</primary></indexterm>
	LDAP implementations have been built across a wide variety of platforms. It lies at the core of Microsoft
	Windows Active Directory services (ADS), Novell's eDirectory, as well as many others. Implementation of the
	directory services LDAP involves interaction with legacy as well as new generation applications, all of which
	depend on some form of authentication services.
	</para>

	<para>
<indexterm significance="normal"><primary>LDAP directory</primary></indexterm>
<indexterm significance="normal"><primary>authentication</primary></indexterm>
<indexterm significance="normal"><primary>access controls</primary></indexterm>
<indexterm significance="normal"><primary>intermediate tools</primary></indexterm>
<indexterm significance="normal"><primary>middle-ware</primary></indexterm>
<indexterm significance="normal"><primary>central environment</primary></indexterm>
<indexterm significance="normal"><primary>infrastructure</primary></indexterm>
<indexterm significance="normal"><primary>login shells</primary></indexterm>
<indexterm significance="normal"><primary>mail</primary></indexterm>
<indexterm significance="normal"><primary>messaging systems</primary></indexterm>
<indexterm significance="normal"><primary>quota controls</primary></indexterm>
<indexterm significance="normal"><primary>printing systems</primary></indexterm>
<indexterm significance="normal"><primary>DNS servers</primary></indexterm>
<indexterm significance="normal"><primary>DHCP servers</primary></indexterm>
	UNIX services can utilize LDAP directory information for authentication and access controls
	through intermediate tools and utilities. The total environment that consists of the LDAP directory
	and the middle-ware tools and utilities makes it possible for all user access to the UNIX platform
	to be managed from a central environment and yet distributed to wherever the point of need may
	be physically located. Applications that benefit from this infrastructure include: UNIX login
	shells, mail and messaging systems, quota controls, printing systems, DNS servers, DHCP servers,
	and also Samba.
	</para>

	<para>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
<indexterm significance="normal"><primary>passdb backend</primary></indexterm>
<indexterm significance="normal"><primary>scalable</primary></indexterm>
<indexterm significance="normal"><primary>SAM backend</primary></indexterm>
<indexterm significance="normal"><primary>LDAP directory</primary></indexterm>
<indexterm significance="normal"><primary>management costs</primary></indexterm>
	Many sites are installing LDAP for the first time in order to provide a scalable passdb backend
	for Samba. Others are faced with the need to adapt an existing LDAP directory to new uses such
	as for the Samba SAM backend. Whatever your particular need and attraction to Samba may be,
	decisions made in respect of the design of the LDAP directory structure and its implementation
	are of a durable nature for the site. These have far-reaching implications that affect long-term
	information systems management costs.
	</para>

	<para>
<indexterm significance="normal"><primary>LDAP deployment</primary></indexterm>
<indexterm significance="normal"><primary>Directory Information Tree</primary><see>DIT</see></indexterm>
	Do not rush into an LDAP deployment. Take the time to understand how the design of the Directory
	Information Tree (DIT) may impact current and future site needs, as well as the ability to meet
	them. The way that Samba SAM information should be stored within the DIT varies from site to site
	and with each implementation new experience is gained. It is well understood by LDAP veterans that
	first implementations create awakening, second implementations of LDAP create fear, and
	third-generation deployments bring peace and tranquility.
	</para>

	<sect3>
	<title>Caution Regarding LDAP and Samba</title>

	<para>
<indexterm significance="normal"><primary>POSIX identity</primary></indexterm>
<indexterm significance="normal"><primary>networking environment</primary></indexterm>
<indexterm significance="normal"><primary>user accounts</primary></indexterm>
<indexterm significance="normal"><primary>group accounts</primary></indexterm>
<indexterm significance="normal"><primary>machine trust accounts</primary></indexterm>
<indexterm significance="normal"><primary>interdomain trust accounts</primary></indexterm>
<indexterm significance="normal"><primary>intermediate information</primary></indexterm>
	Samba requires UNIX POSIX identity information as well as a place to store information that is
	specific to Samba and the Windows networking environment. The most used information that must
	be dealt with includes: user accounts, group accounts, machine trust accounts, interdomain
	trust accounts, and intermediate information specific to Samba internals.
	</para>

	<para>
<indexterm significance="normal"><primary>deployment guidelines</primary></indexterm>
<indexterm significance="normal"><primary>HOWTO documents</primary></indexterm>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
	The example deployment guidelines in this book, as well as other books and HOWTO documents
	available from the internet may not fit with established directory designs and implementations.
	The existing DIT may not be able to accommodate the simple information layout proposed in common
	sources. Additionally, you may find that the common scripts and tools that are used to provision
	the LDAP directory for use with Samba may not suit your needs.
	</para>

	<para>
<indexterm significance="normal"><primary>existing LDAP DIT</primary></indexterm>
	It is not uncommon, for sites that have existing LDAP DITs to find necessity to generate a
	set of site-specific scripts and utilities to make it possible to deploy Samba within the
	scope of site operations. The way that user and group accounts are distributed throughout
	the DIT may make this a challenging matter. The solution will, of course, be rewarding, but
	the journey to it may be challenging. Take time to understand site needs and do not rush
	into deployment.
	</para>

	<para>
<indexterm significance="normal"><primary>scripts</primary></indexterm>
<indexterm significance="normal"><primary>tools</primary></indexterm>
	Above all, do not blindly use scripts and tools that are not suitable for your site. Check
	and validate all scripts before you execute them to make sure that the existing infrastructure
	will not be damaged by inadvertent use of an inappropriate tool.
	</para>

	</sect3>

	</sect2>

	<sect2>
	<title>LDAP Directories and Windows Computer Accounts</title>

		<para>
<indexterm significance="normal"><primary>turnkey solution</primary></indexterm>
<indexterm significance="normal"><primary>LDAP.</primary></indexterm>
<indexterm significance="normal"><primary>frustrating experience</primary></indexterm>
		Samba doesn't provide a turnkey solution to LDAP. It is best to deal with the design and
		configuration of an LDAP directory prior to integration with Samba. A working knowledge
		of LDAP makes Samba integration easy, and the lack of a working knowledge of LDAP can make
		it a frustrating experience.
		</para>

		<para>
<indexterm significance="normal"><primary>computer accounts</primary></indexterm>
<indexterm significance="normal"><primary>machine accounts</primary></indexterm>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
		Computer (machine) accounts can be placed wherever you like in an LDAP directory subject
		to some constraints that are described in this chapter.
		</para>

		<para>
<indexterm significance="normal"><primary>POSIX</primary></indexterm>
<indexterm significance="normal"><primary>sambaSamAccount</primary></indexterm>
<indexterm significance="normal"><primary>computer accounts</primary></indexterm>
<indexterm significance="normal"><primary>machine accounts</primary></indexterm>
<indexterm significance="normal"><primary>Windows NT4/200X</primary></indexterm>
<indexterm significance="normal"><primary>user account</primary></indexterm>
<indexterm significance="normal"><primary>trust accounts</primary></indexterm>
		The POSIX and sambaSamAccount components of computer (machine) accounts are both used by Samba.
		Thus, machine accounts are treated inside Samba in the same way that Windows NT4/200X treats
		them. A user account and a machine account are indistinquishable from each other, except that
		the machine account ends in a $ character, as do trust accounts.
		</para>

		<para>
<indexterm significance="normal"><primary>user</primary></indexterm>
<indexterm significance="normal"><primary>group</primary></indexterm>
<indexterm significance="normal"><primary>machine</primary></indexterm>
<indexterm significance="normal"><primary>trust</primary></indexterm>
<indexterm significance="normal"><primary>UID</primary></indexterm>
		The need for Windows user, group, machine, trust, and other accounts to be tied to a valid UNIX
		UID is a design decision that was made a long way back in the history of Samba development. It
		is unlikely that this decision will be reversed or changed during the remaining life of the
		Samba-3.x series.
		</para>

		<para>
<indexterm significance="normal"><primary>UID</primary></indexterm>
<indexterm significance="normal"><primary>SID</primary></indexterm>
<indexterm significance="normal"><primary>NSS</primary></indexterm>
		The resolution of a UID from the Windows SID is achieved within Samba through a mechanism that
		must refer back to the host operating system on which Samba is running. The NSS is the preferred
		mechanism that shields applications (like Samba) from the need to know everything about every
		host OS it runs on.
		</para>

		<para>
<indexterm significance="normal"><primary>UID</primary></indexterm>
<indexterm significance="normal"><primary>passwd</primary></indexterm>
<indexterm significance="normal"><primary>shadow</primary></indexterm>
<indexterm significance="normal"><primary>group</primary></indexterm>
<indexterm significance="normal"><primary>NSS</primary></indexterm>
<indexterm significance="normal"><primary>winbindd</primary></indexterm>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
		Samba asks the host OS to provide a UID via the <quote>passwd</quote>, <quote>shadow</quote>,
		and <quote>group</quote> facilities in the NSS control (configuration) file. The best tool
		for achieving this is left up to the UNIX administrator to determine. It is not imposed by
		Samba. Samba provides winbindd with its support libraries as one method. It is
		possible to do this via LDAP, and for that Samba provides the appropriate hooks so that
		all account entities can be located in an LDAP directory.
		</para>

		<para>
<indexterm significance="normal"><primary>PADL</primary></indexterm>
<indexterm significance="normal"><primary>nss_ldap</primary></indexterm>
<indexterm significance="normal"><primary>UID</primary></indexterm>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
<indexterm significance="normal"><primary>documentation</primary></indexterm>
		For many the weapon of choice is to use the PADL nss_ldap utility. This utility must
		be configured so that computer accounts can be resolved to a POSIX/UNIX account UID. That
		is fundamentally an LDAP design question.  The information provided on the Samba list and
		in the documentation is directed at providing working examples only. The design
		of an LDAP directory is a complex subject that is beyond the scope of this documentation.
		</para>

	</sect2>

</sect1>

<sect1 id="acctmgmttools">
<title>Account Management Tools</title>

<para>
<indexterm significance="normal"><primary>pdbedit</primary></indexterm>
<indexterm significance="normal"><primary>machine accounts</primary></indexterm>
<indexterm significance="normal"><primary>management tools</primary></indexterm>
Samba provides two tools for management of user and machine accounts:
<literal>smbpasswd</literal> and <literal>pdbedit</literal>.
</para>

<para>
<indexterm significance="normal"><primary>pdbedit</primary></indexterm>
<indexterm significance="normal"><primary>password aging</primary></indexterm>
<indexterm significance="normal"><primary>failed logins</primary></indexterm>
The <literal>pdbedit</literal> can be used to manage account policies in addition to
Samba user account information. The policy management capability is used to administer
domain default settings for password aging and management controls to handle failed login
attempts.
</para>

<para>
<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
<indexterm significance="normal"><primary>storage mechanism</primary></indexterm>
<indexterm significance="normal"><primary>SambaSAMAccount</primary></indexterm>
<indexterm significance="normal"><primary>net</primary></indexterm>
Some people are confused when reference is made to <literal moreinfo="none">smbpasswd</literal> because the
name refers to a storage mechanism for SambaSAMAccount information, but it is also the name
of a utility tool. That tool is destined to eventually be replaced by new functionality that
is being added to the <literal>net</literal> toolset (see <link linkend="NetCommand">the Net Command</link>.
</para>

	<sect2>
	<title>The <literal>smbpasswd</literal> Tool</title>

		<para>
<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
<indexterm significance="normal"><primary>passwd</primary></indexterm>
<indexterm significance="normal"><primary>yppasswd</primary></indexterm>
<indexterm significance="normal"><primary>passdb backend</primary></indexterm>
<indexterm significance="normal"><primary>storage methods</primary></indexterm>
		The <literal>smbpasswd</literal> utility is similar to the <literal>passwd</literal>
		and <literal>yppasswd</literal> programs. It maintains the two 32 byte password
		fields in the passdb backend. This utility operates independently of the actual
		account and password storage methods used (as specified by the <parameter moreinfo="none">passdb
		backend</parameter> in the <filename moreinfo="none">smb.conf</filename> file.
		</para>

		<para>
<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
<indexterm significance="normal"><primary>client-server mode</primary></indexterm>
		<literal>smbpasswd</literal> works in a client-server mode where it contacts the
		local smbd to change the user's password on its behalf. This has enormous benefits.
		</para>

		<para>
<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
<indexterm significance="normal"><primary>change passwords</primary></indexterm>
		<literal>smbpasswd</literal> has the capability to change passwords on Windows NT
		servers (this only works when the request is sent to the NT PDC if changing an NT
		domain user's password).
		</para>

		<para>
		<indexterm significance="normal"><primary>user management</primary></indexterm>
		<indexterm significance="normal"><primary>user account</primary><secondary>Adding/Deleting</secondary></indexterm>
		<literal>smbpasswd</literal> can be used to:
		</para>

		<itemizedlist>
			<listitem><para><emphasis>add</emphasis> user or machine accounts.</para></listitem>
			<listitem><para><emphasis>delete</emphasis> user or machine accounts.</para></listitem>
			<listitem><para><emphasis>enable</emphasis> user or machine accounts.</para></listitem>
			<listitem><para><emphasis>disable</emphasis> user or machine accounts.</para></listitem>
			<listitem><para><emphasis>set to NULL</emphasis> user passwords.</para></listitem>
			<listitem><para><emphasis>manage</emphasis> interdomain trust accounts.</para></listitem>
		</itemizedlist>

		<para>
		To run smbpasswd as a normal user, just type:
		</para>

		<para>
<screen format="linespecific">
<prompt moreinfo="none">$ </prompt><userinput moreinfo="none">smbpasswd</userinput>
<prompt moreinfo="none">Old SMB password: </prompt><userinput moreinfo="none"><replaceable>secret</replaceable></userinput>
</screen>
		For <replaceable>secret</replaceable>, type the old value here or press return if
		there is no old password.
<screen format="linespecific">
<prompt moreinfo="none">New SMB Password: </prompt><userinput moreinfo="none"><replaceable>new secret</replaceable></userinput>
<prompt moreinfo="none">Repeat New SMB Password: </prompt><userinput moreinfo="none"><replaceable>new secret</replaceable></userinput>
</screen>
		</para>

		<para>
		If the old value does not match the current value stored for that user, or the two
		new values do not match each other, then the password will not be changed.
		</para>

		<para>
<indexterm significance="normal"><primary>SMB password</primary></indexterm>
		When invoked by an ordinary user, the command will allow only the user to change his or her own
		SMB password.
		</para>

		<para>
<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
<indexterm significance="normal"><primary>SMB password</primary></indexterm>
		When run by root, <literal>smbpasswd</literal> may take an optional argument specifying
		the username whose SMB password you wish to change. When run as root, <literal>smbpasswd</literal>
		does not prompt for or check the old password value, thus allowing root to set passwords
		for users who have forgotten their passwords.
		</para>

		<para>
<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
<indexterm significance="normal"><primary>passwd</primary></indexterm>
<indexterm significance="normal"><primary>yppasswd</primary></indexterm>
<indexterm significance="normal"><primary>change capabilities</primary></indexterm>
		<literal>smbpasswd</literal> is designed to work in the way familiar to UNIX
		users who use the <literal>passwd</literal> or <literal>yppasswd</literal> commands.
		While designed for administrative use, this tool provides essential user-level
		password change capabilities.
		</para>

		<para>
<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
		For more details on using <literal>smbpasswd</literal>, refer to the man page (the
		definitive reference).
		</para>
	</sect2>

	<sect2 id="pdbeditthing">
	<title>The <literal>pdbedit</literal> Tool</title>

		<para>
		<indexterm significance="normal"><primary>pdbedit</primary></indexterm>
		<indexterm significance="normal"><primary>User Management</primary></indexterm>
		<indexterm significance="normal"><primary>account policy</primary></indexterm>
		<indexterm significance="normal"><primary>User Accounts</primary><secondary>Adding/Deleting</secondary></indexterm>
		<literal>pdbedit</literal> is a tool that can be used only by root. It is used to
		manage the passdb backend, as well as domain-wide account policy settings. <literal>pdbedit</literal>
		can be used to:
		</para>

		<itemizedlist>
			<listitem><para>add, remove, or modify user accounts.</para></listitem>
			<listitem><para>list user accounts.</para></listitem>
			<listitem><para>migrate user accounts.</para></listitem>
			<listitem><para>migrate group accounts.</para></listitem>
			<listitem><para>manage account policies.</para></listitem>
			<listitem><para>manage domain access policy settings.</para></listitem>
		</itemizedlist>

		<para>
		<indexterm significance="normal"><primary>Sarbanes-Oxley</primary></indexterm>
		Under the terms of the Sarbanes-Oxley Act of 2002, American businesses and organizations are mandated to
		implement a series of <literal moreinfo="none">internal controls</literal> and procedures to communicate, store,
		and protect financial data. The Sarbanes-Oxley Act has far reaching implications in respect of:
		</para>

		<orderedlist inheritnum="ignore" continuation="restarts">
			<listitem><para>Who has access to information systems that store financial data.</para></listitem>
			<listitem><para>How personal and financial information is treated among employees and business
				partners.</para></listitem>
			<listitem><para>How security vulnerabilities are managed.</para></listitem>
			<listitem><para>Security and patch level maintenance for all information systems.</para></listitem>
			<listitem><para>How information systems changes are documented and tracked.</para></listitem>
			<listitem><para>How information access controls are implemented and managed.</para></listitem>
			<listitem><para>Auditability of all information systems in respect of change and security.</para></listitem>
			<listitem><para>Disciplinary procedures and controls to ensure privacy.</para></listitem>
		</orderedlist>

		<para>
		<indexterm significance="normal"><primary>accountability</primary></indexterm>
		<indexterm significance="normal"><primary>compliance</primary></indexterm>
		In short, the Sarbanes-Oxley Act of 2002 is an instrument that enforces accountability in respect of
		business related information systems so as to ensure the compliance of all information systems that
		are used to store personal information and particularly for financial records processing. Similar
		accountabilities are being demanded around the world.
		</para>

		<para>
		<indexterm significance="normal"><primary>laws</primary></indexterm>
		<indexterm significance="normal"><primary>regulations</primary></indexterm>
		<indexterm significance="normal"><primary>pdbedit</primary></indexterm>
		<indexterm significance="normal"><primary>access controls</primary></indexterm>
		<indexterm significance="normal"><primary>manage accounts</primary></indexterm>
		The need to be familiar with the Samba tools and facilities that permit information systems operation
		in compliance with government laws and regulations is clear to all. The <literal>pdbedit</literal> is
		currently the only Samba tool that provides the capacity to manage account and systems access controls
		and policies. During the remaining life-cycle of the Samba-3 series it is possible the new tools may
		be implemented to aid in this important area.
		</para>

		<para>
		Domain global policy controls available in Windows NT4 compared with Samba
		is shown in <link linkend="policycontrols">NT4 Domain v's Samba Policy Controls</link>.
		</para>

		<table id="policycontrols">
		<title>NT4 Domain v's Samba Policy Controls</title>
            <tgroup cols="5">
                <colspec align="left" colwidth="2*"/>
                <colspec align="left" colwidth="2*"/>
                <colspec align="center" colwidth="1*"/>
                <colspec align="center" colwidth="1*"/>
                <colspec align="center" colwidth="1*"/>
                <thead>
                    <row>
                        <entry><para>NT4 policy Name</para></entry>
                        <entry><para>Samba Policy Name</para></entry>
                        <entry><para>NT4 Range</para></entry>
                        <entry><para>Samba Range</para></entry>
                        <entry><para>Samba Default</para></entry>
                    </row>
                </thead>
                <tbody>
                    <row>
					<entry><para>Maximum Password Age</para></entry>
					<entry><para>maximum password age</para></entry>
					<entry><para>0 - 999 (days)</para></entry>
					<entry><para>0 - 4294967295 (sec)</para></entry>
					<entry><para>4294967295</para></entry>
                    </row>
                    <row>
					<entry><para>Minimum Password Age</para></entry>
					<entry><para>minimum password age</para></entry>
					<entry><para>0 - 999 (days)</para></entry>
					<entry><para>0 - 4294967295 (sec)</para></entry>
					<entry><para>0</para></entry>
                    </row>
                    <row>
					<entry><para>Mimimum Password Length</para></entry>
					<entry><para>min password length</para></entry>
					<entry><para>1 - 14 (Chars)</para></entry>
					<entry><para>0 - 4294967295 (Chars)</para></entry>
					<entry><para>5</para></entry>
                    </row>
                    <row>
					<entry><para>Password Uniqueness</para></entry>
					<entry><para>password history</para></entry>
					<entry><para>0 - 23 (#)</para></entry>
					<entry><para>0 - 4294967295 (#)</para></entry>
					<entry><para>0</para></entry>
                    </row>
                    <row>
					<entry><para>Account Lockout - Reset count after</para></entry>
					<entry><para>reset count minutes</para></entry>
					<entry><para>1 - 99998 (min)</para></entry>
					<entry><para>0 - 4294967295 (min)</para></entry>
					<entry><para>30</para></entry>
                    </row>
                    <row>
					<entry><para>Lockout after bad logon attempts</para></entry>
					<entry><para>bad lockout attempt</para></entry>
					<entry><para>0 - 998 (#)</para></entry>
					<entry><para>0 - 4294967295 (#)</para></entry>
					<entry><para>0</para></entry>
                    </row>
                    <row>
					<entry><para>*** Not Known ***</para></entry>
					<entry><para>disconnect time</para></entry>
					<entry><para>TBA</para></entry>
					<entry><para>0 - 4294967295</para></entry>
					<entry><para>0</para></entry>
                    </row>
                    <row>
					<entry><para>Lockout Duration</para></entry>
					<entry><para>lockout duration</para></entry>
					<entry><para>1 - 99998 (min)</para></entry>
					<entry><para>0 - 4294967295 (min)</para></entry>
					<entry><para>30</para></entry>
                    </row>
                    <row>
					<entry><para>Users must log on in order to change password</para></entry>
					<entry><para>user must logon to change password</para></entry>
					<entry><para>0/1</para></entry>
					<entry><para>0 - 4294967295</para></entry>
					<entry><para>0</para></entry>
                    </row>
                    <row>
					<entry><para>*** Registry Setting ***</para></entry>
					<entry><para>refuse machine password change</para></entry>
					<entry><para>0/1</para></entry>
					<entry><para>0 - 4294967295</para></entry>
					<entry><para>0</para></entry>
                    </row>
				</tbody>
			</tgroup>
		</table>

		<para>
		<indexterm significance="normal"><primary>pdbedit</primary></indexterm>
<indexterm significance="normal"><primary>policy settings</primary></indexterm>
<indexterm significance="normal"><primary>account security</primary></indexterm>
<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
		The <literal>pdbedit</literal> tool is the only one that can manage the account
		security and policy settings. It is capable of all operations that smbpasswd can
		do as well as a superset of them.
		</para>

		<para>
		<indexterm significance="normal"><primary>pdbedit</primary></indexterm>
<indexterm significance="normal"><primary>account import/export</primary></indexterm>
<indexterm significance="normal"><primary>passdb backend</primary></indexterm>
		One particularly important purpose of the <literal>pdbedit</literal> is to allow
		the import/export of account information from one passdb backend to another.
		</para>

		<sect3>
		<title>User Account Management</title>

		<para>
<indexterm significance="normal"><primary>pdbedit</primary></indexterm>
<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
<indexterm significance="normal"><primary>system accounts</primary></indexterm>
<indexterm significance="normal"><primary>user account</primary></indexterm>
<indexterm significance="normal"><primary>domain user manager</primary></indexterm>
<indexterm significance="normal"><primary>add user script</primary></indexterm>
<indexterm significance="normal"><primary>interface scripts</primary></indexterm>
		The <literal>pdbedit</literal> tool, like the <literal>smbpasswd</literal> tool, requires
		that a POSIX user account already exists in the UNIX/Linux system accounts database (backend).
		Neither tool will call out to the operating system to create a user account because this is
		considered to be the responsibility of the system administrator. When the Windows NT4 domain
		user manager is used to add an account, Samba will implement the <literal moreinfo="none">add user script</literal>
		(as well as the other interface scripts) to ensure that user, group and machine accounts are
		correctly created and changed. The use of the <literal>pdbedit</literal> tool does not
		make use of these interface scripts.
		</para>

		<para>
<indexterm significance="normal"><primary>pdbedit</primary></indexterm>
<indexterm significance="normal"><primary>POSIX account</primary></indexterm>
		Before attempting to use the <literal>pdbedit</literal> tool to manage user and machine
		accounts, make certain that a system (POSIX) account has already been created.
		</para>

		<sect4>
		<title>Listing User and Machine Accounts</title>

		<para>
<indexterm significance="normal"><primary>tdbsam</primary></indexterm>
<indexterm significance="normal"><primary>password backend</primary></indexterm>
		The following is an example of the user account information that is stored in
		a tdbsam password backend. This listing was produced by running:
<screen format="linespecific">
<prompt moreinfo="none">$ </prompt><userinput moreinfo="none">pdbedit -Lv met</userinput>
UNIX username:        met
NT username:          met
Account Flags:        [U          ]
User SID:             S-1-5-21-1449123459-1407424037-3116680435-2004
Primary Group SID:    S-1-5-21-1449123459-1407424037-3116680435-1201
Full Name:            Melissa E Terpstra
Home Directory:       \\frodo\met\Win9Profile
HomeDir Drive:        H:
Logon Script:         scripts\logon.bat
Profile Path:         \\frodo\Profiles\met
Domain:               MIDEARTH
Account desc:
Workstations:         melbelle
Munged dial:
Logon time:           0
Logoff time:          Mon, 18 Jan 2038 20:14:07 GMT
Kickoff time:         Mon, 18 Jan 2038 20:14:07 GMT
Password last set:    Sat, 14 Dec 2002 14:37:03 GMT
Password can change:  Sat, 14 Dec 2002 14:37:03 GMT
Password must change: Mon, 18 Jan 2038 20:14:07 GMT
</screen>
		</para>

		<para>
<indexterm significance="normal"><primary>smbpasswd format</primary></indexterm>
		Accounts can also be listed in the older <literal moreinfo="none">smbpasswd</literal> format:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">pdbedit -Lw</userinput>
root:0:84B0D8E14D158FF8417EAF50CFAC29C3:
     AF6DD3FD4E2EA8BDE1695A3F05EFBF52:[U          ]:LCT-42681AB8:
jht:1000:6BBC4159020A52741486235A2333E4D2:
     CC099521AD554A3C3CF2556274DBCFBC:[U          ]:LCT-40D75B5B:
rcg:1002:E95D4331A6F23AF8AAD3B435B51404EE:
     BB0F2C39B04CA6100F0E535DF8314B43:[U          ]:LCT-40D7C5A3:
afw:1003:1AAFA7F9F6DC1DEAAAD3B435B51404EE:
     CE92C2F9471594CDC4E7860CA6BC62DB:[T          ]:LCT-40DA501F:
met:1004:A2848CB7E076B435AAD3B435B51404EE:
     F25F5D3405085C555236B80B7B22C0D2:[U          ]:LCT-4244FAB8:
aurora$:1005:060DE593EA638B8ACC4A19F14D2FF2BB:
     060DE593EA638B8ACC4A19F14D2FF2BB:[W          ]:LCT-4173E5CC:
temptation$:1006:XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX:
     A96703C014E404E33D4049F706C45EE9:[W          ]:LCT-42BF0C57:
vaioboss$:1001:XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX:
     88A30A095160072784C88F811E89F98A:[W          ]:LCT-41C3878D:
frodo$:1008:15891DC6B843ECA41249940C814E316B:
     B68EADCCD18E17503D3DAD3E6B0B9A75:[W          ]:LCT-42B7979F:
marvel$:1011:BF709959C3C94E0B3958B7B84A3BB6F3:
     C610EFE9A385A3E8AA46ADFD576E6881:[W          ]:LCT-40F07A4
</screen>
<indexterm significance="normal"><primary>login id</primary></indexterm>
<indexterm significance="normal"><primary>UID</primary></indexterm>
<indexterm significance="normal"><primary>LanManger password</primary></indexterm>
<indexterm significance="normal"><primary>NT password</primary></indexterm>
<indexterm significance="normal"><primary>Account Flags</primary></indexterm>
<indexterm significance="normal"><primary>LCT</primary><see>last change time</see></indexterm>
		The account information that was returned by this command in order from left to right
		consists of the following colon separated data:
		</para>

		<itemizedlist>
			<listitem><para>Login ID.</para></listitem>
			<listitem><para>UNIX UID.</para></listitem>
			<listitem>
				<para>Microsoft LanManager password hash (password converted to upper-case then hashed.</para>
			</listitem>
			<listitem><para>Microsoft NT password hash (hash of the case-preserved password).</para></listitem>
			<listitem><para>Samba SAM Account Flags.</para></listitem>
			<listitem><para>The LCT data (password last change time).</para></listitem>
		</itemizedlist>

		<para>
<indexterm significance="normal"><primary>Account Flags</primary></indexterm>
<indexterm significance="normal"><primary>pdbedit</primary></indexterm>
		The Account Flags parameters are documented in the <literal>pdbedit</literal> man page, and are
		briefly documented in <link linkend="TOSHARG-acctflags">the Account Flags Management section</link>.
		</para>

		<para>
<indexterm significance="normal"><primary>last change time</primary></indexterm>
		The LCT data consists of 8 hexadecimal characters representing the time since January 1, 1970, of
		the time when the password was last changed.
		</para>

		</sect4>

		<sect4>
		<title>Adding User Accounts</title>

		<para>
<indexterm significance="normal"><primary>pdbedit</primary></indexterm>
<indexterm significance="normal"><primary>add a user account</primary></indexterm>
<indexterm significance="normal"><primary>standalone server</primary></indexterm>
<indexterm significance="normal"><primary>domain</primary></indexterm>
<indexterm significance="normal"><primary>SambaSAMAccount</primary></indexterm>
		The <literal>pdbedit</literal> can be used to add a user account to a standalone server
		or to a domain. In the example shown here the account for the user <literal moreinfo="none">vlaan</literal>
		has been created before attempting to add the SambaSAMAccount.
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> pdbedit -a vlaan
new password: secretpw
retype new password: secretpw
Unix username:        vlaan
NT username:          vlaan
Account Flags:        [U          ]
User SID:             S-1-5-21-726309263-4128913605-1168186429-3014
Primary Group SID:    S-1-5-21-726309263-4128913605-1168186429-513
Full Name:            Victor Laan
Home Directory:       \\frodo\vlaan
HomeDir Drive:        H:
Logon Script:         scripts\logon.bat
Profile Path:         \\frodo\profiles\vlaan
Domain:               MIDEARTH
Account desc:         Guest User
Workstations:
Munged dial:
Logon time:           0
Logoff time:          Mon, 18 Jan 2038 20:14:07 GMT
Kickoff time:         Mon, 18 Jan 2038 20:14:07 GMT
Password last set:    Wed, 29 Jun 2005 19:35:12 GMT
Password can change:  Wed, 29 Jun 2005 19:35:12 GMT
Password must change: Mon, 18 Jan 2038 20:14:07 GMT
Last bad password   : 0
Bad password count  : 0
Logon hours         : FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF
</screen>
		</para>

		</sect4>

		<sect4>
		<title>Deleting Accounts</title>

		<para>
<indexterm significance="normal"><primary>account deleted</primary></indexterm>
<indexterm significance="normal"><primary>SambaSAMAccount</primary></indexterm>
<indexterm significance="normal"><primary>pdbedit</primary></indexterm>
<indexterm significance="normal"><primary>passdb backend</primary></indexterm>
		An account can be deleted from the SambaSAMAccount database
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> pdbedit -x vlaan
</screen>
		The account is removed without further screen output. The account is removed only from the
		SambaSAMAccount (passdb backend) database, it is not removed from the UNIX account backend.
		</para>

		<para>
<indexterm significance="normal"><primary>delete user script</primary></indexterm>
<indexterm significance="normal"><primary>pdbedit</primary></indexterm>
		The use of the NT4 domain user manager to delete an account will trigger the <parameter moreinfo="none">delete user
		script</parameter>, but not the <literal>pdbedit</literal> tool.
		</para>

		</sect4>

		<sect4>
		<title>Changing User Accounts</title>

		<para>
<indexterm significance="normal"><primary>pdbedit</primary></indexterm>
		Refer to the <literal>pdbedit</literal> man page for a full synopsis of all operations
		that are available with this tool.
		</para>

		<para>
<indexterm significance="normal"><primary>pdbedit</primary></indexterm>
		An example of a simple change in the user account information is the change of the full name
		information shown here:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> pdbedit -r --fullname="Victor Aluicious Laan" vlaan
...
Primary Group SID:    S-1-5-21-726309263-4128913605-1168186429-513
Full Name:            Victor Aluicious Laan
Home Directory:       \\frodo\vlaan
...
</screen>
		</para>

		<para>
<indexterm significance="normal"><primary>grace time</primary></indexterm>
<indexterm significance="normal"><primary>password expired</primary></indexterm>
<indexterm significance="normal"><primary>expired password</primary></indexterm>
		Let us assume for a moment that a user's password has expired and the user is unable to
		change the password at this time. It may be necessary to give the user additional grace time
		so that it is possible to continue to work with the account and the original password. This
		demonstrates how the password expiration settings may be updated
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> pdbedit -Lv vlaan
...
Password last set:    Sun, 09 Sep 2001 22:21:40 GMT
Password can change:  Thu, 03 Jan 2002 15:08:35 GMT
Password must change: Thu, 03 Jan 2002 15:08:35 GMT
Last bad password   : Thu, 03 Jan 2002 15:08:35 GMT
Bad password count  : 2
...
</screen>
<indexterm significance="normal"><primary>bad logon attempts</primary></indexterm>
<indexterm significance="normal"><primary>lock the account</primary></indexterm>
		The user has recorded 2 bad logon attempts and the next will lock the account, but the
		password is also expired. Here is how this account can be reset:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> pdbedit -z vlaan
...
Password last set:    Sun, 09 Sep 2001 22:21:40 GMT
Password can change:  Thu, 03 Jan 2002 15:08:35 GMT
Password must change: Thu, 03 Jan 2002 15:08:35 GMT
Last bad password   : 0
Bad password count  : 0
...
</screen>
		The <literal moreinfo="none">Password must change:</literal> parameter can be reset like this:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> pdbedit --pwd-must-change-time=1200000000 vlaan
...
Password last set:    Sun, 09 Sep 2001 22:21:40 GMT
Password can change:  Thu, 03 Jan 2002 15:08:35 GMT
Password must change: Thu, 10 Jan 2008 14:20:00 GMT
...
</screen>
		Another way to use this tools is to set the date like this:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> pdbedit --pwd-must-change-time="2010-01-01" \
              --time-format="%Y-%m-%d" vlaan
...
Password last set:    Sun, 09 Sep 2001 22:21:40 GMT
Password can change:  Thu, 03 Jan 2002 15:08:35 GMT
Password must change: Fri, 01 Jan 2010 00:00:00 GMT
...
</screen>
<indexterm significance="normal"><primary>strptime</primary></indexterm>
<indexterm significance="normal"><primary>time format</primary></indexterm>
		Refer to the strptime man page for specific time format information.
		</para>

		<para>
<indexterm significance="normal"><primary>pdbedit</primary></indexterm>
<indexterm significance="normal"><primary>SambaSAMAccount</primary></indexterm>
		Please refer to the pdbedit man page for further information relating to SambaSAMAccount
		management.
		</para>

		<sect5 id="TOSHARG-acctflags">
		<title>Account Flags Management</title>

		<para>
<indexterm significance="normal"><primary>Samba SAM account flags</primary></indexterm>
<indexterm significance="normal"><primary>account control block</primary><see>ACB</see></indexterm>
<indexterm significance="normal"><primary>account encode_bits</primary></indexterm>
<indexterm significance="normal"><primary>account control flags</primary></indexterm>
		The Samba SAM account flags are properly called the ACB (account control block) within
		the Samba source code. In some parts of the Samba source code they are referred to as the
		account encode_bits, and also as the account control flags.
		</para>

		<para>
<indexterm significance="normal"><primary>pdbedit</primary></indexterm>
<indexterm significance="normal"><primary>user account</primary></indexterm>
<indexterm significance="normal"><primary>machine account</primary></indexterm>
<indexterm significance="normal"><primary>trust account</primary></indexterm>
<indexterm significance="normal"><primary>damaged data</primary></indexterm>
		The manual adjustment of user, machine (workstation or server) or an inter-domain trust
		account account flgas should not be necessary under normal conditions of use of Samba. On the other hand,
		where this information becomes corrupted for some reason, the ability to correct the damaged data is certainly
		useful. The tool of choice by which such correction can be affected is the <literal>pdbedit</literal> utility.
		</para>

		<para>
<indexterm significance="normal"><primary>account flags</primary></indexterm>
<indexterm significance="normal"><primary>LDAP directory</primary></indexterm>
		There have been a few requests for information regarding the account flags from developers
		who are creating their own Samba management tools. An example of a need for information regarding
		the proper management of the account flags is evident when developing scripts that will be used
		to manage an LDAP directory.
		</para>

		<para>
<indexterm significance="normal"><primary>pdbedit</primary></indexterm>
<indexterm significance="normal"><primary>account flag order</primary></indexterm>
		The account flag field can contain up to 16 characters. Presently, only 11 are in use.
		These are listed in <link linkend="accountflags">Samba SAM Account Control Block Flags</link>.
		The order in which the flags are specified to the <literal>pdbedit</literal> command is not important.
		In fact, they can be set without problem in any order in the SambaAcctFlags record in the LDAP directory.
		</para>

		<table frame="all" id="accountflags">
		<title>Samba SAM Account Control Block Flags</title>
        <tgroup cols="2" align="center">
            <thead>
            <row><entry align="center">Flag</entry><entry>Description</entry></row>
            </thead>
            <tbody>
            <row>
				<entry align="center">D</entry>
				<entry align="left">Account is disabled.</entry>
			</row>
            <row>
				<entry align="center">H</entry>
				<entry align="left">A home directory is required.</entry>
			</row>
            <row>
				<entry align="center">I</entry>
				<entry align="left">An inter-domain trust account.</entry>
			</row>
            <row>
				<entry align="center">L</entry>
				<entry align="left">Account has been auto-locked.</entry>
			</row>
            <row>
				<entry align="center">M</entry>
				<entry align="left">An MNS (Microsoft network service) logon account.</entry>
			</row>
            <row>
				<entry align="center">N</entry>
				<entry align="left">Password not required.</entry>
			</row>
            <row>
				<entry align="center">S</entry>
				<entry align="left">A server trust account.</entry>
			</row>
            <row>
				<entry align="center">T</entry>
				<entry align="left">Temporary duplicate account entry.</entry>
			</row>
            <row>
				<entry align="center">U</entry>
				<entry align="left">A normal user account.</entry>
			</row>
            <row>
				<entry align="center">W</entry>
				<entry align="left">A workstation trust account.</entry>
			</row>
            <row>
				<entry align="center">X</entry>
				<entry align="left">Password does not expire.</entry>
			</row>
            </tbody>
        </tgroup>
        </table>

		<para>
<indexterm significance="normal"><primary>pdbedit</primary></indexterm>
<indexterm significance="normal"><primary>account control flags</primary></indexterm>
		An example of use of the <literal>pdbedit</literal> utility to set the account control flags
		is shown here:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> pdbedit -r -c "[DLX]" jht
Unix username:        jht
NT username:          jht
Account Flags:        [DHULX      ]
User SID:             S-1-5-21-729263-4123605-1186429-3000
Primary Group SID:    S-1-5-21-729263-4123605-1186429-513
Full Name:            John H Terpstra,Utah Office
Home Directory:       \\aurora\jht
HomeDir Drive:        H:
Logon Script:         scripts\logon.bat
Profile Path:         \\aurora\profiles\jht
Domain:               MIDEARTH
Account desc:         BluntObject
Workstations:
Logon time:           0
Logoff time:          Mon, 18 Jan 2038 20:14:07 GMT
Kickoff time:         0
Password last set:    Sun, 03 Jul 2005 23:19:18 GMT
Password can change:  Sun, 03 Jul 2005 23:19:18 GMT
Password must change: Mon, 18 Jan 2038 20:14:07 GMT
Last bad password   : 0
Bad password count  : 0
Logon hours         : FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF
</screen>
<indexterm significance="normal"><primary>default settings</primary></indexterm>
	The flags can be reset to the default settings by executing:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> pdbedit -r -c "[]" jht
Unix username:        jht
NT username:          jht
Account Flags:        [U          ]
User SID:             S-1-5-21-729263-4123605-1186429-3000
Primary Group SID:    S-1-5-21-729263-4123605-1186429-513
Full Name:            John H Terpstra,Utah Office
Home Directory:       \\aurora\jht
HomeDir Drive:        H:
Logon Script:         scripts\logon.bat
Profile Path:         \\aurora\profiles\jht
Domain:               MIDEARTH
Account desc:         BluntObject
Workstations:
Logon time:           0
Logoff time:          Mon, 18 Jan 2038 20:14:07 GMT
Kickoff time:         0
Password last set:    Sun, 03 Jul 2005 23:19:18 GMT
Password can change:  Sun, 03 Jul 2005 23:19:18 GMT
Password must change: Mon, 18 Jan 2038 20:14:07 GMT
Last bad password   : 0
Bad password count  : 0
Logon hours         : FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF
</screen>
		</para>

		</sect5>

		</sect4>

		<sect4>
		<title>Domain Account Policy Managment</title>

		<para>
<indexterm significance="normal"><primary>domain account access policies</primary></indexterm>
<indexterm significance="normal"><primary>access policies</primary></indexterm>
		To view the domain account access policies that may be configured execute:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> pdbedit -P ?
No account policy by that name
Account policy names are :
min password length
password history
user must logon to change password
maximum password age
minimum password age
lockout duration
reset count minutes
bad lockout attempt
disconnect time
refuse machine password change
</screen>
		</para>

		<para>
		Commands will be executed to establish controls for our domain as follows:
		</para>

		<orderedlist inheritnum="ignore" continuation="restarts">
			<listitem><para>min password length = 8 characters.</para></listitem>
			<listitem><para>password history = last 4 passwords.</para></listitem>
			<listitem><para>maximum password age = 90 days.</para></listitem>
			<listitem><para>minimum password age = 7 days.</para></listitem>
			<listitem><para>bad lockout attempt = 8 bad logon attempts.</para></listitem>
			<listitem><para>lockout duration = forever, account must be manually reenabled.</para></listitem>
		</orderedlist>

		<para>
		The following command execution will achieve these settings:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> pdbedit -P "min password length" -C 8
account policy value for min password length was 5
account policy value for min password length is now 8
<prompt moreinfo="none">root# </prompt> pdbedit -P "password history" -C 4
account policy value for password history was 0
account policy value for password history is now 4
<prompt moreinfo="none">root# </prompt> pdbedit -P "maximum password age" -C 7776000
account policy value for maximum password age was 4294967295
account policy value for maximum password age is now 7776000
<prompt moreinfo="none">root# </prompt> pdbedit -P "minimum password age" -C 7
account policy value for minimum password age was 0
account policy value for minimum password age is now 7
<prompt moreinfo="none">root# </prompt> pdbedit -P "bad lockout attempt" -C 8
account policy value for bad lockout attempt was 0
account policy value for bad lockout attempt is now 8
<prompt moreinfo="none">root# </prompt> pdbedit -P "lockout duration" -C -1
account policy value for lockout duration was 30
account policy value for lockout duration is now 4294967295
</screen>
		</para>

<note><para>
To set the maximum (infinite) lockout time use the value of -1.
</para></note>

<warning><para>
Account policies must be set individually on each PDC and BDC. At this time (Samba 3.0.11 to Samba 3.0.14a)
account policies are not replicated automatically. This may be fixed before Samba 3.0.20 ships or some
time there after. Please check the WHATSNEW.txt file in the Samba-3 tarball for specific update notiations
regarding this facility.
</para></warning>

		</sect4>

		</sect3>

		<sect3>
		<title>Account Import/Export</title>

		<para>
		<indexterm significance="normal"><primary>pdbedit</primary></indexterm>
<indexterm significance="normal"><primary>account import/export</primary></indexterm>
<indexterm significance="normal"><primary>authentication</primary></indexterm>
		The <literal>pdbedit</literal> tool allows import/export of authentication (account)
		databases from one backend to another. For example, to import/export accounts from an
		old <filename moreinfo="none">smbpasswd</filename> database to a <parameter moreinfo="none">tdbsam</parameter>
		backend:
		</para>

		<procedure>
			<step performance="required"><para>
<indexterm significance="normal"><primary>pdbedit</primary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">pdbedit -i smbpasswd -e tdbsam</userinput>
</screen>
			</para></step>

			<step performance="required"><para>
<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
			Replace the <parameter moreinfo="none">smbpasswd</parameter> with <parameter moreinfo="none">tdbsam</parameter> in the
			<parameter moreinfo="none">passdb backend</parameter> configuration in <filename moreinfo="none">smb.conf</filename>.
			</para></step>
		</procedure>

	</sect3>
	</sect2>
</sect1>

<sect1>
<title>Password Backends</title>

<para>
<indexterm significance="normal"><primary>account database</primary></indexterm>
<indexterm significance="normal"><primary>SMB/CIFS server</primary></indexterm>
Samba offers flexibility in backend account database design. The flexibility is immediately obvious as one
begins to explore this capability. Recent changes to Samba (since 3.0.23) have removed the mulitple backend
feature in order to simplify problems that broke some installations. This removal has made the internal
operation of Samba-3 more consistent and predictable.
</para>

<para>
<indexterm significance="normal"><primary>multiple backends</primary></indexterm>
<indexterm significance="normal"><primary>tdbsam databases</primary></indexterm>
Beginning with Samba 3.0.23 it is no longer possible to specify use of mulitple passdb backends. Earlier
versions of Samba-3 made it possible to specify multiple password backends, and even multiple
backends of the same type. The multiple passdb backend capability caused many problems with name to SID and
SID to name ID resolution.  The Samba team wrestled with the challenges and decided that this feature needed
to be removed.
</para>

	<sect2>
	<title>Plaintext</title>

		<para>
<indexterm significance="normal"><primary>user database</primary></indexterm>
<indexterm significance="normal"><primary>/etc/samba/smbpasswd</primary></indexterm>
<indexterm significance="normal"><primary>/etc/smbpasswd</primary></indexterm>
<indexterm significance="normal"><primary>password encryption</primary></indexterm>
<indexterm significance="normal"><primary>/etc/passwd</primary></indexterm>
<indexterm significance="normal"><primary>PAM</primary></indexterm>
		Older versions of Samba retrieved user information from the UNIX user database
		and eventually some other fields from the file <filename moreinfo="none">/etc/samba/smbpasswd</filename>
		or <filename moreinfo="none">/etc/smbpasswd</filename>. When password encryption is disabled, no
		SMB-specific data is stored at all. Instead, all operations are conducted via the way
		that the Samba host OS will access its <filename moreinfo="none">/etc/passwd</filename> database.
		On most Linux systems, for example, all user and group resolution is done via PAM.
		</para>

	</sect2>

	<sect2>
	<title>smbpasswd: Encrypted Password Database</title>

		<para>
		<indexterm significance="normal"><primary>SAM backend</primary><secondary>smbpasswd</secondary></indexterm>
<indexterm significance="normal"><primary>user account</primary></indexterm>
<indexterm significance="normal"><primary>LM/NT password hashes</primary></indexterm>
<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
		Traditionally, when configuring <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="ENCRYPTPASSWORDS" xlink:href="smb.conf.5.html#ENCRYPTPASSWORDS">encrypt passwords = yes</link>
		in Samba's <filename moreinfo="none">smb.conf</filename> file, user account information such as username, LM/NT password hashes,
		password change times, and account flags have been stored in the <filename moreinfo="none">smbpasswd(5)</filename>
		file. There are several disadvantages to this approach for sites with large numbers of users
		(counted in the thousands).
		</para>

		<itemizedlist>
		<listitem><para>
<indexterm significance="normal"><primary>lookups</primary></indexterm>
		The first problem is that all lookups must be performed sequentially. Given that
		there are approximately two lookups per domain logon (one during intial logon validation
		and one for a session connection setup, such as when mapping a network drive or printer), this
		is a performance bottleneck for large sites. What is needed is an indexed approach
		such as that used in databases.
		</para></listitem>

		<listitem><para>
<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
<indexterm significance="normal"><primary>replicate</primary></indexterm>
<indexterm significance="normal"><primary>rsync</primary></indexterm>
<indexterm significance="normal"><primary>ssh</primary></indexterm>
<indexterm significance="normal"><primary>custom scripts</primary></indexterm>
		The second problem is that administrators who desire to replicate an smbpasswd file
		to more than one Samba server are left to use external tools such as
		<literal>rsync(1)</literal> and <literal>ssh(1)</literal> and write custom,
		in-house scripts.
		</para></listitem>

		<listitem><para>
<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
<indexterm significance="normal"><primary>home directory</primary></indexterm>
<indexterm significance="normal"><primary>password expiration</primary></indexterm>
<indexterm significance="normal"><primary>relative identifier</primary></indexterm>
<indexterm significance="normal"><primary>relative identifier</primary><see>RID</see></indexterm>
		Finally, the amount of information that is stored in an smbpasswd entry leaves
		no room for additional attributes such as a home directory, password expiration time,
		or even a relative identifier (RID).
		</para></listitem>
		</itemizedlist>

		<para>
<indexterm significance="normal"><primary>user attributes</primary></indexterm>
<indexterm significance="normal"><primary>smbd</primary></indexterm>
<indexterm significance="normal"><primary>API</primary></indexterm>
<indexterm significance="normal"><primary>samdb interface</primary></indexterm>
		As a result of these deficiencies, a more robust means of storing user attributes
		used by smbd was developed. The API that defines access to user accounts
		is commonly referred to as the samdb interface (previously, this was called the passdb
		API and is still so named in the Samba source code trees).
		</para>

		<para>
<indexterm significance="normal"><primary>passdb backends</primary></indexterm>
<indexterm significance="normal"><primary>smbpasswd plaintext database</primary></indexterm>
<indexterm significance="normal"><primary>tdbsam</primary></indexterm>
<indexterm significance="normal"><primary>ldapsam</primary></indexterm>
<indexterm significance="normal"><primary>enterprise</primary></indexterm>
		Samba provides an enhanced set of passdb backends that overcome the deficiencies
		of the smbpasswd plaintext database. These are tdbsam and ldapsam.
		Of these, ldapsam will be of most interest to large corporate or enterprise sites.
		</para>

	</sect2>

	<sect2>
	<title>tdbsam</title>

		<para>
		<indexterm significance="normal"><primary>SAM backend</primary><secondary>tdbsam</secondary></indexterm>
<indexterm significance="normal"><primary>trivial database</primary><see>TDB</see></indexterm>
<indexterm significance="normal"><primary>machine account</primary></indexterm>
		Samba can store user and machine account data in a <quote>TDB</quote> (trivial database).
		Using this backend does not require any additional configuration. This backend is
		recommended for new installations that do not require LDAP.
		</para>

		<para>
<indexterm significance="normal"><primary>tdbsam</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>BDC</primary></indexterm>
<indexterm significance="normal"><primary>scalability</primary></indexterm>
		As a general guide, the Samba Team does not recommend using the tdbsam backend for sites
		that have 250 or more users. Additionally, tdbsam is not capable of scaling for use
		in sites that require PDB/BDC implementations that require replication of the account
		database. Clearly, for reason of scalability, the use of ldapsam should be encouraged.
		</para>

		<para>
<indexterm significance="normal"><primary>250-user limit</primary></indexterm>
<indexterm significance="normal"><primary>performance-based</primary></indexterm>
<indexterm significance="normal"><primary>tdbsam</primary></indexterm>
		The recommendation of a 250-user limit is purely based on the notion that this
		would generally involve a site that has routed networks, possibly spread across
		more than one physical location. The Samba Team has not at this time established
		the performance-based scalability limits of the tdbsam architecture.
		</para>

		<para>
<indexterm significance="normal"><primary>4,500 user accounts</primary></indexterm>
<indexterm significance="normal"><primary>passdb backend</primary></indexterm>
<indexterm significance="normal"><primary>tdbsam</primary></indexterm>
<indexterm significance="normal"><primary>SambaSAMAccount</primary></indexterm>
		There are sites that have thousands of users and yet require only one server.
		One site recently reported having 4,500 user accounts on one UNIX system and
		reported excellent performance with the <literal moreinfo="none">tdbsam</literal> passdb backend.
		The limitation of where the <literal moreinfo="none">tdbsam</literal> passdb backend can be used
		is not one pertaining to a limitation in the TDB storage system, it is based
		only on the need for a reliable distribution mechanism for the SambaSAMAccount
		backend.
		</para>

	</sect2>

	<sect2>
	<title>ldapsam</title>

		<para>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
<indexterm significance="normal"><primary>ldapsam</primary></indexterm>
		<indexterm significance="normal"><primary>SAM backend</primary><secondary>ldapsam</secondary></indexterm>
		There are a few points to stress that the ldapsam does not provide. The LDAP
		support referred to in this documentation does not include:
		</para>

		<itemizedlist>
			<listitem><para>A means of retrieving user account information from
			a Windows 200x Active Directory server.</para></listitem>
			<listitem><para>A means of replacing /etc/passwd.</para></listitem>
		</itemizedlist>

		<para>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
<indexterm significance="normal"><primary>NSS</primary></indexterm>
<indexterm significance="normal"><primary>PAM</primary></indexterm>
<indexterm significance="normal"><primary>LGPL</primary></indexterm>
		The second item can be accomplished by using LDAP NSS and PAM modules. LGPL versions of these libraries can be
		obtained from <ulink url="http://www.padl.com/">PADL Software</ulink>.  More information about the
		configuration of these packages may be found in <ulink url="http://safari.oreilly.com/?XmlId=1-56592-491-6">
		<emphasis>LDAP, System Administration</emphasis> by Gerald Carter, Chapter 6, Replacing NIS"</ulink>.
		</para>

		<para>
<indexterm significance="normal"><primary>LDAP directory</primary></indexterm>
<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
<indexterm significance="normal"><primary>directory server</primary></indexterm>
		This document describes how to use an LDAP directory for storing Samba user
		account information traditionally stored in the smbpasswd(5) file. It is
		assumed that the reader already has a basic understanding of LDAP concepts
		and has a working directory server already installed. For more information
		on LDAP architectures and directories, please refer to the following sites:
		</para>

		<itemizedlist>
			<listitem><para><ulink url="http://www.openldap.org/">OpenLDAP</ulink></para></listitem>
			<listitem><para><ulink url="http://www.sun.com/software/products/directory_srvr_ee/index.xml">
				Sun One Directory Server</ulink></para></listitem>
			<listitem><para><ulink url="http://www.novell.com/products/edirectory/">Novell eDirectory</ulink></para></listitem>
			<listitem><para><ulink url="http://www-306.ibm.com/software/tivoli/products/directory-server/">IBM
				Tivoli Directory Server</ulink></para></listitem>
			<listitem><para><ulink url="http://www.redhat.com/software/rha/directory/">Red Hat Directory
				Server</ulink></para></listitem>
			<listitem><para><ulink url="http://www.linuxsecurity.com/content/view/119229">Fedora Directory
				Server</ulink></para></listitem>
		</itemizedlist>

		<para>
		Two additional Samba resources that may prove to be helpful are:
		</para>

		<itemizedlist>
			<listitem><para>
<indexterm significance="normal"><primary>Samba-PDC-LDAP-HOWTO</primary></indexterm>
			The <ulink url="http://www.unav.es/cti/ldap-smb/ldap-smb-3-howto.html">Samba-PDC-LDAP-HOWTO</ulink>
			maintained by Ignacio Coupeau.
			</para></listitem>

			<listitem><para>
<indexterm significance="normal"><primary>IDEALX</primary></indexterm>
<indexterm significance="normal"><primary>NT migration scripts</primary></indexterm>
<indexterm significance="normal"><primary>smbldap-tools</primary></indexterm>
			The NT migration scripts from <ulink url="http://samba.idealx.org/">IDEALX</ulink> that are
			geared to manage users and groups in such a Samba-LDAP domain controller configuration.
			Idealx also produced the smbldap-tools and the Interactive Console Management tool.
			</para></listitem>
		</itemizedlist>

		<sect3>
		<title>Supported LDAP Servers</title>

			<para>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
<indexterm significance="normal"><primary>ldapsam</primary></indexterm>
<indexterm significance="normal"><primary>OpenLDAP</primary></indexterm>
<indexterm significance="normal"><primary>Netscape's Directory Server</primary></indexterm>
			The LDAP ldapsam code was developed and tested using the OpenLDAP 2.x server and
			client libraries. The same code should work with Netscape's Directory Server and client SDK.
			However, there are bound to be compile errors and bugs. These should not be hard to fix.
			Please submit fixes via the process outlined in <link linkend="bugreport">Reporting Bugs</link>.
			</para>

			<para>
			Samba is capable of working with any standards-compliant LDAP server.
			</para>

		</sect3>

		<sect3>
		<title>Schema and Relationship to the RFC 2307 posixAccount</title>


			<para>
			Samba-3.0 includes the necessary schema file for OpenLDAP 2.x in the
			<filename moreinfo="none">examples/LDAP/samba.schema</filename> directory of the source code distribution
			tarball. The schema entry for the sambaSamAccount ObjectClass is shown here:
<programlisting format="linespecific">
ObjectClass (1.3.6.1.4.1.7165.2.2.6 NAME 'sambaSamAccount' SUP top AUXILIARY
    DESC 'Samba-3.0 Auxiliary SAM Account'
    MUST ( uid $ sambaSID )
    MAY  ( cn $ sambaLMPassword $ sambaNTPassword $ sambaPwdLastSet $
          sambaLogonTime $ sambaLogoffTime $ sambaKickoffTime $
          sambaPwdCanChange $ sambaPwdMustChange $ sambaAcctFlags $
          displayName $ sambaHomePath $ sambaHomeDrive $ sambaLogonScript $
          sambaProfilePath $ description $ sambaUserWorkstations $
          sambaPrimaryGroupSID $ sambaDomainName ))
</programlisting>
			</para>

			<para>
<indexterm significance="normal"><primary>samba.schema</primary></indexterm>
<indexterm significance="normal"><primary>OpenLDAP</primary></indexterm>
<indexterm significance="normal"><primary>OID</primary></indexterm>
			The <filename moreinfo="none">samba.schema</filename> file has been formatted for OpenLDAP 2.0/2.1.
			The Samba Team owns the OID space used by the above schema and recommends its use.
			If you translate the schema to be used with Netscape DS, please submit the modified
			schema file as a patch to <ulink url="mailto:jerry@samba.org">jerry@samba.org</ulink>.
			</para>

			<para>
<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
<indexterm significance="normal"><primary>/etc/passwd</primary></indexterm>
<indexterm significance="normal"><primary>sambaSamAccount</primary></indexterm>
<indexterm significance="normal"><primary>AUXILIARY</primary></indexterm>
<indexterm significance="normal"><primary>ObjectClass</primary></indexterm>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
<indexterm significance="normal"><primary>RFC 2307.</primary></indexterm>
			Just as the smbpasswd file is meant to store information that provides information
			additional to  a user's <filename moreinfo="none">/etc/passwd</filename> entry, so is the sambaSamAccount
			object meant to supplement the UNIX user account information. A sambaSamAccount is an
			<constant>AUXILIARY</constant> ObjectClass, so it can be used to augment existing
			user account information in the LDAP directory, thus providing information needed
			for Samba account handling. However, there are several fields (e.g., uid) that overlap
			with the posixAccount ObjectClass outlined in RFC 2307. This is by design.
			</para>

			<para>
<indexterm significance="normal"><primary>account information</primary></indexterm>
<indexterm significance="normal"><primary>sambaSamAccount</primary></indexterm>
<indexterm significance="normal"><primary>posixAccount</primary></indexterm>
<indexterm significance="normal"><primary>ObjectClasses</primary></indexterm>
<indexterm significance="normal"><primary>smbd</primary></indexterm>
<indexterm significance="normal"><primary>getpwnam</primary></indexterm>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
<indexterm significance="normal"><primary>NIS</primary></indexterm>
<indexterm significance="normal"><primary>NSS</primary></indexterm>
			In order to store all user account information (UNIX and Samba) in the directory,
			it is necessary to use the sambaSamAccount and posixAccount ObjectClasses in
			combination. However, <literal>smbd</literal> will still obtain the user's UNIX account
			information via the standard C library calls, such as getpwnam().
			This means that the Samba server must also have the LDAP NSS library installed
			and functioning correctly. This division of information makes it possible to
			store all Samba account information in LDAP, but still maintain UNIX account
			information in NIS while the network is transitioning to a full LDAP infrastructure.
			</para>
		</sect3>

		<sect3>
		<title>OpenLDAP Configuration</title>

		<para>
<indexterm significance="normal"><primary>sambaSamAccount</primary></indexterm>
<indexterm significance="normal"><primary>OpenLDAP</primary></indexterm>
<indexterm significance="normal"><primary>slapd</primary></indexterm>
<indexterm significance="normal"><primary>samba.schema</primary></indexterm>
		To include support for the sambaSamAccount object in an OpenLDAP directory
		server, first copy the samba.schema file to slapd's configuration directory.
		The samba.schema file can be found in the directory <filename moreinfo="none">examples/LDAP</filename>
		in the Samba source distribution.
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">cp samba.schema /etc/openldap/schema/</userinput>
</screen>
		</para>

		<para>
<indexterm significance="normal"><primary>samba.schema</primary></indexterm>
<indexterm significance="normal"><primary>slapd.conf</primary></indexterm>
<indexterm significance="normal"><primary>sambaSamAccount</primary></indexterm>
<indexterm significance="normal"><primary>cosine.schema</primary></indexterm>
<indexterm significance="normal"><primary>uid</primary></indexterm>
<indexterm significance="normal"><primary>inetorgperson.schema</primary></indexterm>
<indexterm significance="normal"><primary>displayName</primary></indexterm>
<indexterm significance="normal"><primary>attribute</primary></indexterm>
		Next, include the <filename moreinfo="none">samba.schema</filename> file in <filename moreinfo="none">slapd.conf</filename>.
		The sambaSamAccount object contains two attributes that depend on other schema
		files. The <parameter moreinfo="none">uid</parameter> attribute is defined in <filename moreinfo="none">cosine.schema</filename> and
		the <parameter moreinfo="none">displayName</parameter> attribute is defined in the <filename moreinfo="none">inetorgperson.schema</filename>
		file. Both of these must be included before the <filename moreinfo="none">samba.schema</filename> file.
<programlisting format="linespecific">
## /etc/openldap/slapd.conf

## schema files (core.schema is required by default)
include	           /etc/openldap/schema/core.schema

## needed for sambaSamAccount
include            /etc/openldap/schema/cosine.schema
include            /etc/openldap/schema/inetorgperson.schema
include            /etc/openldap/schema/nis.schema
include            /etc/openldap/schema/samba.schema
....
</programlisting>
		</para>

		<para>
<indexterm significance="normal"><primary>sambaSamAccount</primary></indexterm>
<indexterm significance="normal"><primary>posixAccount</primary></indexterm>
<indexterm significance="normal"><primary>posixGroup</primary></indexterm>
<indexterm significance="normal"><primary>ObjectClasses</primary></indexterm>
		It is recommended that you maintain some indices on some of the most useful attributes,
		as in the following example, to speed up searches made on sambaSamAccount ObjectClasses
		(and possibly posixAccount and posixGroup as well):
		</para>

<para>
<programlisting format="linespecific">
# Indices to maintain
## required by OpenLDAP
index objectclass             eq

index cn                      pres,sub,eq
index sn                      pres,sub,eq
## required to support pdb_getsampwnam
index uid                     pres,sub,eq
## required to support pdb_getsambapwrid()
index displayName             pres,sub,eq

## uncomment these if you are storing posixAccount and
## posixGroup entries in the directory as well
##index uidNumber               eq
##index gidNumber               eq
##index memberUid               eq

index   sambaSID              eq
index   sambaPrimaryGroupSID  eq
index   sambaDomainName       eq
index   default               sub
</programlisting>
</para>

		<para>
		Create the new index by executing:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt>./sbin/slapindex -f slapd.conf
</screen>
		</para>

		<para>
		Remember to restart slapd after making these changes:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">/etc/init.d/slapd restart</userinput>
</screen>
		</para>

		</sect3>

		<sect3>
		<title>Initialize the LDAP Database</title>

		<para>
<indexterm significance="normal"><primary>LDAP database</primary></indexterm>
<indexterm significance="normal"><primary>account containers</primary></indexterm>
<indexterm significance="normal"><primary>LDIF file</primary></indexterm>
<indexterm significance="normal"><primary>DNS</primary></indexterm>
		Before you can add accounts to the LDAP database, you must create the account containers
		that they will be stored in. The following LDIF file should be modified to match your
		needs (DNS entries, and so on):
<programlisting format="linespecific">
# Organization for Samba Base
dn: dc=quenya,dc=org
objectclass: dcObject
objectclass: organization
dc: quenya
o: Quenya Org Network
description: The Samba-3 Network LDAP Example

# Organizational Role for Directory Management
dn: cn=Manager,dc=quenya,dc=org
objectclass: organizationalRole
cn: Manager
description: Directory Manager

# Setting up container for Users OU
dn: ou=People,dc=quenya,dc=org
objectclass: top
objectclass: organizationalUnit
ou: People

# Setting up admin handle for People OU
dn: cn=admin,ou=People,dc=quenya,dc=org
cn: admin
objectclass: top
objectclass: organizationalRole
objectclass: simpleSecurityObject
userPassword: {SSHA}c3ZM9tBaBo9autm1dL3waDS21+JSfQVz

# Setting up container for groups
dn: ou=Groups,dc=quenya,dc=org
objectclass: top
objectclass: organizationalUnit
ou: Groups

# Setting up admin handle for Groups OU
dn: cn=admin,ou=Groups,dc=quenya,dc=org
cn: admin
objectclass: top
objectclass: organizationalRole
objectclass: simpleSecurityObject
userPassword: {SSHA}c3ZM9tBaBo9autm1dL3waDS21+JSfQVz

# Setting up container for computers
dn: ou=Computers,dc=quenya,dc=org
objectclass: top
objectclass: organizationalUnit
ou: Computers

# Setting up admin handle for Computers OU
dn: cn=admin,ou=Computers,dc=quenya,dc=org
cn: admin
objectclass: top
objectclass: organizationalRole
objectclass: simpleSecurityObject
userPassword: {SSHA}c3ZM9tBaBo9autm1dL3waDS21+JSfQVz
</programlisting>
		</para>

		<para>
<indexterm significance="normal"><primary>userPassword</primary></indexterm>
<indexterm significance="normal"><primary>slappasswd</primary></indexterm>
		The userPassword shown above should be generated using <literal>slappasswd</literal>.
		</para>

		<para>
<indexterm significance="normal"><primary>LDIF</primary></indexterm>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
		The following command will then load the contents of the LDIF file into the LDAP
		database.
<indexterm significance="normal"><primary>slapadd</primary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">$ </prompt><userinput moreinfo="none">slapadd -v -l initldap.dif</userinput>
</screen>
		</para>

		<para>
		Do not forget to secure your LDAP server with an adequate access control list
		as well as an admin password.
		</para>

		<note><para>
<indexterm significance="normal"><primary>secrets.tdb</primary></indexterm>
		Before Samba can access the LDAP server, you need to store the LDAP admin password
		in the Samba-3 <filename moreinfo="none">secrets.tdb</filename> database by:
<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">smbpasswd -w <replaceable>secret</replaceable></userinput>
</screen>
		</para></note>

		</sect3>

		<sect3>
		<title>Configuring Samba</title>

			<para>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
<indexterm significance="normal"><primary>smbd</primary></indexterm>
			The following parameters are available in <filename moreinfo="none">smb.conf</filename> only if your version of Samba was built with
			LDAP support. Samba automatically builds with LDAP support if the LDAP libraries are found. The
			best method to verify that Samba was built with LDAP support is:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> smbd -b | grep LDAP
   HAVE_LDAP_H
   HAVE_LDAP
   HAVE_LDAP_DOMAIN2HOSTLIST
   HAVE_LDAP_INIT
   HAVE_LDAP_INITIALIZE
   HAVE_LDAP_SET_REBIND_PROC
   HAVE_LIBLDAP
   LDAP_SET_REBIND_PROC_ARGS
</screen>
			If the build of the <literal>smbd</literal> command you are using does not produce output
			that includes <literal moreinfo="none">HAVE_LDAP_H</literal> it is necessary to discover why the LDAP headers
			and libraries were not found during compilation.
			</para>

			<para>LDAP-related smb.conf options include these:
			<simplelist>
			<member><indexterm>passdb backend</indexterm><parameter>passdb backend = ldapsam:url</parameter></member>
			<member><indexterm>ldap admin dn</indexterm></member>
			<member><indexterm>ldap delete dn</indexterm></member>
			<member><indexterm>ldap filter</indexterm></member>
			<member><indexterm>ldap group suffix</indexterm></member>
			<member><indexterm>ldap idmap suffix</indexterm></member>
			<member><indexterm>ldap machine suffix</indexterm></member>
			<member><indexterm>ldap passwd sync</indexterm></member>
			<member><indexterm>ldap ssl</indexterm></member>
			<member><indexterm>ldap suffix</indexterm></member>
			<member><indexterm>ldap user suffix</indexterm></member>
			<member><indexterm>ldap replication sleep</indexterm></member>
			<member><indexterm>ldap timeout</indexterm></member>
			<member><indexterm>ldap page size</indexterm></member>
			</simplelist>
			</para>

			<para>
			These are described in the <filename moreinfo="none">smb.conf</filename> man page and so are not repeated here. However, an example
			for use with an LDAP directory is shown in <link linkend="confldapex">the Configuration with LDAP.</link>
			</para>

<example id="confldapex">
<title>Configuration with LDAP</title>
<simplelist>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>security</indexterm><parameter>security = user</parameter></member>
<member><indexterm>encrypt passwords</indexterm><parameter>encrypt passwords = yes</parameter></member>
<member><indexterm>netbios name</indexterm><parameter>netbios name = MORIA</parameter></member>
<member><indexterm>workgroup</indexterm><parameter>workgroup = NOLDOR</parameter></member>

<member># LDAP related parameters:</member>

<member># Define the DN used when binding to the LDAP servers.</member>
<member># The password for this DN is not stored in smb.conf</member>
<member># Set it using 'smbpasswd -w secret' to store the</member>
<member># passphrase in the secrets.tdb file.</member>
<member># If the "ldap admin dn" value changes, it must be reset.</member>
<member><indexterm>ldap admin dn</indexterm><parameter>ldap admin dn = "cn=Manager,dc=quenya,dc=org"</parameter></member>

<member># SSL directory connections can be configured by:</member>
<member># ('off', 'start tls', or 'on' (default))</member>
<member><indexterm>ldap ssl</indexterm><parameter>ldap ssl = start tls</parameter></member>

<member># syntax: passdb backend = ldapsam:ldap://server-name[:port]</member>
<member><indexterm>passdb backend</indexterm><parameter>passdb backend = ldapsam:ldap://frodo.quenya.org</parameter></member>

<member># smbpasswd -x delete the entire dn-entry</member>
<member><indexterm>ldap delete dn</indexterm><parameter>ldap delete dn = no</parameter></member>

<member># The machine and user suffix are added to the base suffix</member>
<member># wrote WITHOUT quotes. NULL suffixes by default</member>
<member><indexterm>ldap user suffix</indexterm><parameter>ldap user suffix = ou=People</parameter></member>
<member><indexterm>ldap group suffix</indexterm><parameter>ldap group suffix = ou=Groups</parameter></member>
<member><indexterm>ldap machine suffix</indexterm><parameter>ldap machine suffix = ou=Computers</parameter></member>

<member># Trust UNIX account information in LDAP</member>
<member>#  (see the smb.conf man page for details)</member>

<member># Specify the base DN to use when searching the directory</member>
<member><indexterm>ldap suffix</indexterm><parameter>ldap suffix = dc=quenya,dc=org</parameter></member>
</simplelist>
</example>

		</sect3>

		<sect3>
		<title>Accounts and Groups Management</title>

			<para>
			<indexterm significance="normal"><primary>User Management</primary></indexterm>
			<indexterm significance="normal"><primary>User Accounts</primary><secondary>Adding/Deleting</secondary></indexterm>
			Because user accounts are managed through the sambaSamAccount ObjectClass, you should
			modify your existing administration tools to deal with sambaSamAccount attributes.
			</para>

			<para>
<indexterm significance="normal"><primary>sambaSamAccount</primary></indexterm>
<indexterm significance="normal"><primary>/etc/openldap/sldap.conf</primary></indexterm>
<indexterm significance="normal"><primary>NSS</primary></indexterm>
			Machine accounts are managed with the sambaSamAccount ObjectClass, just
			like user accounts. However, it is up to you to store those accounts
			in a different tree of your LDAP namespace. You should use
			<quote>ou=Groups,dc=quenya,dc=org</quote> to store groups and
			<quote>ou=People,dc=quenya,dc=org</quote> to store users. Just configure your
			NSS and PAM accordingly (usually, in the <filename moreinfo="none">/etc/openldap/sldap.conf</filename>
			configuration file).
			</para>

			<para>
<indexterm significance="normal"><primary>POSIX</primary></indexterm>
<indexterm significance="normal"><primary>posixGroup</primary></indexterm>
<indexterm significance="normal"><primary>Domain Groups</primary></indexterm>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
			In Samba-3, the group management system is based on POSIX
			groups. This means that Samba makes use of the posixGroup ObjectClass.
			For now, there is no NT-like group system management (global and local
			groups). Samba-3 knows only about <constant>Domain Groups</constant>
			and, unlike MS Windows 2000 and Active Directory, Samba-3 does not
			support nested groups.
			</para>

		</sect3>

		<sect3>
		<title>Security and sambaSamAccount</title>


			<para>
<indexterm significance="normal"><primary>sambaSAMAccount</primary></indexterm>
			There are two important points to remember when discussing the security
			of sambaSAMAccount entries in the directory.
			</para>

			<itemizedlist>
				<listitem><para><emphasis>Never</emphasis> retrieve the SambaLMPassword or
<indexterm significance="normal"><primary>SambaNTPassword</primary></indexterm>
				SambaNTPassword attribute values over an unencrypted LDAP session.</para></listitem>
				<listitem><para><emphasis>Never</emphasis> allow non-admin users to
				view the SambaLMPassword or SambaNTPassword attribute values.</para></listitem>
			</itemizedlist>

			<para>
<indexterm significance="normal"><primary>clear-text</primary></indexterm>
<indexterm significance="normal"><primary>impersonate</primary></indexterm>
<indexterm significance="normal"><primary>LM/NT password hashes</primary></indexterm>
			These password hashes are clear-text equivalents and can be used to impersonate
			the user without deriving the original clear-text strings. For more information
			on the details of LM/NT password hashes, refer to <link linkend="passdb">the
			Account Information Database section</link>.
			</para>

			<para>
<indexterm significance="normal"><primary>encrypted session</primary></indexterm>
<indexterm significance="normal"><primary>StartTLS</primary></indexterm>
<indexterm significance="normal"><primary>LDAPS</primary></indexterm>
<indexterm significance="normal"><primary>secure communications</primary></indexterm>
			To remedy the first security issue, the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LDAPSSL" xlink:href="smb.conf.5.html#LDAPSSL">ldap ssl</link> <filename moreinfo="none">smb.conf</filename>
			parameter defaults to require an encrypted session (<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LDAPSSL" xlink:href="smb.conf.5.html#LDAPSSL">ldap    ssl = on</link>) using the default port of <constant>636</constant> when
			contacting the directory server. When using an OpenLDAP server, it
			is possible to use the StartTLS LDAP extended operation in the place of LDAPS.
			In either case, you are strongly encouraged to use secure communications protocols
			(so do not set <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LDAPSSL" xlink:href="smb.conf.5.html#LDAPSSL">ldap ssl = off</link>).
			</para>

			<para>
<indexterm significance="normal"><primary>LDAPS</primary></indexterm>
<indexterm significance="normal"><primary>StartTLS</primary></indexterm>
<indexterm significance="normal"><primary>LDAPv3</primary></indexterm>
			Note that the LDAPS protocol is deprecated in favor of the LDAPv3 StartTLS
			extended operation. However, the OpenLDAP library still provides support for
			the older method of securing communication between clients and servers.
			</para>

			<para>
<indexterm significance="normal"><primary>harvesting password hashes</primary></indexterm>
<indexterm significance="normal"><primary>ACL</primary></indexterm>
<indexterm significance="normal"><primary>slapd.conf</primary></indexterm>
			The second security precaution is to prevent non-administrative users from
			harvesting password hashes from the directory. This can be done using the
			following ACL in <filename moreinfo="none">slapd.conf</filename>:
			</para>

<para>
<programlisting format="linespecific">
## allow the "ldap admin dn" access, but deny everyone else
access to attrs=SambaLMPassword,SambaNTPassword
     by dn="cn=Samba Admin,ou=People,dc=quenya,dc=org" write
     by * none
</programlisting>
</para>

		</sect3>

		<sect3>
		<title>LDAP Special Attributes for sambaSamAccounts</title>

			<para> The sambaSamAccount ObjectClass is composed of the attributes shown in next tables: <link linkend="attribobjclPartA">Part A</link>, and <link linkend="attribobjclPartB">Part B</link>.
			</para>

			<table frame="all" id="attribobjclPartA">
				<title>Attributes in the sambaSamAccount ObjectClass (LDAP), Part A</title>
			<tgroup cols="2" align="justify">
					<colspec align="left"/>
					<colspec align="justify" colwidth="1*"/>
			<tbody>
				<row><entry><constant>sambaLMPassword</constant></entry><entry>The LanMan password 16-byte hash stored as a character
						representation of a hexadecimal string.</entry></row>
				<row><entry><constant>sambaNTPassword</constant></entry><entry>The NT password 16-byte hash stored as a character
						representation of a hexadecimal string.</entry></row>
				<row><entry><constant>sambaPwdLastSet</constant></entry><entry>The integer time in seconds since 1970 when the
						<constant>sambaLMPassword</constant> and <constant>sambaNTPassword</constant> attributes were last set.
				</entry></row>

				<row><entry><constant>sambaAcctFlags</constant></entry><entry>String of 11 characters surrounded by square brackets [ ]
						representing account flags such as U (user), W (workstation), X (no password expiration),
						I (domain trust account), H (home dir required), S (server trust account),
						and D (disabled).</entry></row>

				<row><entry><constant>sambaLogonTime</constant></entry><entry>Integer value currently unused.</entry></row>

				<row><entry><constant>sambaLogoffTime</constant></entry><entry>Integer value currently unused.</entry></row>

				<row><entry><constant>sambaKickoffTime</constant></entry><entry>Specifies the time (UNIX time format) when the user
				will be locked down and cannot login any longer. If this attribute is omitted, then the account will never expire.
				Using this attribute together with shadowExpire of the shadowAccount ObjectClass will enable accounts to
				expire completely on an exact date.</entry></row>

				<row><entry><constant>sambaPwdCanChange</constant></entry><entry>Specifies the time (UNIX time format)
				after which the user is allowed to change his password. If this attribute is not set, the user will be free
				to change his password whenever he wants.</entry></row>

				<row><entry><constant>sambaPwdMustChange</constant></entry><entry>Specifies the time (UNIX time format) when the user is
				forced to change his password. If this value is set to 0, the user will have to change his password at first login.
				If this attribute is not set, then the password will never expire.</entry></row>

				<row><entry><constant>sambaHomeDrive</constant></entry><entry>Specifies the drive letter to which to map the
				UNC path specified by sambaHomePath. The drive letter must be specified in the form <quote>X:</quote>
				where X is the letter of the drive to map. Refer to the <quote>logon drive</quote> parameter in the
				smb.conf(5) man page for more information.</entry></row>

				<row><entry><constant>sambaLogonScript</constant></entry><entry>The sambaLogonScript property specifies the path of
				the user's logon script, .CMD, .EXE, or .BAT file. The string can be null. The path
				is relative to the netlogon share. Refer to the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOGONSCRIPT" xlink:href="smb.conf.5.html#LOGONSCRIPT">logon script</link> parameter in the
				<filename moreinfo="none">smb.conf</filename> man page for more information.</entry></row>

				<row><entry><constant>sambaProfilePath</constant></entry><entry>Specifies a path to the user's profile.
				This value can be a null string, a local absolute path, or a UNC path. Refer to the
				<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOGONPATH" xlink:href="smb.conf.5.html#LOGONPATH">logon path</link> parameter in the <filename moreinfo="none">smb.conf</filename> man page for more information.</entry></row>

				<row><entry><constant>sambaHomePath</constant></entry><entry>The sambaHomePath property specifies the path of
				the home directory for the user. The string can be null. If sambaHomeDrive is set and specifies
				a drive letter, sambaHomePath should be a UNC path. The path must be a network
				UNC path of the form <filename moreinfo="none">\\server\share\directory</filename>. This value can be a null string.
				Refer to the <literal>logon home</literal> parameter in the <filename moreinfo="none">smb.conf</filename> man page for more information.
				</entry></row>
			</tbody>
			</tgroup></table>


			<table frame="all" id="attribobjclPartB">
				<title>Attributes in the sambaSamAccount ObjectClass (LDAP), Part B</title>
			<tgroup cols="2" align="justify">
					<colspec align="left"/>
					<colspec align="justify" colwidth="1*"/>
			<tbody>
				<row><entry><constant>sambaUserWorkstations</constant></entry><entry>Here you can give a comma-separated list of machines
				on which the user is allowed to login. You may observe problems when you try to connect to a Samba domain member.
				Because domain members are not in this list, the domain controllers will reject them. Where this attribute is omitted,
				the default implies no restrictions.
				</entry></row>

				<row><entry><constant>sambaSID</constant></entry><entry>The security identifier(SID) of the user.
				The Windows equivalent of UNIX UIDs.</entry></row>

				<row><entry><constant>sambaPrimaryGroupSID</constant></entry><entry>The security identifier (SID) of the primary group
				of the user.</entry></row>

				<row><entry><constant>sambaDomainName</constant></entry><entry>Domain the user is part of.</entry></row>
			</tbody>
			</tgroup></table>


			<para>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>sambaSamAccount</primary></indexterm>
			The majority of these parameters are only used when Samba is acting as a PDC of
			a domain (refer to <link linkend="samba-pdc">Domain Control</link>, for details on
			how to configure Samba as a PDC). The following four attributes
			are only stored with the sambaSamAccount entry if the values are non-default values:
			</para>

			<itemizedlist>
<indexterm significance="normal"><primary>sambaHomePath</primary></indexterm>
<indexterm significance="normal"><primary>sambaLogonScript</primary></indexterm>
<indexterm significance="normal"><primary>sambaProfilePath</primary></indexterm>
<indexterm significance="normal"><primary>sambaHomeDrive</primary></indexterm>
				<listitem><para>sambaHomePath</para></listitem>
				<listitem><para>sambaLogonScript</para></listitem>
				<listitem><para>sambaProfilePath</para></listitem>
				<listitem><para>sambaHomeDrive</para></listitem>
			</itemizedlist>

			<para>
<indexterm significance="normal"><primary>sambaSamAccount</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>smbHome</primary></indexterm>
			These attributes are only stored with the sambaSamAccount entry if
			the values are non-default values. For example, assume MORIA has now been
			configured as a PDC and that <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOGONHOME" xlink:href="smb.conf.5.html#LOGONHOME">logon home = \\%L\%u</link> was defined in
			its <filename moreinfo="none">smb.conf</filename> file. When a user named <quote>becky</quote> logs on to the domain,
			the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOGONHOME" xlink:href="smb.conf.5.html#LOGONHOME">logon home</link> string is expanded to \\MORIA\becky.
			If the smbHome attribute exists in the entry <quote>uid=becky,ou=People,dc=samba,dc=org</quote>,
			this value is used. However, if this attribute does not exist, then the value
			of the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOGONHOME" xlink:href="smb.conf.5.html#LOGONHOME">logon home</link> parameter is used in its place. Samba
			will only write the attribute value to the directory entry if the value is
			something other than the default (e.g., <filename moreinfo="none">\\MOBY\becky</filename>).
			</para>

		</sect3>

		<sect3>
		<title>Example LDIF Entries for a sambaSamAccount</title>

			<para>
			The following is a working LDIF that demonstrates the use of the SambaSamAccount ObjectClass:
<programlisting format="linespecific">
dn: uid=guest2, ou=People,dc=quenya,dc=org
sambaLMPassword: 878D8014606CDA29677A44EFA1353FC7
sambaPwdMustChange: 2147483647
sambaPrimaryGroupSID: S-1-5-21-2447931902-1787058256-3961074038-513
sambaNTPassword: 552902031BEDE9EFAAD3B435B51404EE
sambaPwdLastSet: 1010179124
sambaLogonTime: 0
objectClass: sambaSamAccount
uid: guest2
sambaKickoffTime: 2147483647
sambaAcctFlags: [UX         ]
sambaLogoffTime: 2147483647
sambaSID: S-1-5-21-2447931902-1787058256-3961074038-5006
sambaPwdCanChange: 0
</programlisting>
			</para>

			<para>
			The following is an LDIF entry for using both the sambaSamAccount and
			posixAccount ObjectClasses:
<programlisting format="linespecific">
dn: uid=gcarter, ou=People,dc=quenya,dc=org
sambaLogonTime: 0
displayName: Gerald Carter
sambaLMPassword: 552902031BEDE9EFAAD3B435B51404EE
sambaPrimaryGroupSID: S-1-5-21-2447931902-1787058256-3961074038-1201
objectClass: posixAccount
objectClass: sambaSamAccount
sambaAcctFlags: [UX         ]
userPassword: {crypt}BpM2ej8Rkzogo
uid: gcarter
uidNumber: 9000
cn: Gerald Carter
loginShell: /bin/bash
logoffTime: 2147483647
gidNumber: 100
sambaKickoffTime: 2147483647
sambaPwdLastSet: 1010179230
sambaSID: S-1-5-21-2447931902-1787058256-3961074038-5004
homeDirectory: /home/moria/gcarter
sambaPwdCanChange: 0
sambaPwdMustChange: 2147483647
sambaNTPassword: 878D8014606CDA29677A44EFA1353FC7
</programlisting>
		</para>

		</sect3>

		<sect3>
		<title>Password Synchronization</title>

		<para>
		Samba-3 and later can update the non-Samba (LDAP) password stored with an account. When
		using pam_ldap, this allows changing both UNIX and Windows passwords at once.
		</para>

		<para>The <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LDAPPASSWDSYNC" xlink:href="smb.conf.5.html#LDAPPASSWDSYNC">ldap passwd sync</link> options can have the  values shown in
		<link linkend="ldappwsync">Possible <emphasis>ldap passwd sync</emphasis> Values</link>.</para>

		<table frame="all" id="ldappwsync">
		<title>Possible <parameter moreinfo="none">ldap passwd sync</parameter> Values</title>
		<tgroup cols="2">
			<colspec align="left" colwidth="1*"/>
			<colspec align="justify" colwidth="4*"/>
		<thead>
			<row><entry align="left">Value</entry><entry align="center">Description</entry></row>
		</thead>
		<tbody>
		<row><entry>yes</entry><entry><para>When the user changes his password, update
			       <constant>SambaNTPassword</constant>, <constant>SambaLMPassword</constant>,
			       and the <constant>password</constant> fields.</para></entry></row>

		<row><entry>no</entry><entry><para>Only update <constant>SambaNTPassword</constant> and
					<constant>SambaLMPassword</constant>.</para></entry></row>

		<row><entry>only</entry><entry><para>Only update the LDAP password and let the LDAP server
			worry about the other fields.  This option is only available on some LDAP servers and
			only when the LDAP server supports LDAP_EXOP_X_MODIFY_PASSWD.</para></entry></row>
		</tbody>
		</tgroup>
		</table>


		<para>More information can be found in the <filename moreinfo="none">smb.conf</filename> man page.</para>

		</sect3>

		<sect3>
		<title>Using OpenLDAP Overlay for Password Syncronization</title>

		<para>
		Howard Chu has written a special overlay called <literal>smbk5pwd</literal>. This tool  modifies the
		<literal moreinfo="none">SambaNTPassword</literal>, <literal moreinfo="none">SambaLMPassword</literal> and <literal moreinfo="none">Heimdal</literal>
		hashes in an OpenLDAP entry when an LDAP_EXOP_X_MODIFY_PASSWD operation is performed.
		</para>

		<para>
		The overlay is shipped with OpenLDAP-2.3 and can be found in the
		<filename moreinfo="none">contrib/slapd-modules/smbk5pwd</filename> subdirectory. This module can also be used with
		OpenLDAP-2.2.
		</para>

		</sect3>

	</sect2>

</sect1>

<sect1>
<title>Common Errors</title>

	<sect2>
		<title>Users Cannot Logon</title>

		<para><quote>I've installed Samba, but now I can't log on with my UNIX account! </quote></para>

		<para>Make sure your user has been added to the current Samba <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PASSDBBACKEND" xlink:href="smb.conf.5.html#PASSDBBACKEND">passdb backend</link>.
		Read the <link linkend="acctmgmttools">Account Management Tools,</link> for details.</para>

	</sect2>

	<sect2>
	<title>Configuration of <parameter moreinfo="none">auth methods</parameter></title>

	<para>
	When explicitly setting an <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="AUTHMETHODS" xlink:href="smb.conf.5.html#AUTHMETHODS">auth methods</link> parameter,
	<parameter moreinfo="none">guest</parameter> must be specified as the first entry on the line <?latex --- ?>
	for example, <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="AUTHMETHODS" xlink:href="smb.conf.5.html#AUTHMETHODS">auth methods = guest sam</link>.
	</para>

	</sect2>

</sect1>

</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="groupmapping">
<chapterinfo>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
	<author>
		<firstname>Jean François</firstname><surname>Micouleau</surname>
	</author>
	<author>
<firstname>Gerald</firstname><surname>Carter</surname><othername>(Jerry)</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jerry@samba.org</email></address>
</affiliation></author>
</chapterinfo>
<title>Group Mapping: MS Windows and UNIX</title>


	<para>
<indexterm significance="preferred"><primary>groups</primary><secondary>mapping</secondary></indexterm>
<indexterm significance="normal"><primary>SID</primary></indexterm>
<indexterm significance="normal"><primary>associations</primary></indexterm>
<indexterm significance="normal"><primary>UNIX groups</primary></indexterm>
<indexterm significance="normal"><primary>groupmap</primary></indexterm>
<indexterm significance="normal"><primary>net</primary></indexterm>
	Starting with Samba-3, new group mapping functionality is available to create associations
	between Windows group SIDs and UNIX group GIDs. The <literal>groupmap</literal> subcommand
	included with the <application moreinfo="none">net</application> tool can be used to manage these associations.
	</para>

	<para>
<indexterm significance="normal"><primary>group mapping</primary></indexterm>
<indexterm significance="normal"><primary>domain groups</primary></indexterm>
	The new facility for mapping NT groups to UNIX system groups allows the administrator to decide
	which NT domain groups are to be exposed to MS Windows clients. Only those NT groups that map
	to a UNIX group that has a value other than the default (<constant>-1</constant>) will be exposed
	in group selection lists in tools that access domain users and groups.
	</para>

	<warning>
	<para>
	<indexterm significance="normal"><primary>domain admin group</primary></indexterm>
<indexterm significance="normal"><primary>Windows group</primary></indexterm>
	The <parameter moreinfo="none">domain admin group</parameter> parameter has been removed in Samba-3 and should no longer
	be specified in <filename moreinfo="none">smb.conf</filename>. In Samba-2.2.x, this parameter was used to give the listed users membership in the
	<constant>Domain Admins</constant> Windows group, which gave local admin rights on their workstations
	(in default configurations).
	</para>
	</warning>

<sect1>
<title>Features and Benefits</title>

	<para>
	Samba allows the administrator to create MS Windows NT4/200x group accounts and to
	arbitrarily associate them with UNIX/Linux group accounts.
	</para>

	<para>
	<indexterm significance="normal"><primary>UID</primary></indexterm>
	<indexterm significance="normal"><primary>GID</primary></indexterm>
	<indexterm significance="normal"><primary>idmap uid</primary></indexterm>
<indexterm significance="normal"><primary>MMC</primary></indexterm>
<indexterm significance="normal"><primary>winbindd</primary></indexterm>
<indexterm significance="normal"><primary>ID range</primary></indexterm>
<indexterm significance="normal"><primary>group accounts</primary></indexterm>
	Group accounts can be managed using the MS Windows NT4 or MS Windows 200x/XP Professional MMC tools.
	Appropriate interface scripts should be provided in <filename moreinfo="none">smb.conf</filename> if it is desired that UNIX/Linux system
	accounts should be automatically created when these tools are used. In the absence of these scripts, and
	so long as <literal>winbindd</literal> is running, Samba group accounts that are created using these
	tools will be allocated UNIX UIDs and GIDs from the ID range specified by the
	<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="IDMAPUID" xlink:href="smb.conf.5.html#IDMAPUID">idmap uid</link>/<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="IDMAPGID" xlink:href="smb.conf.5.html#IDMAPGID">idmap gid</link>
	parameters in the <filename moreinfo="none">smb.conf</filename> file.
	</para>

	<figure id="idmap-sid2gid" float="0">
		<title>IDMAP: Group SID-to-GID Resolution.</title>
		<mediaobject><imageobject role="html"><imagedata fileref="images/idmap-sid2gid.png" scale="50" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/idmap-sid2gid.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/idmap-sid2gid"/></imageobject></mediaobject>
	</figure>

	<figure id="idmap-gid2sid" float="0">
		<title>IDMAP: GID Resolution to Matching SID.</title>
	<mediaobject><imageobject role="html"><imagedata fileref="images/idmap-gid2sid.png" scale="50" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/idmap-gid2sid.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/idmap-gid2sid"/></imageobject></mediaobject>
	</figure>

	<para>
	<indexterm significance="normal"><primary>IDMAP</primary></indexterm>
<indexterm significance="normal"><primary>SID-to-GID</primary></indexterm>
<indexterm significance="normal"><primary>net</primary><secondary>groupmap</secondary></indexterm>
<indexterm significance="normal"><primary>group mappings</primary></indexterm>
	In both cases, when winbindd is not running, only locally resolvable groups can be recognized. Please refer to
	<link linkend="idmap-sid2gid">IDMAP: Group SID-to-GID Resolution</link> and <link linkend="idmap-gid2sid">IDMAP: GID Resolution to Matching SID</link>.  The <literal>net groupmap</literal> is
	used to establish UNIX group to NT SID mappings as shown in <link linkend="idmap-store-gid2sid">IDMAP: storing
	group mappings</link>.
	</para>

	<figure id="idmap-store-gid2sid" float="0">
		<title>IDMAP Storing Group Mappings.</title>
		<mediaobject><imageobject role="html"><imagedata fileref="images/idmap-store-gid2sid.png" scale="50" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/idmap-store-gid2sid.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/idmap-store-gid2sid"/></imageobject></mediaobject>
	</figure>

	<para>
	<indexterm significance="normal"><primary>groupadd</primary></indexterm>
	<indexterm significance="normal"><primary>groupdel</primary></indexterm>
<indexterm significance="normal"><primary>shadow utilities</primary></indexterm>
<indexterm significance="normal"><primary>groupmod</primary></indexterm>
	Administrators should be aware that where <filename moreinfo="none">smb.conf</filename> group interface scripts make
	direct calls to the UNIX/Linux system tools (the shadow utilities, <literal>groupadd</literal>,
	<literal>groupdel</literal>, and <literal>groupmod</literal>), the resulting UNIX/Linux group names will be subject
	to any limits imposed by these tools. If the tool does not allow uppercase characters
	or space characters, then the creation of an MS Windows NT4/200x-style group of
	<literal moreinfo="none">Engineering Managers</literal> will attempt to create an identically named
	UNIX/Linux group, an attempt that will of course fail.
	</para>

	<para>
	<indexterm significance="normal"><primary>GID</primary></indexterm>
	<indexterm significance="normal"><primary>SID</primary></indexterm>
	There are several possible workarounds for the operating system tools limitation. One
	method is to use a script that generates a name for the UNIX/Linux system group that
	fits the operating system limits and that then just passes the UNIX/Linux group ID (GID)
	back to the calling Samba interface. This will provide a dynamic workaround solution.
	</para>

	<para>
<indexterm significance="normal"><primary>net</primary><secondary>groupmap</secondary></indexterm>
	Another workaround is to manually create a UNIX/Linux group, then manually create the
	MS Windows NT4/200x group on the Samba server, and then use the <literal>net groupmap</literal>
	tool to connect the two to each other.
	</para>

</sect1>

<sect1>
<title>Discussion</title>

	<para>
<indexterm significance="normal"><primary>Windows NT4/200x</primary></indexterm>
<indexterm significance="normal"><primary>group privileges</primary></indexterm>
	When you install <application moreinfo="none">MS Windows NT4/200x</application> on a computer, the installation
	program creates default users and groups, notably the <constant>Administrators</constant> group,
	and gives that group privileges necessary to perform essential system tasks,
	such as the ability to change the date and time or to kill (or close) any process running on the
	local machine.
	</para>
	
	<para>
	<indexterm significance="normal"><primary>Administrator</primary></indexterm>
	The <constant>Administrator</constant> user is a member of the <constant>Administrators</constant> group, and thus inherits
	<constant>Administrators</constant> group privileges. If a <constant>joe</constant> user is created to be a member of the
	<constant>Administrators</constant> group, <constant>joe</constant> has exactly the same rights as the user
	<constant>Administrator</constant>.
	</para>

	<para>
<indexterm significance="normal"><primary>domain member</primary></indexterm>
<indexterm significance="normal"><primary>Domain Admins</primary></indexterm>
<indexterm significance="normal"><primary>inherits rights</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
	When an MS Windows NT4/200x/XP machine is made a domain member, the <quote>Domain Admins</quote> group of the
	PDC is added to the local <constant>Administrators</constant> group of the workstation. Every member of the
	<constant>Domain Admins</constant> group inherits the rights of the local <constant>Administrators</constant> group when
	logging on the workstation.
	</para>

	<para>
<indexterm significance="normal"><primary>Domain Admins</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
	The following steps describe how to make Samba PDC users members of the <constant>Domain Admins</constant> group.
	</para>

	<orderedlist inheritnum="ignore" continuation="restarts">
		<listitem><para>
		Create a UNIX group (usually in <filename moreinfo="none">/etc/group</filename>); let's call it <constant>domadm</constant>.
		</para></listitem>

		<listitem><para>
<indexterm significance="normal"><primary>/etc/group</primary></indexterm>
		Add to this group the users that must be <quote>Administrators</quote>. For example,
		if you want <constant>joe, john</constant>, and <constant>mary</constant> to be administrators,
		your entry in <filename moreinfo="none">/etc/group</filename> will look like this:
		</para>

		<para><programlisting format="linespecific">
		domadm:x:502:joe,john,mary
		</programlisting>
		</para></listitem>

		<listitem><para>
		Map this domadm group to the <quote>Domain Admins</quote> group by executing the command:
		</para>

		<para>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">net groupmap add ntgroup="Domain Admins" unixgroup=domadm rid=512 type=d</userinput>
</screen>
		</para>
		
		<para>
		<indexterm significance="normal"><primary>Domain Admins group</primary></indexterm>
		The quotes around <quote>Domain Admins</quote> are necessary due to the space in the group name.
		Also make sure to leave no white space surrounding the equal character (=).
		</para></listitem>
	</orderedlist>

	<para>
	Now <constant>joe, john</constant>, and <constant>mary</constant> are domain administrators.
	</para>

	<para>
	<indexterm significance="normal"><primary>groups</primary><secondary>domain</secondary></indexterm>
	It is possible to map any arbitrary UNIX group to any Windows NT4/200x group as well as
	to make any UNIX group a Windows domain group. For example, if you wanted to include a
	UNIX group (e.g., acct) in an ACL on a local file or printer on a Domain Member machine,
	you would flag that group as a domain group by running the following on the Samba PDC:
	</para>

	<para>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">net groupmap add rid=1000 ntgroup="Accounting" unixgroup=acct type=d</userinput>
</screen>
	The <literal moreinfo="none">ntgroup</literal> value must be in quotes if it contains space characters to prevent
	the space from being interpreted as a command delimiter.
	</para>

	<para>
<indexterm significance="normal"><primary>RID</primary></indexterm>
<indexterm significance="normal"><primary>assigned RID</primary></indexterm>
	Be aware that the RID parameter is an unsigned 32-bit integer that should
	normally start at 1000. However, this RID must not overlap with any RID assigned
	to a user. Verification for this is done differently depending on the passdb backend
	you are using. Future versions of the tools may perform the verification automatically,
	but for now the burden is on you.
	</para>

	<sect2>
	<title>Warning: User Private Group Problems</title>

	<para>
<indexterm significance="normal"><primary>group accounts</primary></indexterm>
<indexterm significance="normal"><primary>Red Hat Linux</primary></indexterm>
<indexterm significance="normal"><primary>private groups</primary></indexterm>
	Windows does not permit user and group accounts to have the same name.
	This has serious implications for all sites that use private group accounts.
	A private group account is an administrative practice whereby users are each
	given their own group account. Red Hat Linux, as well as several free distributions
	of Linux, by default create private groups.
	</para>

	<para>
<indexterm significance="normal"><primary>UNIX/Linux group</primary></indexterm>
<indexterm significance="normal"><primary>Windows group</primary></indexterm>
	When mapping a UNIX/Linux group to a Windows group account, all conflict can
	be avoided by assuring that the Windows domain group name does not overlap
	with any user account name.
	</para>

	</sect2>

	<sect2>
	<title>Nested Groups: Adding Windows Domain Groups to Windows Local Groups</title>

	<indexterm significance="normal"><primary>groups</primary><secondary>nested</secondary></indexterm>

	<para>
<indexterm significance="normal"><primary>nested groups</primary></indexterm>
	This functionality is known as <constant>nested groups</constant> and was first added to
	Samba-3.0.3.
	</para>

	<para>
<indexterm significance="normal"><primary>nested groups</primary></indexterm>
	All MS Windows products since the release of Windows NT 3.10 support the use of nested groups.
	Many Windows network administrators depend on this capability because it greatly simplifies security
	administration.
	</para>

	<para>
<indexterm significance="normal"><primary>nested group</primary></indexterm>
<indexterm significance="normal"><primary>group membership</primary></indexterm>
<indexterm significance="normal"><primary>domain security</primary></indexterm>
<indexterm significance="normal"><primary>domain member server</primary></indexterm>
<indexterm significance="normal"><primary>local groups</primary></indexterm>
<indexterm significance="normal"><primary>domain global groups</primary></indexterm>
<indexterm significance="normal"><primary>domain global users</primary></indexterm>
	The nested group architecture was designed with the premise that day-to-day user and group membership
	management should be performed on the domain security database. The application of group security
	should be implemented on domain member servers using only local groups. On the domain member server,
	all file system security controls are then limited to use of the local groups, which will contain
	domain global groups and domain global users.
	</para>

	<para>
<indexterm significance="normal"><primary>individual domain user</primary></indexterm>
<indexterm significance="normal"><primary>domain group settings</primary></indexterm>
<indexterm significance="normal"><primary>Account Unknown</primary></indexterm>
	You may ask, What are the benefits of this arrangement? The answer is obvious to those who have plumbed
	the dark depths of Windows networking architecture. Consider for a moment a server on which are stored
	200,000 files, each with individual domain user and domain group settings. The company that owns the
	file server is bought by another company, resulting in the server being moved to another location, and then
	it is made a member of a different domain. Who would you think now owns all the files and directories?
	Answer: Account Unknown.
	</para>

	<para>
<indexterm significance="normal"><primary>directory access control</primary></indexterm>
<indexterm significance="normal"><primary>local groups</primary></indexterm>
<indexterm significance="normal"><primary>ACL</primary></indexterm>
<indexterm significance="normal"><primary>Account Unknown</primary></indexterm>
	Unraveling the file ownership mess is an unenviable administrative task that can be avoided simply
	by using local groups to control all file and directory access control. In this case, only the members
	of the local groups will have been lost. The files and directories in the storage subsystem will still
	be owned by the local groups. The same goes for all ACLs on them. It is administratively much simpler
	to delete the <constant>Account Unknown</constant> membership entries inside local groups with appropriate
	entries for domain global groups in the new domain that the server has been made a member of.
	</para>

	<para>
<indexterm significance="normal"><primary>nested groups</primary></indexterm>
<indexterm significance="normal"><primary>administrative privileges</primary></indexterm>
<indexterm significance="normal"><primary>domain member workstations</primary></indexterm>
<indexterm significance="normal"><primary>domain member servers</primary></indexterm>
<indexterm significance="normal"><primary>member machine</primary></indexterm>
<indexterm significance="normal"><primary>full rights</primary></indexterm>
<indexterm significance="normal"><primary>Domain Admins</primary></indexterm>
<indexterm significance="normal"><primary>local administrative privileges</primary></indexterm>
	Another prominent example of the use of nested groups involves implementation of administrative privileges
	on domain member workstations and servers. Administrative privileges are given to all members of the
	built-in local group <constant>Administrators</constant> on each domain member machine. To ensure that all domain
	administrators have full rights on the member server or workstation, on joining the domain, the
	<constant>Domain Admins</constant> group is added to the local Administrators group. Thus everyone who is
	logged into the domain as a member of the Domain Admins group is also granted local administrative
	privileges on each domain member.
	</para>

	<para>
<indexterm significance="normal"><primary>nested groups</primary></indexterm>
<indexterm significance="normal"><primary>auxiliary members</primary></indexterm>
<indexterm significance="normal"><primary>/etc/group</primary></indexterm>
<indexterm significance="normal"><primary>winbind</primary></indexterm>
	UNIX/Linux has no concept of support for nested groups, and thus Samba has for a long time not supported
	them either. The problem is that you would have to enter UNIX groups as auxiliary members of a group in
	<filename moreinfo="none">/etc/group</filename>. This does not work because it was not a design requirement at the time
	the UNIX file system security model was implemented. Since Samba-2.2, the winbind daemon can provide
	<filename moreinfo="none">/etc/group</filename> entries on demand by obtaining user and group information from the domain
	controller that the Samba server is a member of.
	</para>

	<para>
<indexterm significance="normal"><primary>/etc/group</primary></indexterm>
<indexterm significance="normal"><primary>libnss_winbind</primary></indexterm>
<indexterm significance="normal"><primary>local groups</primary></indexterm>
<indexterm significance="normal"><primary>Domain Users</primary></indexterm>
<indexterm significance="normal"><primary>alias group</primary></indexterm>
	In effect, Samba supplements the <filename moreinfo="none">/etc/group</filename> data via the dynamic
	<literal>libnss_winbind</literal> mechanism. Beginning with Samba-3.0.3, this facility is used to provide
	local groups in the same manner as Windows. It works by expanding the local groups on the
	fly as they are accessed. For example, the <constant>Domain Users</constant> group of the domain is made
	a member of the local group <constant>demo</constant>. Whenever Samba needs to resolve membership of the
	<constant>demo</constant> local (alias) group, winbind asks the domain controller for demo members of the Domain Users
	group. By definition, it can only contain user objects, which can then be faked to be member of the
	UNIX/Linux group <constant>demo</constant>.
	</para>

	<para>
<indexterm significance="normal"><primary>nested groups</primary></indexterm>
<indexterm significance="normal"><primary>winbindd</primary></indexterm>
<indexterm significance="normal"><primary>NSS</primary></indexterm>
<indexterm significance="normal"><primary>winbind</primary></indexterm>
<indexterm significance="normal"><primary>local groups</primary></indexterm>
<indexterm significance="normal"><primary>Domain User Manager</primary></indexterm>
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>group</tertiary></indexterm>
	To enable the use of nested groups, <literal>winbindd</literal> must be used with NSS winbind.
	Creation and administration of the local groups is done best via the Windows Domain User Manager or its
	Samba equivalent, the utility <literal>net rpc group</literal>. Creating the local group
	<constant>demo</constant> is achieved by executing:
	<screen format="linespecific">
	<prompt moreinfo="none">root# </prompt> net rpc group add demo -L -Uroot%not24get
	</screen>
<indexterm significance="normal"><primary>addmem</primary></indexterm>
<indexterm significance="normal"><primary>delmem</primary></indexterm>
	Here the -L switch means that you want to create a local group. It may be necessary to add -S and -U
	switches for accessing the correct host with appropriate user or root privileges. Adding and removing
	group members can be done via the <constant>addmem</constant> and <constant>delmem</constant> subcommands of
	<literal>net rpc group</literal> command. For example, addition of <quote>DOM\Domain Users</quote> to the
	local group <constant>demo</constant> is done by executing:
	<screen format="linespecific">
	net rpc group addmem demo "DOM\Domain Users"
	</screen>
<indexterm significance="normal"><primary>getent group demo</primary></indexterm>
<indexterm significance="normal"><primary>trusted domain</primary></indexterm>
<indexterm significance="normal"><primary>foreign domain</primary></indexterm>
<indexterm significance="normal"><primary>local access permissions</primary></indexterm>
	Having completed these two steps, the execution of <literal>getent group demo</literal> will show demo
	members of the global <constant>Domain Users</constant> group as members of  the group
	<constant>demo</constant>.  This also works with any local or domain user. In case the domain DOM trusts
	another domain, it is also possible to add global users and groups of the trusted domain as members of
	<constant>demo</constant>. The users from the foreign domain who are members of the group that has been
	added to the <constant>demo</constant> group now have the same local access permissions as local domain
	users have. 
	</para>

	</sect2>

	<sect2>
	<title>Important Administrative Information</title>

	<para>
	Administrative rights are necessary in two specific forms:
	</para>

	<orderedlist inheritnum="ignore" continuation="restarts">
		<listitem><para>For Samba-3 domain controllers and domain member servers/clients.</para></listitem>
		<listitem><para>To manage domain member Windows workstations.</para></listitem>
	</orderedlist>

	<para>
<indexterm significance="normal"><primary>rights and privileges</primary></indexterm>
<indexterm significance="normal"><primary>domain member client</primary></indexterm>
<indexterm significance="normal"><primary>group account</primary></indexterm>
	Versions of Samba up to and including 3.0.10 do not provide a means for assigning rights and privileges
	that are necessary for system administration tasks from a Windows domain member client machine, so
	domain administration tasks such as adding, deleting, and changing user and group account information, and
	managing workstation domain membership accounts, can be handled by any account other than root.
	</para>

	<para>
<indexterm significance="normal"><primary>privilege management</primary></indexterm>
<indexterm significance="normal"><primary>delegated</primary></indexterm>
<indexterm significance="normal"><primary>Administrator</primary></indexterm>
	Samba-3.0.11 introduced a new privilege management interface (see <link linkend="rights">User Rights and Privileges</link>)
	that permits these tasks to be delegated to non-root (i.e., accounts other than the equivalent of the
	MS Windows Administrator) accounts.
	</para>

	<para>
<indexterm significance="normal"><primary>mapped</primary></indexterm>
<indexterm significance="normal"><primary>Domain Admins</primary></indexterm>
	Administrative tasks on a Windows domain member workstation can be done by anyone who is a member of the
	<constant>Domain Admins</constant> group. This group can be mapped to any convenient UNIX group.
	</para>

	<sect3>
	<title>Applicable Only to Versions Earlier than 3.0.11</title>

	<para>
<indexterm significance="normal"><primary>privilege</primary></indexterm>
	Administrative tasks on UNIX/Linux systems, such as adding users or groups, requires
	<constant>root</constant>-level privilege. The addition of a Windows client to a Samba domain involves the
	addition of a user account for the Windows client.
	</para>

	<para>
<indexterm significance="normal"><primary>system security</primary></indexterm>
<indexterm significance="normal"><primary>privileges</primary></indexterm>
	Many UNIX administrators continue to request that the Samba Team make it possible to add Windows workstations, or 
	the ability to add, delete, or modify user accounts, without requiring <constant>root</constant> privileges. 
	Such a request violates every understanding of basic UNIX system security.
	</para>

	<para>
<indexterm significance="normal"><primary>privileges</primary></indexterm>
<indexterm significance="normal"><primary>/etc/passwd</primary></indexterm>
<indexterm significance="normal"><primary>Domain Server Manager</primary></indexterm>
<indexterm significance="normal"><primary>Domain User Manager</primary></indexterm>
<indexterm significance="normal"><primary>manage share-level ACL</primary></indexterm>
<indexterm significance="normal"><primary>share-level ACLs</primary></indexterm>
	There is no safe way to provide access on a UNIX/Linux system without providing
	<constant>root</constant>-level privileges. Provision of <constant>root</constant> privileges can be done
	either by logging on to the Domain as the user <constant>root</constant> or by permitting particular users to
	use a UNIX account that has a UID=0 in the <filename moreinfo="none">/etc/passwd</filename> database. Users of such accounts
	can use tools like the NT4 Domain User Manager and the NT4 Domain Server Manager to manage user and group
	accounts as well as domain member server and client accounts. This level of privilege is also needed to manage
	share-level ACLs.
	</para>

	</sect3>

	</sect2>

	<sect2>
	<title>Default Users, Groups, and Relative Identifiers</title>

	<para>
	<indexterm significance="normal"><primary>Relative Identifier</primary><see>RID</see></indexterm>
	<indexterm significance="normal"><primary>RID</primary></indexterm>
<indexterm significance="normal"><primary>Windows NT4/200x/XP</primary></indexterm>
<indexterm significance="normal"><primary>well-known RID</primary></indexterm>
<indexterm significance="normal"><primary>domain groups</primary></indexterm>
<indexterm significance="normal"><primary>tdbsam</primary></indexterm>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
<indexterm significance="normal"><primary>NT groups</primary></indexterm>
	When first installed, Windows NT4/200x/XP are preconfigured with certain user, group, and
	alias entities. Each has a well-known RID. These must be preserved for continued
	integrity of operation. Samba must be provisioned with certain essential domain groups that require
	the appropriate RID value. When Samba-3 is configured to use <constant>tdbsam</constant>, the essential
	domain groups are automatically created. It is the LDAP administrator's responsibility to create
	(provision) the default NT groups.
	</para>

	<para>
<indexterm significance="normal"><primary>default users</primary></indexterm>
<indexterm significance="normal"><primary>default groups</primary></indexterm>
<indexterm significance="normal"><primary>default aliases</primary></indexterm>
<indexterm significance="normal"><primary>RID</primary></indexterm>
	Each essential domain group must be assigned its respective well-known RID. The default users, groups,
	aliases, and RIDs are shown in <link linkend="WKURIDS">Well-Known User Default RIDs</link>.
	</para>

	<note><para>
<indexterm significance="normal"><primary>passdb backend</primary></indexterm>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
<indexterm significance="normal"><primary>ldapsam</primary></indexterm>
<indexterm significance="normal"><primary>domain groups</primary></indexterm>
<indexterm significance="normal"><primary>RID</primary></indexterm>
	It is the administrator's responsibility to create the essential domain groups and to assign each
	its default RID.
	</para></note>

	<para>
<indexterm significance="normal"><primary>domain groups</primary></indexterm>
<indexterm significance="normal"><primary>RID</primary></indexterm>
	It is permissible to create any domain group that may be necessary; just make certain that the essential
	domain groups (well known) have been created and assigned their default RIDs. Other groups you create may
	be assigned any arbitrary RID you care to use.
	</para>

	<para>
	Be sure to map each domain group to a UNIX system group. That is the only way to ensure that the group
	will be available for use as an NT domain group.
	</para>

	<para>
	<table frame="all" id="WKURIDS">
	<title>Well-Known User Default RIDs</title>
		<tgroup cols="4" align="left">
			<colspec align="left"/>
			<colspec align="left"/>
			<colspec align="left"/>
			<colspec align="center"/>
			<thead>
				<row>
					<entry>Well-Known Entity</entry>
					<entry>RID</entry>
					<entry>Type</entry>
					<entry>Essential</entry>
				</row>
			</thead>
			<tbody>
				<row>
					<entry>Domain Administrator</entry>
					<entry>500</entry>
					<entry>User</entry>
					<entry>No</entry>
				</row>
				<row>
					<entry>Domain Guest</entry>
					<entry>501</entry>
					<entry>User</entry>
					<entry>No</entry>
				</row>
				<row>
					<entry>Domain KRBTGT</entry>
					<entry>502</entry>
					<entry>User</entry>
					<entry>No</entry>
				</row>
				<row>
					<entry>Domain Admins</entry>
					<entry>512</entry>
					<entry>Group</entry>
					<entry>Yes</entry>
				</row>
				<row>
					<entry>Domain Users</entry>
					<entry>513</entry>
					<entry>Group</entry>
					<entry>Yes</entry>
				</row>
				<row>
					<entry>Domain Guests</entry>
					<entry>514</entry>
					<entry>Group</entry>
					<entry>Yes</entry>
				</row>
				<row>
					<entry>Domain Computers</entry>
					<entry>515</entry>
					<entry>Group</entry>
					<entry>No</entry>
				</row>
				<row>
					<entry>Domain Controllers</entry>
					<entry>516</entry>
					<entry>Group</entry>
					<entry>No</entry>
				</row>
				<row>
					<entry>Domain Certificate Admins</entry>
					<entry>517</entry>
					<entry>Group</entry>
					<entry>No</entry>
				</row>
				<row>
					<entry>Domain Schema Admins</entry>
					<entry>518</entry>
					<entry>Group</entry>
					<entry>No</entry>
				</row>
				<row>
					<entry>Domain Enterprise Admins</entry>
					<entry>519</entry>
					<entry>Group</entry>
					<entry>No</entry>
				</row>
				<row>
					<entry>Domain Policy Admins</entry>
					<entry>520</entry>
					<entry>Group</entry>
					<entry>No</entry>
				</row>
				<row>
					<entry>Builtin Admins</entry>
					<entry>544</entry>
					<entry>Alias</entry>
					<entry>No</entry>
				</row>
				<row>
					<entry>Builtin users</entry>
					<entry>545</entry>
					<entry>Alias</entry>
					<entry>No</entry>
				</row>
				<row>
					<entry>Builtin Guests</entry>
					<entry>546</entry>
					<entry>Alias</entry>
					<entry>No</entry>
				</row>
				<row>
					<entry>Builtin Power Users</entry>
					<entry>547</entry>
					<entry>Alias</entry>
					<entry>No</entry>
				</row>
				<row>
					<entry>Builtin Account Operators</entry>
					<entry>548</entry>
					<entry>Alias</entry>
					<entry>No</entry>
				</row>
				<row>
					<entry>Builtin System Operators</entry>
					<entry>549</entry>
					<entry>Alias</entry>
					<entry>No</entry>
				</row>
				<row>
					<entry>Builtin Print Operators</entry>
					<entry>550</entry>
					<entry>Alias</entry>
					<entry>No</entry>
				</row>
				<row>
					<entry>Builtin Backup Operators</entry>
					<entry>551</entry>
					<entry>Alias</entry>
					<entry>No</entry>
				</row>
				<row>
					<entry>Builtin Replicator</entry>
					<entry>552</entry>
					<entry>Alias</entry>
					<entry>No</entry>
				</row>
				<row>
					<entry>Builtin RAS Servers</entry>
					<entry>553</entry>
					<entry>Alias</entry>
					<entry>No</entry>
				</row>
			</tbody>
		</tgroup>
	</table>
	</para>

	</sect2>

	<sect2>
	<title>Example Configuration</title>

		<para>
<indexterm significance="normal"><primary>net</primary><secondary>groupmap</secondary><tertiary>list</tertiary></indexterm>
		You can list the various groups in the mapping database by executing
		<literal>net groupmap list</literal>. Here is an example:
		</para>

		<para>
<indexterm significance="normal"><primary>net</primary><secondary>groupmap</secondary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> <userinput moreinfo="none">net groupmap list</userinput>
Domain Admins (S-1-5-21-2547222302-1596225915-2414751004-512) -&gt; domadmin
Domain Users (S-1-5-21-2547222302-1596225915-2414751004-513) -&gt; domuser
Domain Guests (S-1-5-21-2547222302-1596225915-2414751004-514) -&gt; domguest
</screen>
		</para>

		<para>
		For complete details on <literal>net groupmap</literal>, refer to the net(8) man page.
		</para>

	</sect2>

</sect1>

<sect1>
<title>Configuration Scripts</title>

	<para>
	Everyone needs tools. Some of us like to create our own, others prefer to use canned tools
	(i.e., prepared by someone else for general use). 
	</para>

	<sect2>
	<title>Sample <filename moreinfo="none">smb.conf</filename> Add Group Script</title>

		<para>
		<indexterm significance="normal"><primary>smbgrpadd.sh</primary></indexterm>
		<indexterm significance="normal"><primary>groupadd limitations</primary></indexterm>
		<indexterm significance="normal"><primary>smbgrpadd.sh</primary></indexterm>
<indexterm significance="normal"><primary>/etc/group</primary></indexterm>
<indexterm significance="normal"><primary>groupadd</primary></indexterm>
		A script to create complying group names for use by the Samba group interfaces
		is provided in <link linkend="smbgrpadd.sh">smbgrpadd.sh</link>. This script
		adds a temporary entry in the <filename moreinfo="none">/etc/group</filename> file and then renames
		it to the desired name. This is an example of a method to get around operating
		system maintenance tool limitations such as those present in some version of the
		<literal>groupadd</literal> tool.
<example id="smbgrpadd.sh">
<title>smbgrpadd.sh</title>
<programlisting format="linespecific">
#!/bin/bash

# Add the group using normal system groupadd tool.
groupadd smbtmpgrp00

thegid=`cat /etc/group | grep ^smbtmpgrp00 | cut -d ":" -f3`

# Now change the name to what we want for the MS Windows networking end
cp /etc/group /etc/group.bak
cat /etc/group.bak | sed "s/^smbtmpgrp00/$1/g" &gt; /etc/group
rm /etc/group.bak

# Now return the GID as would normally happen.
echo $thegid
exit 0
</programlisting>
</example>
</para>

		<para>
		The <filename moreinfo="none">smb.conf</filename> entry for the above script shown in <link linkend="smbgrpadd">the configuration of
		<filename moreinfo="none">smb.conf</filename> for the add group Script</link> demonstrates how it may be used.

<example id="smbgrpadd">
<title>Configuration of <filename moreinfo="none">smb.conf</filename> for the add group Script</title>
<simplelist>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>add group script</indexterm><parameter>add group script = /path_to_tool/smbgrpadd.sh "%g"</parameter></member>
</simplelist>
</example>
		</para>

	</sect2>
	
	<sect2>
	<title>Script to Configure Group Mapping</title>

	<para>
<indexterm significance="normal"><primary>initGroups.sh</primary></indexterm>
	In our example we have created a UNIX/Linux group called <literal moreinfo="none">ntadmin</literal>.
	Our script will create the additional groups <literal moreinfo="none">Orks</literal>, <literal moreinfo="none">Elves</literal>, and <literal moreinfo="none">Gnomes</literal>.
	It is a good idea to save this shell script for later use just in case you ever need to rebuild your mapping database.
	For the sake of convenience we elect to save this script as a file called <filename moreinfo="none">initGroups.sh</filename>.
	This script is given in <link linkend="set-group-map">intGroups.sh</link>.
<indexterm significance="normal"><primary>initGroups.sh</primary></indexterm>
<example id="set-group-map">
<title>Script to Set Group Mapping</title>
<programlisting format="linespecific">
#!/bin/bash

net groupmap add ntgroup="Domain Admins" unixgroup=ntadmin rid=512 type=d
net groupmap add ntgroup="Domain Users" unixgroup=users rid=513 type=d
net groupmap add ntgroup="Domain Guests" unixgroup=nobody rid=514 type=d

groupadd Orks
groupadd Elves
groupadd Gnomes

net groupmap add ntgroup="Orks"   unixgroup=Orks   type=d
net groupmap add ntgroup="Elves"  unixgroup=Elves  type=d
net groupmap add ntgroup="Gnomes" unixgroup=Gnomes type=d
</programlisting>
</example>
	</para>

	<para>
	Of course it is expected that the administrator will modify this to suit local needs.
	For information regarding the use of the <literal>net groupmap</literal> tool please
	refer to the man page.
	</para>

	<note><para>
	Versions of Samba-3 prior to 3.0.23 automatically create default group mapping for the
	<literal moreinfo="none">Domain Admins, Domain Users</literal> and <literal moreinfo="none">Domain Guests</literal> Windows
	groups, but do not map them to UNIX GIDs. This was a cause of administrative confusion and 
	trouble. Commencing with Samba-3.0.23 this annomaly has been fixed - thus all Windows groups
	must now be manually and explicitly created and mapped to a valid UNIX GID by the Samba 
	administrator.
	</para></note>

	</sect2>

</sect1>

<sect1>
<title>Common Errors</title>

<para>
At this time there are many little surprises for the unwary administrator. In a real sense
it is imperative that every step of automated control scripts be carefully tested
manually before putting it into active service.
</para>

	<sect2>
	<title>Adding Groups Fails</title>

		<para>
<indexterm significance="normal"><primary>groupadd</primary></indexterm>
		This is a common problem when the <literal>groupadd</literal> is called directly
		by the Samba interface script for the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="ADDGROUPSCRIPT" xlink:href="smb.conf.5.html#ADDGROUPSCRIPT">add group script</link> in
		the <filename moreinfo="none">smb.conf</filename> file.
		</para>

		<para>
<indexterm significance="normal"><primary>uppercase character</primary></indexterm>
<indexterm significance="normal"><primary>space character</primary></indexterm>
		The most common cause of failure is an attempt to add an MS Windows group account
		that has an uppercase character and/or a space character in it.
		</para>

		<para>
<indexterm significance="normal"><primary>groupadd</primary></indexterm>
		There are three possible workarounds. First, use only group names that comply
		with the limitations of the UNIX/Linux <literal>groupadd</literal> system tool.
		Second, it involves the use of the script mentioned earlier in this chapter, and
		third is the option is to manually create a UNIX/Linux group account that can substitute
		for the MS Windows group name, then use the procedure listed above to map that group
		to the MS Windows group.
		</para>

	</sect2>

	<sect2>
	<title>Adding Domain Users to the Workstation Power Users Group</title>

		<para><quote>
		What must I do to add domain users to the Power Users group?
		</quote></para>

		<para>
<indexterm significance="normal"><primary>Domain Users group</primary></indexterm>
		The Power Users group is a group that is local to each Windows 200x/XP Professional workstation.
		You cannot add the Domain Users group to the Power Users group automatically, it must be done on
		each workstation by logging in as the local workstation <emphasis>administrator</emphasis> and
		then using the following procedure:
		</para>

		<procedure>
			<step performance="required"><para>
			Click <guimenu moreinfo="none">Start -&gt; Control Panel -&gt; Users and Passwords</guimenu>.
			</para></step>

			<step performance="required"><para>
			Click the <guimenuitem moreinfo="none">Advanced</guimenuitem> tab.
			</para></step>

			<step performance="required"><para>
			Click the <guibutton moreinfo="none">Advanced</guibutton> button.
			</para></step>

			<step performance="required"><para>
			Click <constant>Groups</constant>.
			</para></step>

			<step performance="required"><para>
			Double-click <constant>Power Users</constant>. This will launch the panel to add users or groups
			to the local machine <constant>Power Users</constant> group.
			</para></step>

			<step performance="required"><para>
			Click the <guibutton moreinfo="none">Add</guibutton> button.
			</para></step>

			<step performance="required"><para>
			Select the domain from which the <constant>Domain Users</constant> group is to be added.
			</para></step>

			<step performance="required"><para>
			Double-click the <constant>Domain Users</constant> group.
			</para></step>

			<step performance="required"><para>
			Click the <guibutton moreinfo="none">OK</guibutton> button. If a logon box is presented during this process, 
			please remember to enter the connect as <constant>DOMAIN\UserName</constant>, that is, for the
			domain <constant>MIDEARTH</constant> and the user <constant>root</constant> enter
			<constant>MIDEARTH\root</constant>.
			</para></step>
		</procedure>
	</sect2>

</sect1>

</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="NetCommand">
<chapterinfo>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
	<author>
<firstname>Volker</firstname><surname>Lendecke</surname>
<affiliation>
		<orgname>Samba Team</orgname>
		<address format="linespecific"><email>Volker.Lendecke@SerNet.DE</email></address>
</affiliation></author>
	<author>
<firstname>Guenther</firstname><surname>Deschner</surname>
<affiliation>
	<orgname>SuSE</orgname>
	<address format="linespecific"><email>gd@suse.de</email></address>
</affiliation></author>
	<pubdate>May 9, 2005</pubdate>
</chapterinfo>

<title>Remote and Local Management: The Net Command</title>

<para>
<indexterm significance="normal"><primary>net</primary></indexterm>
<indexterm significance="normal"><primary>remote management</primary></indexterm>
<indexterm significance="normal"><primary>command-line</primary></indexterm>
<indexterm significance="normal"><primary>scripted control</primary></indexterm>
The <literal>net</literal> command is one of the new features of Samba-3 and is an attempt to provide a useful
tool for the majority of remote management operations necessary for common tasks. The <literal>net</literal>
tool is flexible by design and is intended for command-line use as well as for scripted control application.
</para>

<para>
<indexterm significance="normal"><primary>net</primary></indexterm>
<indexterm significance="normal"><primary>network administrator's toolbox</primary></indexterm>
<indexterm significance="normal"><primary>smbgroupedit</primary></indexterm>
<indexterm significance="normal"><primary>rpcclient</primary></indexterm>
Originally introduced with the intent to mimic the Microsoft Windows command that has the same name, the
<literal>net</literal> command has morphed into a very powerful instrument that has become an essential part
of the Samba network administrator's toolbox. The Samba Team has introduced tools, such as
<literal>smbgroupedit</literal> and <literal>rpcclient</literal>, from which really useful capabilities have
been integrated into the <literal>net</literal>. The <literal>smbgroupedit</literal> command was absorbed
entirely into the <literal>net</literal>, while only some features of the <literal>rpcclient</literal> command
have been ported to it. Anyone who finds older references to these utilities and to the functionality they
provided should look at the <literal>net</literal> command before searching elsewhere.
</para>

<para>
A Samba-3 administrator cannot afford to gloss over this chapter because to do so will almost certainly cause
the infliction of self-induced pain, agony, and desperation. Be warned: this is an important chapter.
</para>

	<sect1>
	<title>Overview</title>

	<para>
<indexterm significance="normal"><primary>standalone</primary></indexterm>
<indexterm significance="normal"><primary>domain member</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>BDC</primary></indexterm>
<indexterm significance="normal"><primary>DMS</primary></indexterm>
<indexterm significance="normal"><primary>authentication</primary></indexterm>
	The tasks that follow the installation of a Samba-3 server, whether standalone or domain member, of a
	domain controller (PDC or BDC) begins with the need to create administrative rights. Of course, the
	creation of user and group accounts is essential for both a standalone server and a PDC.
	In the case of a BDC or a Domain Member server (DMS), domain user and group accounts are obtained from
	the central domain authentication backend.
	</para>

	<para>
<indexterm significance="normal"><primary>server type</primary></indexterm>
<indexterm significance="normal"><primary>local UNIX groups</primary></indexterm>
<indexterm significance="normal"><primary>mapped</primary></indexterm>
<indexterm significance="normal"><primary>domain global group</primary></indexterm>
<indexterm significance="normal"><primary>UID</primary></indexterm>
<indexterm significance="normal"><primary>GID</primary></indexterm>
<indexterm significance="normal"><primary>access rights</primary></indexterm>
<indexterm significance="normal"><primary>net</primary></indexterm>
	Regardless of the type of server being installed, local UNIX groups must be mapped to the Windows
	networking domain global group accounts. Do you ask why? Because Samba always limits its access to
	the resources of the host server by way of traditional UNIX UID and GID controls. This means that local
	groups must be mapped to domain global groups so that domain users who are members of the domain
	global groups can be given access rights based on UIDs and GIDs local to the server that is hosting
	Samba. Such mappings are implemented using the <literal>net</literal> command.
	</para>

	<para>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>BDC</primary></indexterm>
<indexterm significance="normal"><primary>DMS</primary></indexterm>
<indexterm significance="normal"><primary>security account</primary></indexterm>
<indexterm significance="normal"><primary>domain authentication</primary></indexterm>
<indexterm significance="normal"><primary>trust accounts</primary></indexterm>
<indexterm significance="normal"><primary>net</primary></indexterm>
	UNIX systems that are hosting a Samba-3 server that is running as a member (PDC, BDC, or DMS) must have
	a machine security account in the domain authentication database (or directory). The creation of such
	security (or trust) accounts is also handled using the <literal>net</literal> command.
	</para>

	<para>
<indexterm significance="normal"><primary>interdomain trusts</primary></indexterm>
<indexterm significance="normal"><primary>net</primary></indexterm>
<indexterm significance="normal"><primary>administrative duties</primary></indexterm>
<indexterm significance="normal"><primary>user management</primary></indexterm>
<indexterm significance="normal"><primary>group management</primary></indexterm>
<indexterm significance="normal"><primary>share management</primary></indexterm>
<indexterm significance="normal"><primary>printer management</primary></indexterm>
<indexterm significance="normal"><primary>printer migration</primary></indexterm>
<indexterm significance="normal"><primary>SID management</primary></indexterm>
	The establishment of interdomain trusts is achieved using the <literal>net</literal> command also, as
	may a plethora of typical administrative duties such as user management, group management, share and
	printer management, file and printer migration, security identifier management, and so on.
	</para>

	<para>
<indexterm significance="normal"><primary>net</primary></indexterm>
<indexterm significance="normal"><primary>man pages</primary></indexterm>
	The overall picture should be clear now: the <literal>net</literal> command plays a central role
	on the Samba-3 stage. This role will continue to be developed. The inclusion of this chapter is
	evidence of its importance, one that has grown in complexity to the point that it is no longer considered
	prudent to cover its use fully in the online UNIX man pages.
	</para>

	</sect1>
	
	<sect1>
	<title>Administrative Tasks and Methods</title>

	<para>
<indexterm significance="normal"><primary>net</primary></indexterm>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
<indexterm significance="normal"><primary>Distributed Computing Environment</primary><see>DCE</see></indexterm>
<indexterm significance="normal"><primary>Remote Procedure Call</primary><see>RPC</see></indexterm>
	The basic operations of the <literal>net</literal> command are documented here. This documentation is not
	exhaustive, and thus it is incomplete. Since the primary focus is on migration from Windows servers to a Samba
	server, the emphasis is on the use of the Distributed Computing Environment Remote Procedure Call (DCE RPC)
	mode of operation. When used against a server that is a member of an Active Directory domain, it is preferable
	(and often necessary) to use ADS mode operations. The <literal>net</literal> command supports both, but not
	for every operation. For most operations, if the mode is not specified, <literal>net</literal> will
	automatically fall back via the <constant>ads</constant>, <constant>rpc</constant>, and
	<constant>rap</constant> modes.  Please refer to the man page for a more comprehensive overview of the
	capabilities of this utility.
	</para>

	</sect1>

	<sect1>
	<title>UNIX and Windows Group Management</title>

	<para>
<indexterm significance="normal"><primary>Active Directory</primary></indexterm>
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary></indexterm>
<indexterm significance="normal"><primary>net</primary><secondary>ads</secondary></indexterm>
<indexterm significance="normal"><primary>net</primary><secondary>rap</secondary></indexterm>
<indexterm significance="normal"><primary>RAP</primary></indexterm>
	As stated, the focus in most of this chapter is on use of the <literal>net rpc</literal> family of
	operations that are supported by Samba. Most of them are supported by the <literal>net ads</literal>
	mode when used in connection with Active Directory. The <literal>net rap</literal> operating mode is
	also supported for some of these operations. RAP protocols are used by IBM OS/2 and by several
	earlier SMB servers.
	</para>

	<para>
<indexterm significance="normal"><primary>net</primary></indexterm>
<indexterm significance="normal"><primary>user management</primary></indexterm>
<indexterm significance="normal"><primary>group management</primary></indexterm>
	Samba's <literal>net</literal> tool implements sufficient capability to permit all common administrative
	tasks to be completed from the command line. In this section each of the essential user and group management
	facilities are explored.
	</para>

	<para>
<indexterm significance="normal"><primary>groups</primary></indexterm>
<indexterm significance="normal"><primary>domain</primary><secondary>groups</secondary></indexterm>
<indexterm significance="normal"><primary>local</primary><secondary>groups</secondary></indexterm>
<indexterm significance="normal"><primary>domain user accounts</primary></indexterm>
	Samba-3 recognizes two types of groups: <emphasis>domain groups</emphasis> and <emphasis>local
	groups</emphasis>. Domain groups can contain (have as members) only domain user accounts. Local groups
	can contain local users, domain users, and domain groups as members.
	</para>

	<para>
	The purpose of a local group is to permit file permission to be set for a group account that, like the
	usual UNIX/Linux group, is persistent across redeployment of a Windows file server.
	</para>

	<sect2>
	<title>Adding, Renaming, or Deletion of Group Accounts</title>

	<para>
	Samba provides file and print services to Windows clients. The file system resources it makes available
	to the Windows environment must, of necessity, be provided in a manner that is compatible with the
	Windows networking environment. UNIX groups are created and deleted as required to serve operational
	needs in the UNIX operating system and its file systems.
	</para>

	<para>
	In order to make available to the Windows environment, Samba has a facility by which UNIX groups can
	be mapped to a logical entity, called a Windows (or domain) group. Samba supports two types of Windows
	groups, local and global. Global groups can contain as members, global users. This membership is
	affected in the normal UNIX manner, but adding UNIX users to UNIX groups. Windows user accounts consist
	of a mapping between a user SambaSAMAccount (logical entity) and a UNIX user account. Therefore, 
	a UNIX user is mapped to a Windows user (i.e., is given a Windows user account and password) and the
	UNIX groups to which that user belongs, is mapped to a Windows group account. The result is that in
	the Windows account environment that user is also a member of the Windows group account by virtue
	of UNIX group memberships.
	</para>

	<para>
	The following sub-sections that deal with management of Windows groups demonstrates the relationship
	between the UNIX group account and its members to the respective Windows group accounts. It goes on to
	show how UNIX group members automatically pass-through to Windows group membership as soon as a logical
	mapping has been created.
	</para>

	<sect3>
	<title>Adding or Creating a New Group</title>

	<para>
	Before attempting to add a Windows group account, the currently available groups can be listed as shown
	here:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>group</tertiary></indexterm>
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>group list</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc group list -Uroot%not24get
Password:
Domain Admins
Domain Users
Domain Guests
Print Operators
Backup Operators
Replicator
Domain Computers
Engineers
</screen>
	</para>

<?latex \newpage ?>
	<para>
	A Windows group account called <quote>SupportEngrs</quote> can be added by executing the following
command:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>group add</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc group add "SupportEngrs" -Uroot%not24get
</screen>
	The addition will result in immediate availability of the new group account as validated by executing
this command:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc group list -Uroot%not24get
Password:
Domain Admins
Domain Users
Domain Guests
Print Operators
Backup Operators
Replicator
Domain Computers
Engineers
SupportEngrs
</screen>
	</para>

	<para>
<indexterm significance="normal"><primary>POSIX</primary></indexterm>
<indexterm significance="normal"><primary>smbldap-groupadd</primary></indexterm>
<indexterm significance="normal"><primary>getent</primary></indexterm>
	The following demonstrates that the POSIX (UNIX/Linux system account) group has been created by calling
	the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="ADDGROUPSCRIPT" xlink:href="smb.conf.5.html#ADDGROUPSCRIPT">add group script = /opt/IDEALX/sbin/smbldap-groupadd -p "%g"</link> interface
	script:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> getent group
...
Domain Admins:x:512:root
Domain Users:x:513:jht,lct,ajt,met
Domain Guests:x:514:
Print Operators:x:550:
Backup Operators:x:551:
Replicator:x:552:
Domain Computers:x:553:
Engineers:x:1002:jht
SupportEngrs:x:1003:
</screen>
	The following demonstrates that the use of the <literal>net</literal> command to add a group account
results in immediate mapping of the POSIX group that has been created to the Windows group account as shown
here:
<indexterm significance="normal"><primary>net</primary><secondary>groupmap</secondary><tertiary>list</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net groupmap list
Domain Admins (S-1-5-21-72630-4128915-11681869-512) -&gt; Domain Admins
Domain Users (S-1-5-21-72630-4128915-11681869-513) -&gt; Domain Users
Domain Guests (S-1-5-21-72630-4128915-11681869-514) -&gt; Domain Guests
Print Operators (S-1-5-21-72630-4128915-11681869-550) -&gt; Print Operators
Backup Operators (S-1-5-21-72630-4128915-11681869-551) -&gt; Backup Operators
Replicator (S-1-5-21-72630-4128915-11681869-552) -&gt; Replicator
Domain Computers (S-1-5-21-72630-4128915-11681869-553) -&gt; Domain Computers
Engineers (S-1-5-21-72630-4128915-11681869-3005) -&gt; Engineers
SupportEngrs (S-1-5-21-72630-4128915-11681869-3007) -&gt; SupportEngrs
</screen>
	</para>

	</sect3>

	<sect3>
	<title>Mapping Windows Groups to UNIX Groups</title>

	<para>
<indexterm significance="normal"><primary>mapped</primary></indexterm>
<indexterm significance="normal"><primary>Windows groups</primary></indexterm>
<indexterm significance="normal"><primary>system groups</primary></indexterm>
<indexterm significance="normal"><primary>access controls</primary></indexterm>
	Windows groups must be mapped to UNIX system (POSIX) groups so that file system access controls
	can be asserted in a manner that is consistent with the methods appropriate to the operating
	system that is hosting the Samba server.
	</para>

	<para>
<indexterm significance="normal"><primary>access controls</primary></indexterm>
<indexterm significance="normal"><primary>UID</primary></indexterm>
<indexterm significance="normal"><primary>GID</primary></indexterm>
<indexterm significance="normal"><primary>locally known UID</primary></indexterm>
	All file system (file and directory) access controls, within the file system of a UNIX/Linux server that is
	hosting a Samba server, are implemented using a UID/GID identity tuple. Samba does not in any way override
	or replace UNIX file system semantics. Thus it is necessary that all Windows networking operations that
	access the file system provide a mechanism that maps a Windows user to a particular UNIX/Linux group
	account. The user account must also map to a locally known UID. Note that the <literal>net</literal>
	command does not call any RPC-functions here but directly accesses the passdb.
	</para>

	<para>
<indexterm significance="normal"><primary>default mappings</primary></indexterm>
<indexterm significance="normal"><primary>Domain Admins</primary></indexterm>
<indexterm significance="normal"><primary>Domain Users</primary></indexterm>
<indexterm significance="normal"><primary>Domain Guests</primary></indexterm>
<indexterm significance="normal"><primary>Windows group</primary></indexterm>
<indexterm significance="normal"><primary>UNIX group</primary></indexterm>
<indexterm significance="normal"><primary>mapping</primary></indexterm>
	Samba depends on default mappings for the <constant>Domain Admins, Domain Users</constant>, and
	<constant>Domain Guests</constant> global groups. Additional groups may be added as shown in the
	examples just given. There are times when it is necessary to map an existing UNIX group account
	to a Windows group. This operation, in effect, creates a Windows group account as a consequence
	of creation of the mapping.
	</para>

	<para>
<indexterm significance="normal"><primary>net</primary><secondary>groupmap</secondary><tertiary>modify</tertiary></indexterm>
<indexterm significance="normal"><primary>net</primary><secondary>groupmap</secondary><tertiary>add</tertiary></indexterm>
<indexterm significance="normal"><primary>net</primary><secondary>groupmap</secondary><tertiary>delete</tertiary></indexterm>
	The operations that are permitted include: <constant>add</constant>, <constant>modify</constant>,
	and <constant>delete</constant>. An example of each operation is shown here.
	</para>

	<note><para>
	Commencing with Samba-3.0.23 Windows Domain Groups must be explicitly created. By default, all
	UNIX groups are exposed to Windows networking as Windows local groups.
	</para></note>

	<para>
	An existing UNIX group may be mapped to an existing Windows group by this example:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net groupmap modify ntgroup="Domain Users" unixgroup=users
</screen>
	An existing UNIX group may be mapped to a new Windows group as shown here:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net groupmap add ntgroup="EliteEngrs" unixgroup=Engineers type=d
</screen>
	Supported mapping types are 'd' (domain global) and 'l' (domain local).
	A Windows group may be deleted, and then a new Windows group can be mapped to the UNIX group by
	executing these commands:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net groupmap delete ntgroup=Engineers
<prompt moreinfo="none">root# </prompt> net groupmap add ntgroup=EngineDrivers unixgroup=Engineers type=d
</screen>
	The deletion and addition operations affected only the logical entities known as Windows groups, or domain
	groups. These operations are inert to UNIX system groups, meaning that they neither delete nor create UNIX
	system groups. The mapping of a UNIX group to a Windows group makes the UNIX group available as Windows
	groups so that files and folders on domain member clients (workstations and servers) can be given
	domain-wide access controls for domain users and groups.
	</para>

	<para>
	Two types of Windows groups can be created: <constant>domain (global)</constant> and <constant>local</constant>.
	In the previous examples the Windows groups created were of type <constant>domain</constant> or global. The
	following command will create a Windows group of type <constant>local</constant>.
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net groupmap add ntgroup=Pixies unixgroup=pixies type=l
</screen>
	Supported mapping types are 'd' (domain global) and 'l' (domain local), a domain local group in Samba is
	treated as local to the individual Samba server. Local groups can be used with Samba to enable multiple
	nested group support.
	</para>

	</sect3>

	<sect3>
	<title>Deleting a Group Account</title>

	<para>
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>group delete</tertiary></indexterm>
	A group account may be deleted by executing the following command:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc group delete SupportEngineers -Uroot%not24get
</screen>
	</para>

	<para>
	Validation of the deletion is advisable. The same commands may be executed as shown above.
	</para>

	</sect3>

	<sect3>
	<title>Rename Group Accounts</title>

	<note><para>
	This command is not documented in the man pages; it is implemented in the source code, but it does not
	work at this time. The example given documents, from the source code, how it should work. Watch the
	release notes of a future release to see when this may have been fixed.
	</para></note>

	<para>
	Sometimes it is necessary to rename a group account. Good administrators know how painful some managers'
	demands can be if this simple request is ignored. The following command demonstrates how the Windows group
	<quote>SupportEngrs</quote> can be renamed to <quote>CustomerSupport</quote>:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>group rename</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc group rename SupportEngrs \
    CustomerSupport -Uroot%not24get
</screen>
	</para>

	</sect3>

	</sect2>

	<sect2 id="grpmemshipchg">
	<title>Manipulating Group Memberships</title>

	<para>
	Three operations can be performed regarding group membership. It is possible to (1) add Windows users
	to a Windows group, to (2) delete Windows users from Windows groups, and to (3) list the Windows users that are
	members of a Windows group.
	</para>

	<para>
	To avoid confusion, it makes sense to check group membership before attempting to make any changes.
	The <literal>getent group</literal> will list UNIX/Linux group membership. UNIX/Linux group members are
	seen also as members of a Windows group that has been mapped using the <literal>net groupmap</literal>
	command (see <link linkend="groupmapping"/>). The following list of UNIX/Linux group membership shows
	that the user <constant>ajt</constant> is a member of the UNIX/Linux group <constant>Engineers</constant>.
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> getent group
...
Domain Admins:x:512:root
Domain Users:x:513:jht,lct,ajt,met,vlendecke
Domain Guests:x:514:
Print Operators:x:550:
Backup Operators:x:551:
Replicator:x:552:
Domain Computers:x:553:
Engineers:x:1000:jht,ajt
</screen>
	The UNIX/Linux groups have been mapped to Windows groups, as is shown here:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net groupmap list
Domain Admins (S-1-5-21-72630-412605-116429-512) -&gt; Domain Admins
Domain Users (S-1-5-21-72630-412605-116429-513) -&gt; Domain Users
Domain Guests (S-1-5-21-72630-412605-116429-514) -&gt; Domain Guests
Print Operators (S-1-5-21-72630-412605-116429-550) -&gt; Print Operators
Backup Operators (S-1-5-21-72630-412605-116429-551) -&gt; Backup Operators
Replicator (S-1-5-21-72630-412605-116429-552) -&gt; Replicator
Domain Computers (S-1-5-21-72630-412605-116429-553) -&gt; Domain Computers
Engineers (S-1-5-21-72630-412605-116429-3001) -&gt; Engineers
</screen>
	</para>

	<para>
	Given that the user <constant>ajt</constant> is already a member of the UNIX/Linux group and, via the
	group mapping, a member of the Windows group, an attempt to add this account again should fail. This is
	demonstrated here:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>group addmem</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc group addmem "MIDEARTH\Engineers" ajt -Uroot%not24get
Could not add ajt to MIDEARTH\Engineers: NT_STATUS_MEMBER_IN_GROUP
</screen>
	This shows that the group mapping between UNIX/Linux groups and Windows groups is effective and
	transparent.
	</para>

	<para>
	To permit the user <constant>ajt</constant> to be added using the <literal>net rpc group</literal> utility,
	this account must first be removed. The removal and confirmation of its effect is shown here:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>group delmem</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc group delmem "MIDEARTH\Engineers" ajt -Uroot%not24get
<prompt moreinfo="none">root# </prompt> getent group Engineers
Engineers:x:1000:jht
<prompt moreinfo="none">root# </prompt> net rpc group members Engineers -Uroot%not24get
MIDEARTH\jht
</screen>
	In this example both at the UNIX/Linux system level, the group no longer has the <constant>ajt</constant>
	as a member. The above also shows this to be the case for Windows group membership.
	</para>

	<para>
	The account is now added again, using the <literal>net rpc group</literal> utility:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc group addmem "MIDEARTH\Engineers" ajt -Uroot%not24get
<prompt moreinfo="none">root# </prompt> getent group Engineers
Engineers:x:1000:jht,ajt
<prompt moreinfo="none">root# </prompt> net rpc group members Engineers -Uroot%not24get
MIDEARTH\jht
MIDEARTH\ajt
</screen>
	</para>

	<para>
	In this example the members of the Windows <constant>Domain Users</constant> account are validated using
	the <literal>net rpc group</literal> utility. Note the this contents of the UNIX/Linux group was shown
	four paragraphs earlier. The Windows (domain) group membership is shown here:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>group members</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc group members "Domain Users" -Uroot%not24get
MIDEARTH\jht
MIDEARTH\lct
MIDEARTH\ajt
MIDEARTH\met
MIDEARTH\vlendecke
</screen>
	This express example shows that Windows group names are treated by Samba (as with
	MS Windows) in a case-insensitive manner:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc group members "DomAiN USerS" -Uroot%not24get
MIDEARTH\jht
MIDEARTH\lct
MIDEARTH\ajt
MIDEARTH\met
MIDEARTH\vlendecke
</screen>
	</para>

	<note><para>
	An attempt to specify the group name as <constant>MIDEARTH\Domain Users</constant> in place of
	just simply <constant>Domain Users</constant> will fail. The default behavior of the net rpc group
	is to direct the command at the local machine. The Windows group is treated as being local to the machine.
	If it is necessary to query another machine, its name can be specified using the <constant>-S
	servername</constant> parameter to the <literal>net</literal> command.
	</para></note>

	</sect2>

	<sect2 id="nestedgrpmgmgt">
	<title>Nested Group Support</title>

	<para>
	It is possible in Windows (and now in Samba also) to create a local group that has members (contains),
	domain users, and domain global groups.  Creation of the local group <constant>demo</constant> is
	achieved by executing:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc group add demo -L -S MORDON -Uroot%not24get
</screen>
	The -L switch means create a local group. Use the -S argument to direct the operation to a particular
	server. The parameters to the -U argument should be for a user who has appropriate administrative right
	and privileges on the machine.
	</para>

	<para>
	Addition and removal of group members can be achieved using the <constant>addmem</constant> and
	<constant>delmem</constant> subcommands of <literal>net rpc group</literal> command. For example,
	addition of <quote>DOM\Domain Users</quote> to the local group <constant>demo</constant> would be
	done by executing:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc group addmem demo "DOM\Domain Users" -Uroot%not24get
</screen>
	</para>

	<para>
	The members of a nested group can be listed by executing the following:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc group members demo -Uroot%not24get
DOM\Domain Users
DOM\Engineers
DOM\jamesf
DOM\jht
</screen>
	</para>

	<para>
	Nested group members can be removed (deleted) as shown here:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc group delmem demo "DOM\jht" -Uroot%not24get
</screen>
	</para>

	<sect3>
	<title>Managing Nest Groups on Workstations from the Samba Server</title>

	<para>
	Windows network administrators often ask on the Samba mailing list how it is possible to grant everyone
	administrative rights on their own workstation. This is of course a very bad practice, but commonly done
	to avoid user complaints. Here is how it can be done remotely from a Samba PDC or BDC:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>group addmem</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc group addmem "Administrators" "Domain Users" \
    -S WINPC032 -Uadministrator%secret
</screen>
	</para>

	<para>
	This can be scripted, and can therefore be performed as a user logs onto the domain from a Windows
	workstation. Here is a simple example that shows how this can be done.
	</para>

	<procedure>
	<title>Automating User Addition to the Workstation Power Users Group</title>

		<step performance="required"><para>
		Create the script shown in <link linkend="autopoweruserscript"/> and locate it in
		the directory <filename moreinfo="none">/etc/samba/scripts</filename>, named as <filename moreinfo="none">autopoweruser.sh</filename>.
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>group addmem</tertiary></indexterm>
<indexterm significance="normal"><primary>autopoweruser.sh</primary></indexterm>
<indexterm significance="normal"><primary>/etc/samba/scripts</primary></indexterm>
		</para></step>

<example id="autopoweruserscript">
<title>Script to Auto-add Domain Users to Workstation Power Users Group</title>
<screen format="linespecific">
#!/bin/bash

/usr/bin/net rpc group addmem "Power Users" "DOMAIN_NAME\$1" \
                   -UAdministrator%secret -S $2

exit 0
</screen>
</example>

		<step performance="required"><para>
		Set the permissions on this script to permit it to be executed as part of the logon process:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> chown root:root /etc/samba/autopoweruser.sh
<prompt moreinfo="none">root# </prompt> chmod 755 /etc/samba/autopoweruser.sh
</screen>
		</para></step>

		<step performance="required"><para>
		Modify the <filename moreinfo="none">smb.conf</filename> file so the <literal moreinfo="none">NETLOGON</literal> stanza contains the parameters
		shown in <link linkend="magicnetlogon">the Netlogon Example smb.conf file</link>.
		</para></step>

<example id="magicnetlogon">
<title>A Magic Netlogon Share</title>
<simplelist>
<member> </member><member><parameter>[netlogon]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = Netlogon Share</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /var/lib/samba/netlogon</parameter></member>
<member><indexterm>root preexec</indexterm><parameter>root preexec = /etc/samba/scripts/autopoweruser.sh %U %m</parameter></member>
<member><indexterm>read only</indexterm><parameter>read only = Yes</parameter></member>
<member><indexterm>guest ok</indexterm><parameter>guest ok = Yes</parameter></member>
</simplelist>
</example>

		<step performance="required"><para>
		Ensure that every Windows workstation Administrator account has the same password that you
		have used in the script shown in <link linkend="magicnetlogon">the Netlogon Example smb.conf
		file</link>
		</para></step>

</procedure>

	<para>
	This script will be executed every time a user logs on to the network. Therefore every user will
	have local Windows workstation management rights. This could of course be assigned using a group,
	in which case there is little justification for the use of this procedure. The key justification
	for the use of this method is that it will guarantee that all users have appropriate rights on
	the workstation.
	</para>

	</sect3>

	</sect2>

	</sect1>

	<sect1>
	<title>UNIX and Windows User Management</title>

	<para>
<indexterm significance="normal"><primary>user account</primary></indexterm>
<indexterm significance="normal"><primary>UNIX/Linux user account</primary></indexterm>
<indexterm significance="normal"><primary>UID</primary></indexterm>
<indexterm significance="normal"><primary>POSIX account</primary></indexterm>
<indexterm significance="normal"><primary>range</primary></indexterm>
<indexterm significance="normal"><primary>Windows user accounts</primary></indexterm>
<indexterm significance="normal"><primary>winbindd</primary></indexterm>
<indexterm significance="normal"><primary>account information</primary></indexterm>
	Every Windows network user account must be translated to a UNIX/Linux user account. In actual fact,
	the only account information the UNIX/Linux Samba server needs is a UID.  The UID is available either
	from a system (POSIX) account or from a pool (range) of UID numbers that is set aside for the purpose
	of being allocated for use by Windows user accounts. In the case of the UID pool, the UID for a
	particular user will be allocated by <literal>winbindd</literal>.
	</para>

	<para>
	Although this is not the appropriate place to discuss the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="USERNAMEMAP" xlink:href="smb.conf.5.html#USERNAMEMAP">username map</link> facility,
	this interface is an important method of mapping a Windows user account to a UNIX account that has a
	different name. Refer to the man page for the <filename moreinfo="none">smb.conf</filename> file for more information regarding this
	facility. User name mappings cannot be managed using the <literal>net</literal> utility.
	</para>

	<sect2 id="sbeuseraddn">
	<title>Adding User Accounts</title>

	<para>
	The syntax for adding a user account via the <literal>net</literal> (according to the man page) is shown
	here:
<screen format="linespecific">
net [&lt;method&gt;] user ADD &lt;name&gt; [-c container] [-F user flags] \
    [misc. options] [targets]
</screen>
	The user account password may be set using this syntax:
<screen format="linespecific">
net rpc password &lt;username&gt; [&lt;password&gt;] -Uadmin_username%admin_pass
</screen>
	</para>

	<para>
	The following demonstrates the addition of an account to the server <constant>FRODO</constant>:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>user add</tertiary></indexterm>
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>user password</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc user add jacko -S FRODO -Uroot%not24get
Added user jacko
</screen>
	The account password can be set with the following methods (all show the same operation):
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc password jacko f4sth0rse -S FRODO -Uroot%not24get
<prompt moreinfo="none">root# </prompt> net rpc user password jacko f4sth0rse \
    -S FRODO -Uroot%not24get
</screen>
	</para>

	</sect2>

	<sect2>
	<title>Deletion of User Accounts</title>

	<para>
	Deletion of a user account can be done using the following syntax:
<screen format="linespecific">
net [&lt;method&gt;] user DELETE &lt;name&gt; [misc. options] [targets]
</screen>
	The following command will delete the user account <constant>jacko</constant>:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>user delete</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc user delete jacko -Uroot%not24get
Deleted user account
</screen>
	</para>

	</sect2>

	<sect2>
	<title>Managing User Accounts</title>

	<para>
	Two basic user account operations are routinely used: change of password and querying which groups a user
	is a member of. The change of password operation is shown in <link linkend="sbeuseraddn"/>.
	</para>

	<para>
	The ability to query Windows group membership can be essential. Here is how a remote server may be
	interrogated to find which groups a user is a member of:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>user info</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc user info jacko -S SAURON -Uroot%not24get
net rpc user info jacko -S SAURON -Uroot%not24get
Domain Users
Domain Admins
Engineers
TorridGroup
BOP Shop
Emergency Services
</screen>
	</para>

	<para>
	It is also possible to rename user accounts:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>user rename</tertiary></indexterm>oldusername newusername
	Note that this operation does not yet work against Samba Servers. It is, however, possible to rename useraccounts on
	Windows Servers.

	</para>

	</sect2>

	<sect2>
	<title>User Mapping</title>

	<para>
<indexterm significance="normal"><primary>logon name</primary></indexterm>
<indexterm significance="normal"><primary>/etc/samba/smbusers</primary></indexterm>
<indexterm significance="normal"><primary>username map</primary></indexterm>
	In some situations it is unavoidable that a user's Windows logon name will differ from the login ID
	that user has on the Samba server. It is possible to create a special file on the Samba server that
	will permit the Windows user name to be mapped to a different UNIX/Linux user name. The <filename moreinfo="none">smb.conf</filename>
	file must also be amended so that the <constant>[global]</constant> stanza contains the parameter:
<screen format="linespecific">
username map = /etc/samba/smbusers
</screen>
	The content of the <filename moreinfo="none">/etc/samba/smbusers</filename> file is shown here:
<screen format="linespecific">
parsonsw: "William Parsons"
marygee: geeringm
</screen>
	In this example the Windows user account <quote>William Parsons</quote> will be mapped to the UNIX user
	<constant>parsonsw</constant>, and the Windows user account <quote>geeringm</quote> will be mapped to the
	UNIX user <constant>marygee</constant>.
	</para>

	</sect2>

	</sect1>

	<sect1>
	<title>Administering User Rights and Privileges</title>

	<para>
<indexterm significance="normal"><primary>credentials</primary></indexterm>
<indexterm significance="normal"><primary>manage printers</primary></indexterm>
<indexterm significance="normal"><primary>manage shares</primary></indexterm>
<indexterm significance="normal"><primary>manage groups</primary></indexterm>
<indexterm significance="normal"><primary>manage users</primary></indexterm>
	With all versions of Samba earlier than 3.0.11 the only account on a Samba server that could
	manage users, groups, shares, printers, and such was the <constant>root</constant> account. This caused
	problems for some users and was a frequent source of scorn over the necessity to hand out the
	credentials for the most security-sensitive account on a UNIX/Linux system.
	</para>

	<para>
<indexterm significance="normal"><primary>delegate administrative privileges</primary></indexterm>
<indexterm significance="normal"><primary>normal user</primary></indexterm>
<indexterm significance="normal"><primary>rights and privilege</primary></indexterm>
<indexterm significance="normal"><primary>privilege management</primary></indexterm>
<indexterm significance="normal"><primary>groups of users</primary></indexterm>
	New to Samba version 3.0.11 is the ability to delegate administrative privileges as necessary to either
	a normal user or to groups of users. The significance of the administrative privileges is documented
	in <link linkend="rights"/>. Examples of use of the <literal>net</literal> for user rights and privilege
	management is appropriate to this chapter.
	</para>

	<note><para>
	When user rights and privileges are correctly set, there is no longer a need for a Windows
	network account for the <constant>root</constant> user (nor for any synonym of it) with a UNIX UID=0.
	Initial user rights and privileges can be assigned by any account that is a member of the <constant>
	Domain Admins</constant> group. Rights can be assigned to user as well as group accounts.
	</para></note>

	<para>
	By default, no privileges and rights are assigned. This is demonstrated by executing the command
	shown here:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc rights list accounts -U root%not24get
BUILTIN\Print Operators
No privileges assigned

BUILTIN\Account Operators
No privileges assigned

BUILTIN\Backup Operators
No privileges assigned

BUILTIN\Server Operators
No privileges assigned

BUILTIN\Administrators
No privileges assigned

Everyone
No privileges assigned
</screen>
	</para>

	<para>
	The <literal>net</literal> command can be used to obtain the currently supported capabilities for rights
	and privileges using this method:
<indexterm significance="normal"><primary>SeMachineAccountPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SePrintOperatorPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeAddUsersPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeRemoteShutdownPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeDiskOperatorPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeBackupPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeRestorePrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeTakeOwnershipPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>rights list</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc rights list -U root%not24get
     SeMachineAccountPrivilege  Add machines to domain
      SePrintOperatorPrivilege  Manage printers
           SeAddUsersPrivilege  Add users and groups to the domain
     SeRemoteShutdownPrivilege  Force shutdown from a remote system
       SeDiskOperatorPrivilege  Manage disk shares
             SeBackupPrivilege  Back up files and directories
            SeRestorePrivilege  Restore files and directories
      SeTakeOwnershipPrivilege  Take ownership of files or other objects
</screen>
	Machine account privilege is necessary to permit a Windows NT4 or later network client to be added to the
	domain. The disk operator privilege is necessary to permit the user to manage share ACLs and file and
	directory ACLs for objects not owned by the user.
	</para>

	<para>
	In this example, all rights are assigned to the <constant>Domain Admins</constant> group. This is a good
	idea since members of this group are generally expected to be all-powerful. This assignment makes that
	the reality:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>rights grant</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc rights grant "MIDEARTH\Domain Admins" \
    SeMachineAccountPrivilege SePrintOperatorPrivilege \
    SeAddUsersPrivilege SeRemoteShutdownPrivilege \
    SeDiskOperatorPrivilege  -U root%not24get
Successfully granted rights.
</screen>
	Next, the domain user <constant>jht</constant> is given the privileges needed for day-to-day
	administration:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc rights grant "MIDEARTH\jht" \
    SeMachineAccountPrivilege SePrintOperatorPrivilege \
    SeAddUsersPrivilege SeDiskOperatorPrivilege \
    -U root%not24get
Successfully granted rights.
</screen>
	</para>

	<para>
	The following step permits validation of the changes just made:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>rights list accounts</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc rights list accounts -U root%not24get
MIDEARTH\jht
SeMachineAccountPrivilege
SePrintOperatorPrivilege
SeAddUsersPrivilege
SeDiskOperatorPrivilege

BUILTIN\Print Operators
No privileges assigned

BUILTIN\Account Operators
No privileges assigned

BUILTIN\Backup Operators
No privileges assigned

BUILTIN\Server Operators
No privileges assigned

BUILTIN\Administrators
No privileges assigned

Everyone
No privileges assigned

MIDEARTH\Domain Admins
SeMachineAccountPrivilege
SePrintOperatorPrivilege
SeAddUsersPrivilege
SeRemoteShutdownPrivilege
SeDiskOperatorPrivilege
</screen>
	</para>

	</sect1>

	<sect1>
	<title>Managing Trust Relationships</title>

	<para>
	There are essentially two types of trust relationships: the first is between domain controllers and domain
	member machines (network clients), the second is between domains (called interdomain trusts). All
	Samba servers that participate in domain security require a domain membership trust account, as do like
	Windows NT/200x/XP workstations.
	</para>

	<sect2>
	<title>Machine Trust Accounts</title>

	<para>
	The net command looks in the <filename moreinfo="none">smb.conf</filename> file to obtain its own configuration settings. Thus, the following
	command 'knows' which domain to join from the <filename moreinfo="none">smb.conf</filename> file.
	</para>

	<para>
	A Samba server domain trust account can be validated as shown in this example:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>testjoin</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc testjoin
Join to 'MIDEARTH' is OK
</screen>
	Where there is no domain membership account, or when the account credentials are not valid, the following
	results will be observed:
<screen format="linespecific">
net rpc testjoin -S DOLPHIN
Join to domain 'WORLDOCEAN' is not valid
</screen>
	</para>

	<para>
	The equivalent command for joining a Samba server to a Windows ADS domain is shown here:
<indexterm significance="normal"><primary>net</primary><secondary>ads</secondary><tertiary>testjoin</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net ads testjoin
Using short domain name -- TAKEAWAY
Joined 'LEMONADE' to realm 'TAKEAWAY.BIZ'
</screen>
	In the event that the ADS trust was not established, or is broken for one reason or another, the following
	error message may be obtained:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net ads testjoin -UAdministrator%secret
Join to domain is not valid
</screen>
	</para>

	<para>
	The following demonstrates the process of creating a machine trust account in the target domain for the
	Samba server from which the command is executed:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>join</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc join -S FRODO -Uroot%not24get
Joined domain MIDEARTH.
</screen>
	The joining of a Samba server to a Samba domain results in the creation of a machine account. An example
	of this is shown here:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> pdbedit -Lw merlin\$
merlin$:1009:9B4489D6B90461FD6A3EC3AB96147E16:\
176D8C554E99914BDF3407DEA2231D80:[S          ]:LCT-42891919:
</screen>
	The S in the square brackets means this is a server (PDC/BDC) account. The domain join can be cast to join
	purely as a workstation, in which case the S is replaced with a W (indicating a workstation account). The
	following command can be used to affect this:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>join member</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc join member -S FRODO -Uroot%not24get
Joined domain MIDEARTH.
</screen>
	Note that the command-line parameter <constant>member</constant> makes this join specific. By default
	the type is deduced from the <filename moreinfo="none">smb.conf</filename> file configuration. To specifically join as a PDC or BDC, the
	command-line parameter will be <constant>[PDC | BDC]</constant>. For example:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>join bdc</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc join bdc -S FRODO -Uroot%not24get
Joined domain MIDEARTH.
</screen>
	It is best to let Samba figure out the domain join type from the settings in the <filename moreinfo="none">smb.conf</filename> file.
	</para>

	<para>
	The command to join a Samba server to a Windows ADS domain is shown here:
<indexterm significance="normal"><primary>net</primary><secondary>ads</secondary><tertiary>join</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net ads join -UAdministrator%not24get
Using short domain name -- GDANSK
Joined 'FRANDIMITZ' to realm 'GDANSK.ABMAS.BIZ'
</screen>
	</para>

	<para>
	There is no specific option to remove a machine account from an NT4 domain. When a domain member that is a
	Windows machine is withdrawn from the domain, the domain membership account is not automatically removed
	either. Inactive domain member accounts can be removed using any convenient tool. If necessary, the
	machine account can be removed using the following <literal>net</literal> command:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>user delete</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc user delete HERRING\$ -Uroot%not24get
Deleted user account.
</screen>
	The removal is made possible because machine accounts are just like user accounts with a trailing $
	character. The account management operations treat user and machine accounts in like manner.
	</para>

	<para>
	A Samba-3 server that is a Windows ADS domain member can execute the following command to detach from the
	domain:
<indexterm significance="normal"><primary>net</primary><secondary>ads</secondary><tertiary>leave</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net ads leave
</screen>
	</para>

	<para>
	Detailed information regarding an ADS domain can be obtained by a Samba DMS machine by executing the
	following:
<indexterm significance="normal"><primary>net</primary><secondary>ads</secondary><tertiary>status</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net ads status
</screen>
	The volume of information is extensive. Please refer to the book <quote>Samba-3 by Example</quote>,
	Chapter 7 for more information regarding its use. This book may be obtained either in print or online from
	the <ulink url="http://www.samba.org/samba/docs/Samba3-ByExample.pdf">Samba-3 by Example</ulink>.
	</para>

	</sect2>

	<sect2>
	<title>Interdomain Trusts</title>

	<para>
	Interdomain trust relationships form the primary mechanism by which users from one domain can be granted
	access rights and privileges in another domain. 
	</para>

	<para>
	To discover what trust relationships are in effect, execute this command:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>trustdom list</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc trustdom list -Uroot%not24get
Trusted domains list:

none

Trusting domains list:

none
</screen>
	There are no interdomain trusts at this time; the following steps will create them.
	</para>

	<para>
	It is necessary to create a trust account in the local domain. A domain controller in a second domain can
	create a trusted connection with this account. That means that the foreign domain is being trusted
	to access resources in the local domain. This command creates the local trust account:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>trustdom add</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc trustdom add DAMNATION f00db4r -Uroot%not24get
</screen>
	The account can be revealed by using the <literal>pdbedit</literal> as shown here:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> pdbedit -Lw DAMNATION\$
DAMNATION$:1016:9AC1F121DF897688AAD3B435B51404EE: \
7F845808B91BB9F7FEF44B247D9DC9A6:[I         ]:LCT-428934B1:
</screen>
	A trust account will always have an I in the field within the square brackets.
	</para>

	<para>
	If the trusting domain is not capable of being reached, the following command will fail:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>trustdom list</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc trustdom list -Uroot%not24get
Trusted domains list:

none

Trusting domains list:

DAMNATION           S-1-5-21-1385457007-882775198-1210191635
</screen>
	The above command executed successfully; a failure is indicated when the following response is obtained:
<screen format="linespecific">
net rpc trustdom list -Uroot%not24get
Trusted domains list:

DAMNATION           S-1-5-21-1385457007-882775198-1210191635

Trusting domains list:

DAMNATION           domain controller is not responding
</screen>
	</para>

	<para>
	Where a trust account has been created on a foreign domain, Samba is able to establish the trust (connect with)
	the foreign account. In the process it creates a one-way trust to the resources on the remote domain. This
	command achieves the objective of joining the trust relationship:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>trustdom establish</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc trustdom establish DAMNATION
Password: xxxxxxx	== f00db4r
Could not connect to server TRANSGRESSION
Trust to domain DAMNATION established
</screen>
	Validation of the two-way trust now established is possible as shown here:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc trustdom list -Uroot%not24get
Trusted domains list:

DAMNATION           S-1-5-21-1385457007-882775198-1210191635

Trusting domains list:

DAMNATION           S-1-5-21-1385457007-882775198-1210191635
</screen>
	</para>

	<para>
	Sometimes it is necessary to remove the ability for local users to access a foreign domain. The trusting
	connection can be revoked as shown here:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>trustdom revoke</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc trustdom revoke DAMNATION -Uroot%not24get
</screen>
	At other times it becomes necessary to remove the ability for users from a foreign domain to be able to
	access resources in the local domain. The command shown here will do that:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc trustdom del DAMNATION -Uroot%not24get
</screen>

	</para>

	</sect2>

	</sect1>

	<sect1>
	<title>Managing Security Identifiers (SIDS)</title>

	<para>
<indexterm significance="normal"><primary>security identifier</primary></indexterm>
<indexterm significance="normal"><primary>SID</primary></indexterm>
<indexterm significance="normal"><primary>desktop profiles</primary></indexterm>
<indexterm significance="normal"><primary>user encoded</primary></indexterm>
<indexterm significance="normal"><primary>group SID</primary></indexterm>
	The basic security identifier that is used by all Windows networking operations is the Windows security
	identifier (SID). All Windows network machines (servers and workstations), users, and groups are
	identified by their respective SID. All desktop profiles are also encoded with user and group SIDs that
	are specific to the SID of the domain to which the user belongs.
	</para>

	<para>
<indexterm significance="normal"><primary>machine SID</primary></indexterm>
<indexterm significance="normal"><primary>domain SID</primary></indexterm>
<indexterm significance="normal"><primary>SID</primary></indexterm>
<indexterm significance="normal"><primary>rejoin</primary></indexterm>
	It is truly prudent to store the machine and/or domain SID in a file for safekeeping. Why? Because 
	a change in hostname or in the domain (workgroup) name may result in a change in the SID. When you
	have the SID on hand, it is a simple matter to restore it. The alternative is to suffer the pain of
	having to recover user desktop profiles and perhaps rejoin all member machines to the domain.
	</para>

	<para>
	First, do not forget to store the local SID in a file. It is a good idea to put this in the directory
	in which the <filename moreinfo="none">smb.conf</filename> file is also stored. Here is a simple action to achieve this:
<indexterm significance="normal"><primary>net</primary><secondary>getlocalsid</secondary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net getlocalsid &gt; /etc/samba/my-sid
</screen>
	Good, there is now a safe copy of the local machine SID. On a PDC/BDC this is the domain SID also.
	</para>

	<para>
	The following command reveals what the former one should have placed into the file called
	<filename moreinfo="none">my-sid</filename>:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net getlocalsid
SID for domain MERLIN is: S-1-5-21-726309263-4128913605-1168186429
</screen>
	</para>

	<para>
	If ever it becomes necessary to restore the SID that has been stored in the <filename moreinfo="none">my-sid</filename>
	file, simply copy the SID (the string of characters that begins with <constant>S-1-5-21</constant>) to
	the command line shown here:
<indexterm significance="normal"><primary>net</primary><secondary>setlocalsid</secondary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net setlocalsid S-1-5-21-1385457007-882775198-1210191635
</screen>
	Restoration of a machine SID is a simple operation, but the absence of a backup copy can be very
	problematic.
	</para>

	<para>
	The following operation is useful only for machines that are being configured as a PDC or a BDC.
	DMS and workstation clients should have their own machine SID to avoid
	any potential namespace collision. Here is the way that the BDC SID can be synchronized to that
	of the PDC (this is the default NT4 domain practice also):
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>getsid</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc getsid -S FRODO -Uroot%not24get
Storing SID S-1-5-21-726309263-4128913605-1168186429 \
    for Domain MIDEARTH in secrets.tdb
</screen>
	Usually it is not necessary to specify the target server (-S FRODO) or the administrator account
	credentials (-Uroot%not24get).
	</para>

	</sect1>
	
	<sect1>
	<title>Share Management</title>

	<para>
	Share management is central to all file serving operations. Typical share operations include:
	</para>

	<itemizedlist>
		<listitem><para>Creation/change/deletion of shares</para></listitem>
		<listitem><para>Setting/changing ACLs on shares</para></listitem>
		<listitem><para>Moving shares from one server to another</para></listitem>
		<listitem><para>Change of permissions of share contents</para></listitem>
	</itemizedlist>

	<para>
	Each of these are dealt with here insofar as they involve the use of the <literal>net</literal>
	command. Operations outside of this command are covered elsewhere in this document.
	</para>

	<sect2>
	<title>Creating, Editing, and Removing Shares</title>

	<para>
	A share can be added using the <literal>net rpc share</literal> command capabilities.
	The target machine may be local or remote and is specified by the -S option. It must be noted
	that the addition and deletion of shares using this tool depends on the availability of a suitable
	interface script. The interface scripts Sambas <literal>smbd</literal> uses are called
	<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="ADDSHARECOMMAND" xlink:href="smb.conf.5.html#ADDSHARECOMMAND">add share command</link>, <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="DELETESHARECOMMAND" xlink:href="smb.conf.5.html#DELETESHARECOMMAND">delete share command</link> and
	<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="CHANGESHARECOMMAND" xlink:href="smb.conf.5.html#CHANGESHARECOMMAND">change share command</link> A set of example scripts are provided in the Samba source
	code tarball in the directory <filename moreinfo="none">~samba/examples/scripts</filename>.
	</para>

	<para>
	The following steps demonstrate the use of the share management capabilities of the <literal>net</literal>
	utility. In the first step a share called <constant>Bulge</constant> is added. The sharepoint within the
	file system is the directory <filename moreinfo="none">/data</filename>. The command that can be executed to perform the
	addition of this share is shown here:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>share add</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc share add Bulge=/data -S MERLIN -Uroot%not24get
</screen>
	Validation is an important process, and by executing the command <literal>net rpc share</literal>
	with no other operators it is possible to obtain a listing of available shares, as shown here:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc share -S MERLIN -Uroot%not24get
profdata
archive
Bulge   &lt;--- This one was added
print$
netlogon
profiles
IPC$
kyocera
ADMIN$
</screen>
	</para>

	<para>
	Often it is desirable also to permit a share to be removed using a command-line tool.
	The following step permits the share that was previously added to be removed:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>share delete</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc share delete Bulge -S MERLIN -Uroot%not24get
</screen>
	A simple validation shown here demonstrates that the share has been removed:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc share -S MERLIN -Uroot%not24get
profdata
archive
print$
netlogon
profiles
IPC$
ADMIN$
kyocera
</screen>
	</para>

	</sect2>

	<sect2>
	<title>Creating and Changing Share ACLs</title>

	<para>
	At this time the <literal>net</literal> tool cannot be used to manage ACLs on Samba shares. In MS Windows 
	language this is called Share Permissions.
	</para>

	<para>
	It is possible to set ACLs on Samba shares using either the SRVTOOLS NT4 Domain Server Manager
	or using the Computer Management MMC snap-in. Neither is covered here,
	but see <link linkend="AccessControls"/>.
	</para>

	</sect2>

	<sect2>
	<title>Share, Directory, and File Migration</title>

	<para>
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>vampire</tertiary></indexterm>
	Shares and files can be migrated in the same manner as user, machine, and group accounts.
	It is possible to preserve access control settings (ACLs) as well as security settings
	throughout the migration process. The <literal>net rpc vampire</literal> facility is used
	to migrate accounts from a Windows NT4 (or later) domain to a Samba server. This process
	preserves passwords and account security settings and is a precursor to the migration
	of shares and files.
	</para>

	<para>
	The <literal>net rpc share</literal> command may be used to migrate shares, directories,
	files, and all relevant data from a Windows server to a Samba server.
	</para>

	<para>
	A set of command-line switches permit the creation of almost direct clones of Windows file
	servers. For example, when migrating a fileserver, file ACLs and DOS file attributes from
	the Windows server can be included in the migration process and will reappear, almost identically,
	on the Samba server when the migration has been completed.
	</para>

	<para>
	The migration process can be completed only with the Samba server already being fully operational.
	The user and group accounts must be migrated before attempting to migrate data
	share, files, and printers. The migration of files and printer configurations involves the use
	of both SMB and MS DCE RPC services. The benefit of the manner in which the migration process has
	been implemented is that the possibility now exists to use a Samba server as a man-in-middle migration
	service that affects a transfer of data from one server to another. For example, if the Samba
	server is called MESSER, the source Windows NT4 server is called PEPPY, and the target Samba
	server is called GONZALES, the machine MESSER can be used to effect the migration of all data
	(files and shares) from PEPPY to GONZALES. If the target machine is not specified, the local
	server is assumed by default - as net's general rule of thumb .
	</para>

	<para>
	The success of server migration requires a firm understanding of the structure of the source
	server (or domain) as well as  the processes on which the migration is critically dependant.
	</para>

	<para>
	There are two known limitations to the migration process:
	</para>

	<orderedlist inheritnum="ignore" continuation="restarts">
		<listitem><para>
		The <literal>net</literal> command requires that the user credentials provided exist on both
		the migration source and the migration target.
		</para></listitem>

		<listitem><para>
		Printer settings may not be fully or may be incorrectly migrated. This might in particular happen
		when migrating a Windows 2003 print server to Samba.
		</para></listitem>
	</orderedlist>

	<sect3>
	<title>Share Migration</title>

	<para>
	The <literal>net rpc share migrate</literal> command operation permits the migration of plain
	share stanzas. A stanza contains the parameters within which a file or print share are defined.
	The use of this migration method will create share stanzas that have as parameters the file
	system directory path, an optional description, and simple security settings that permit write
	access to files. One of the first steps necessary following migration is to review the share
	stanzas to ensure that the settings are suitable for use.
	</para>

	<para>
	The shares are created on the fly as part of the migration process. The <literal>smbd</literal>
	application does this by calling on the operating system to execute the script specified by the 
	<filename moreinfo="none">smb.conf</filename> parameter <parameter moreinfo="none">add share command</parameter>.
	</para>

	<para>
	There is a suitable example script for the <parameter moreinfo="none">add share command</parameter> in the
	<filename moreinfo="none">$SAMBA_SOURCES/examples/scripts</filename> directory. It should be noted that
	the account that is used to drive the migration must, of necessity, have appropriate file system
	access privileges and have the right to create shares and to set ACLs on them. Such rights are
	conferred by these rights: <parameter moreinfo="none">SeAddUsersPrivilege</parameter> and <parameter moreinfo="none">SeDiskOperatorPrivilege</parameter>.
	For more information regarding rights and privileges please refer to <link linkend="rights"/>.
	</para>

	<para>
	The syntax of the share migration command is shown here:
<screen format="linespecific">
net rpc share MIGRATE SHARES &lt;share-name&gt; -S &lt;source&gt;
        [--destination=localhost] [--exclude=share1,share2] [-v]
</screen>
	When the parameter &lt;share-name&gt; is omitted, all shares will be migrated. The potentially
	large list of available shares on the system that is being migrated can be limited using the
	<parameter moreinfo="none">--exclude</parameter> switch. For example:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>share migrate</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc share migrate shares myshare\
         -S win2k -U administrator%secret"
</screen>
	This will migrate the share <constant>myshare</constant> from the server <constant>win2k</constant>
	to the Samba Server using the permissions that are tied to the account <constant>administrator</constant> 
	with the password <constant>secret</constant>. The account that is used must be the same on both the
	migration source server and the target Samba server. The use of the <literal>net rpc
	vampire</literal>, prior to attempting the migration of shares, will ensure that accounts will be
	identical on both systems. One precaution worth taking before commencement of migration of shares is
	to validate that the migrated accounts (on the Samba server) have the needed rights and privileges.
	This can be done as shown here:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>right list accounts</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc right list accounts -Uroot%not24get
</screen>
	The steps taken so far perform only the migration of shares. Directories and directory contents
	are not migrated by the steps covered up to this point.
	</para>

	</sect3>

	<sect3>
	<title>File and Directory Migration</title>

	<para>
	Everything covered to this point has been done in preparation for the migration of file and directory
	data. For many people preparation is potentially boring and the real excitement only begins when file
	data can be used. The next steps demonstrate the techniques that can be used to transfer (migrate)
	data files using the <literal>net</literal> command.
	</para>

	<para>
	Transfer of files from one server to another has always been a challenge for MS Windows
	administrators because Windows NT and 200X servers do not always include the tools needed. The
	<literal>xcopy</literal> from Windows NT is not capable of preserving file and directory ACLs, 
	it does so only with Windows 200x. Microsoft does provide a
	utility that can copy ACLs (security settings) called <literal>scopy</literal>, but it is provided only
	as part of the Windows NT or 200X Server Resource Kit.
	</para>

	<para>
	There are several tools, both commercial and freeware, that can be used from a Windows server to copy files
	and directories with full preservation of security settings. One of the best known of the free tools is
	called <literal>robocopy</literal>.
	</para>

	<para>
	The <literal>net</literal> utility can be used to copy files and directories with full preservation of
	ACLs as well as DOS file attributes. Note that including ACLs makes sense only where the destination
	system will operate within the same security context as the source system. This applies both to a
	DMS and to domain controllers that result from a vampired domain.
	Before file and directory migration, all shares must already exist.
	</para>

	<para>
	The syntax for the migration commands is shown here:
<screen format="linespecific">
net rpc share MIGRATE FILES &lt;share-name&gt; -S &lt;source&gt;
    [--destination=localhost] [--exclude=share1,share2]
    [--acls] [--attrs] [--timestamps] [-v]
</screen>
	If the &lt;share-name&gt; parameter is omitted, all shares will be migrated. The potentially large
	list of shares on the source system can be restricted using the <parameter moreinfo="none">--exclude</parameter> command
	switch.
	</para>

	<para>
	Where it is necessary to preserve all file ACLs, the <parameter moreinfo="none">--acls</parameter> switch should be added
	to the above command line. Original file timestamps can be preserved by specifying the
	<parameter moreinfo="none">--timestamps</parameter> switch, and the DOS file attributes (i.e., hidden, archive, etc.) can
	be preserved by specifying the <parameter moreinfo="none">--attrs</parameter> switch.
	</para>

	<note><para>
	The ability to preserve ACLs depends on appropriate support for ACLs as well as the general file system
	semantics of the host operating system on the target server. A migration from one Windows file server to
	another will perfectly preserve all file attributes. Because of the difficulty of mapping Windows ACLs
	onto a POSIX ACLs-supporting system, there can be no perfect migration of Windows ACLs to a Samba server.
	</para></note>

	<para>
	The ACLs that result on a Samba server will most probably not match the originating ACLs. Windows supports
	the possibility of files that are owned only by a group. Group-alone file ownership is not possible under
	UNIX/Linux. Errors in migrating group-owned files can be avoided by using the <filename moreinfo="none">smb.conf</filename> file
	<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="FORCEUNKNOWNACLUSER" xlink:href="smb.conf.5.html#FORCEUNKNOWNACLUSER">force unknown acl user = yes</link> parameter. This facility will
	automatically convert group-owned files into correctly user-owned files on the Samba server.
	</para>

	<para>
	An example for migration of files from a machine called <constant>nt4box</constant> to the Samba server
	from which the process will be handled is shown here:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>share migrate files</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc share migrate files -S nt4box --acls \
    --attrs -U administrator%secret
</screen>
	</para>

	<para>
	This command  will migrate all files and directories from all file shares on the Windows server called
	<constant>nt4box</constant> to the Samba server from which migration is initiated. Files that are group-owned
	will be owned by the user account <constant>administrator</constant>.
	</para>

	</sect3>
	
	<sect3>
	<title>Share-ACL Migration</title>
	<para>
	It is possible to have share-ACLs (security descriptors) that won't allow you, even as Administrator, to
	copy any files or directories into it. Therefor the migration of the share-ACLs has been put into a separate
	function:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>share migrate security</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc share migrate security -S nt4box -U administrator%secret
</screen>
	</para>

	<para>
	This command will only copy the share-ACL of each share on nt4box to your local samba-system.
	</para>
	</sect3>

	<sect3>
	<title>Simultaneous Share and File Migration</title>

	<para>
	The operating mode shown here is just a combination of the previous three. It first migrates
	share definitions and then all shared files and directories and finally migrates the share-ACLs:
<screen format="linespecific">
net rpc share MIGRATE ALL &lt;share-name&gt; -S &lt;source&gt;
    [--exclude=share1, share2] [--acls] [--attrs] [--timestamps] [-v]
</screen>
	</para>

	<para>
	An example of simultaneous migration is shown here:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>share migrate all</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc share migrate all -S w2k3server -U administrator%secret
</screen>
	This will generate a complete server clone of the <parameter moreinfo="none">w2k3server</parameter> server.
	</para>

	</sect3>

	</sect2>

	<sect2>
	<title>Printer Migration</title>

	<para>
	The installation of a new server, as with the migration to a new network environment, often is similar to
	building a house; progress is very rapid from the laying of foundations up to the stage at which
	the house can be locked up, but the finishing off appears to take longer and longer as building
	approaches completion.
	</para>

	<para>
	Printing needs vary greatly depending on the network environment and may be very simple or complex. If
	the need is very simple, the best solution to the implementation of printing support may well be to
	re-install everything from a clean slate instead of migrating older configurations. On the other hand,
	a complex network that is integrated with many international offices and a multiplexity of local branch
	offices, each of which form an inter-twined maze of printing possibilities, the ability to migrate all
	printer configurations is decidedly beneficial. To manually re-establish a complex printing network
	will take much time and frustration. Often it will not be possible to find driver files that are
	currently in use, necessitating the installation of newer drivers. Newer drivers often implement
	printing features that will necessitate a change in the printer usage. Additionally, with very complex
	printer configurations it becomes almost impossible to re-create the same environment <?latex --- ?> no matter
	how extensively it has been documented.
	</para>

	<para>
	The migration of an existing printing architecture involves the following:
	</para>

	<itemizedlist>
		<listitem><para>Establishment of print queues.</para></listitem>

		<listitem><para>Installation of printer drivers (both for the print server and for Windows clients.</para></listitem>

		<listitem><para>Configuration of printing forms.</para></listitem>

		<listitem><para>Implementation of security settings.</para></listitem>

		<listitem><para>Configuration of printer settings.</para></listitem>
	</itemizedlist>

	<para>
	The Samba <literal>net</literal> utility permits printer migration from one Windows print server
	to another. When this tool is used to migrate printers to a Samba server <literal>smbd</literal>,
	the application that receives the network requests to create the necessary services must call out
	to the operating system in order to create the underlying printers. The call-out is implemented
	by way of an interface script that can be specified by the <filename moreinfo="none">smb.conf</filename> file parameter
	<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="" xlink:href="smb.conf.5.html#"/>. This script is essential to the migration process.
	A suitable example script may be obtained from the <filename moreinfo="none">$SAMBA_SOURCES/examples/scripts</filename>
	directory. Take note that this script must be customized to suit the operating system environment
	and may use its tools to create a print queue.
	</para>

	<para>
	Each of the components listed above can be completed separately, or they can be completed as part of an
	automated operation. Many network administrators prefer to deal with migration issues in a manner that
	gives them the most control, particularly when things go wrong. The syntax for each operation is now
	briefly described.
	</para>

	<para>
	Printer migration from a Windows print server (NT4 or 200x) is shown. This instruction causes the
	printer share to be created together with the underlying print queue:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>printer migrate printers</tertiary></indexterm>
<screen format="linespecific">
net rpc printer MIGRATE PRINTERS [printer] [misc. options] [targets]
</screen>
	Printer drivers can be migrated from the Windows print server to the Samba server using this
	command-line instruction:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>printer migrate drivers</tertiary></indexterm>
<screen format="linespecific">
net rpc printer MIGRATE DRIVERS [printer] [misc. options] [targets]
</screen>
	Printer forms can be migrated with the following operation:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>printer migrate forms</tertiary></indexterm>
<screen format="linespecific">
net rpc printer MIGRATE FORMS [printer] [misc. options] [targets]
</screen>
	Printer security settings (ACLs) can be migrated from the Windows server to the Samba server using this command:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>printer migrate security</tertiary></indexterm>
<screen format="linespecific">
net rpc printer MIGRATE SECURITY [printer] [misc. options] [targets]
</screen>
	Printer configuration settings include factors such as paper size and default paper orientation.
	These can be migrated from the Windows print server to the Samba server with this command:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>printer migrate settings</tertiary></indexterm>
<screen format="linespecific">
net rpc printer MIGRATE SETTINGS [printer] [misc. options] [targets]
</screen>
	</para>

	<para>
	Migration of printers including the above-mentioned sets of information may be completed
	with a single command using this syntax:
<screen format="linespecific">
net rpc printer MIGRATE ALL [printer] [misc. options] [targets]
</screen>
	</para>

	</sect2>

	</sect1>

	<sect1>
	<title>Controlling Open Files</title>

	<para>
	The man page documents the <literal>net file</literal> function suite, which provides the tools to
	close open files using either RAP or RPC function calls. Please refer to the man page for specific
	usage information.
	</para>

	</sect1>

	<sect1>
	<title>Session and Connection Management</title>

	<para>
	The session management interface of the <literal>net session</literal> command uses the old RAP
	method to obtain the list of connections to the Samba server, as shown here:
<indexterm significance="normal"><primary>net</primary><secondary>rap</secondary><tertiary>session</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rap session -S MERLIN -Uroot%not24get
Computer             User name            Client Type        Opens Idle time
------------------------------------------------------------------------------
\\merlin             root                 Unknown Client         0 00:00:00
\\marvel             jht                  Unknown Client         0 00:00:00
\\maggot             jht                  Unknown Client         0 00:00:00
\\marvel             jht                  Unknown Client         0 00:00:00
</screen>
	</para>

	<para>
	A session can be closed by executing a command as shown here:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rap session close marvel -Uroot%not24get
</screen>
	</para>

	</sect1>

	<sect1>
	<title>Printers and ADS</title>

	<para>
	When Samba-3 is used within an MS Windows ADS environment, printers shared via Samba will not be browseable
	until they have been published to the ADS domain. Information regarding published printers may be obtained
	from the ADS server by executing the <literal>net ads print info</literal> command following this syntax:
<indexterm significance="normal"><primary>net</primary><secondary>ads</secondary><tertiary>printer info</tertiary></indexterm>
<screen format="linespecific">
net ads printer info &lt;printer_name&gt; &lt;server_name&gt; -Uadministrator%secret
</screen>
	If the asterisk (*) is used in place of the printer_name argument, a list of all printers will be
	returned.
	</para>

	<para>
	To publish (make available) a printer to ADS, execute the following command:
<indexterm significance="normal"><primary>net</primary><secondary>ads</secondary><tertiary>printer publish</tertiary></indexterm>
<screen format="linespecific">
net ads printer publish &lt;printer_name&gt; -Uadministrator%secret
</screen>
	This publishes a printer from the local Samba server to ADS.
	</para>

	<para>
	Removal of a Samba printer from ADS is achieved by executing this command:
<indexterm significance="normal"><primary>net</primary><secondary>ads</secondary><tertiary>printer remove</tertiary></indexterm>
<screen format="linespecific">
net ads printer remove &lt;printer_name&gt; -Uadministrator%secret
</screen>
	</para>

	<para>
	A generic search (query) can also be made to locate a printer across the entire ADS domain by executing:
<indexterm significance="normal"><primary>net</primary><secondary>ads</secondary><tertiary>printer search</tertiary></indexterm>
<screen format="linespecific">
net ads printer search &lt;printer_name&gt; -Uadministrator%secret
</screen>
	</para>

	</sect1>

	<sect1>
	<title>Manipulating the Samba Cache</title>

	<para>
	Please refer to the <literal>net</literal> command man page for information regarding cache management.
	</para>

	</sect1>

	<sect1>
	<title>Managing IDMAP UID/SID Mappings</title>

	<para>
	The IDMAP UID to SID, and SID to UID, mappings that are created by <literal>winbindd</literal> can be
	backed up to a text file. The text file can be manually edited, although it is highly recommended that
	you attempt this only if you know precisely what you are doing.
	</para>

	<para>
	An IDMAP text dump file can be restored (or reloaded). There are two situations that may necessitate
	this action: a) The existing IDMAP file is corrupt, b) It is necessary to install an editted version
	of the mapping information.
	</para>

	<para>
	Winbind must be shut down to dump the IDMAP file. Before restoring a dump file, shut down
	<literal>winbindd</literal> and delete the old <filename moreinfo="none">winbindd_idmap.tdb</filename> file.
	</para>

	<sect2>
	<title>Creating an IDMAP Database Dump File</title>

	<para>
	The IDMAP database can be dumped to a text file as shown here:
<screen format="linespecific">
net idmap dump &lt;full_path_and_tdb_filename&gt; &gt; dumpfile.txt
</screen>
	Where a particular build of Samba the run-time tdb files are stored in the
	<filename moreinfo="none">/var/lib/samba</filename> directory the following commands to create the dump file will suffice:
<screen format="linespecific">
net idmap dump /var/lib/samba/winbindd_idmap.tdb &gt; idmap_dump.txt
</screen>
	</para>

	</sect2>

	<sect2>
	<title>Restoring the IDMAP Database Dump File</title>

	<para>
	The IDMAP dump file can be restored using the following command:
<screen format="linespecific">
net idmap restore &lt;full_path_and_tdb_filename&gt; &lt; dumpfile.txt
</screen>
	Where the Samba run-time tdb files are stored in the <filename moreinfo="none">/var/lib/samba</filename> directory
    the following command can be used to restore the data to the tdb file:
<screen format="linespecific">
net idmap restore /var/lib/samba/winbindd_idmap.tdb &lt; idmap_dump.txt
</screen>
	</para>

	</sect2>

	</sect1>

	<sect1 id="netmisc1">
	<title>Other Miscellaneous Operations</title>

	<para>
	The following command is useful for obtaining basic statistics regarding a Samba domain. This command does
	not work with current Windows XP Professional clients.
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>info</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc info
Domain Name: RAPIDFLY
Domain SID: S-1-5-21-399034208-633907489-3292421255
Sequence number: 1116312355
Num users: 720
Num domain groups: 27
Num local groups: 6
</screen>
	</para>

	<para>
	Another useful tool is the <literal>net time</literal> tool set. This tool may be used to query the
	current time on the target server as shown here:
<indexterm significance="normal"><primary>net</primary><secondary>time</secondary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net time -S SAURON
Tue May 17 00:50:43 2005
</screen>
	In the event that it is the intent to pass the time information obtained to the UNIX
	<literal>/bin/time</literal>, it is a good idea to obtain the time from the target server in a format
	that is ready to be passed through. This may be done by executing:
<indexterm significance="normal"><primary>net</primary><secondary>time</secondary><tertiary>system</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net time system -S FRODO
051700532005.16
</screen>
	The time can be set on a target server by executing:
<indexterm significance="normal"><primary>net</primary><secondary>time</secondary><tertiary>set</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net time set -S MAGGOT -U Administrator%not24get
Tue May 17 00:55:30 MDT 2005
</screen>
	It is possible to obtain the time zone of a server by executing the following command against it:
<indexterm significance="normal"><primary>net</primary><secondary>time</secondary><tertiary>zone</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net time zone -S SAURON
-0600
</screen>
	</para>

	</sect1>

</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="idmapper">
<chapterinfo>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
</chapterinfo>

<title>Identity Mapping (IDMAP)</title>

<para>
<indexterm significance="normal"><primary>Windows</primary></indexterm>
<indexterm significance="normal"><primary>interoperability</primary></indexterm>
<indexterm significance="normal"><primary>IDMAP</primary></indexterm>
<indexterm significance="normal"><primary>Windows Security Identifiers</primary><see>SID</see></indexterm>
<indexterm significance="normal"><primary>SID</primary></indexterm>
<indexterm significance="normal"><primary>UID</primary></indexterm>
<indexterm significance="normal"><primary>GID</primary></indexterm>
The Microsoft Windows operating system has a number of features that impose specific challenges
to interoperability with the operating systems on which Samba is implemented. This chapter deals
explicitly with the mechanisms Samba-3 (version 3.0.8 and later) uses to overcome one of the
key challenges in the integration of Samba servers into an MS Windows networking environment.
This chapter deals with identity mapping (IDMAP) of Windows security identifiers (SIDs)
to UNIX UIDs and GIDs.
</para>

<para>
To ensure sufficient coverage, each possible Samba deployment type is discussed.
This is followed by an overview of how the IDMAP facility may be implemented.
</para>

<para>
<indexterm significance="normal"><primary>network client</primary></indexterm>
<indexterm significance="normal"><primary>IDMAP</primary></indexterm>
<indexterm significance="normal"><primary>IDMAP infrastructure</primary></indexterm>
<indexterm significance="normal"><primary>default behavior</primary></indexterm>
The IDMAP facility is of concern where more than one Samba server (or Samba network client)
is installed in a domain. Where there is a single Samba server, do not be too concerned regarding
the IDMAP infrastructure <?latex --- ?> the default behavior of Samba is nearly always sufficient.
Where mulitple Samba servers are used it is often necessary to move data off one server and onto
another, and that is where the fun begins!
</para>

<para>
<indexterm significance="normal"><primary>UID</primary></indexterm>
<indexterm significance="normal"><primary>GID</primary></indexterm>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
<indexterm significance="normal"><primary>NSS</primary></indexterm>
<indexterm significance="normal"><primary>nss_ldap</primary></indexterm>
<indexterm significance="normal"><primary>NT4 domain members</primary></indexterm>
<indexterm significance="normal"><primary>ADS domain members</primary></indexterm>
<indexterm significance="normal"><primary>security name-space</primary></indexterm>
Where user and group account information is stored in an LDAP directory every server can have the same
consistent UID and GID for users and groups. This is achieved using NSS and the nss_ldap tool. Samba
can be configured to use only local accounts, in which case the scope of the IDMAP problem is somewhat
reduced. This works reasonably well if the servers belong to a single domain, and interdomain trusts
are not needed. On the other hand, if the Samba servers are NT4 domain members, or ADS  domain members,
or if there is a need to keep the security name-space separate (i.e., the user
<literal moreinfo="none">DOMINICUS\FJones</literal> must not be given access to the account resources of the user 
<literal moreinfo="none">FRANCISCUS\FJones</literal><footnote>Samba local account mode results in both
<literal moreinfo="none">DOMINICUS\FJones</literal> and <literal moreinfo="none">FRANCISCUS\FJones</literal> mapping to the UNIX user
<literal moreinfo="none">FJones</literal>.</footnote> free from inadvertent cross-over, close attention should be given
to the way that the IDMAP facility is configured.
</para>

<para>
<indexterm significance="normal"><primary>IDMAP</primary></indexterm>
<indexterm significance="normal"><primary>domain access</primary></indexterm>
<indexterm significance="normal"><primary>SID</primary></indexterm>
<indexterm significance="normal"><primary>UID</primary></indexterm>
<indexterm significance="normal"><primary>GID</primary></indexterm>
<indexterm significance="normal"><primary>one domain</primary></indexterm>
The use of IDMAP is important where the Samba server will be accessed by workstations or servers from
more than one domain, in which case it is important to run winbind so it can handle the resolution (ID mapping)
of foreign SIDs to local UNIX UIDs and GIDs.
</para>

<para>
<indexterm significance="normal"><primary>winbindd</primary></indexterm>
The use of the IDMAP facility requires the execution of the <literal>winbindd</literal> upon Samba startup.
</para>

<sect1>
<title>Samba Server Deployment Types and IDMAP</title>

<para>
<indexterm significance="normal"><primary>Server Types</primary></indexterm>
There are four basic server deployment types, as documented in <link linkend="ServerType">the chapter
on Server Types and Security Modes</link>.
</para>

	<sect2>
	<title>Standalone Samba Server</title>

	<para>
	<indexterm significance="normal"><primary>stand-alone server</primary></indexterm>
	<indexterm significance="normal"><primary>Active Directory</primary></indexterm>
	<indexterm significance="normal"><primary>NT4 Domain</primary></indexterm>
	A standalone Samba server is an implementation that is not a member of a Windows NT4 domain,
	a Windows 200X Active Directory domain, or a Samba domain.
	</para>

	<para>
	<indexterm significance="normal"><primary>IDMAP</primary></indexterm>
	<indexterm significance="normal"><primary>identity</primary></indexterm>
	<indexterm significance="normal"><primary>local user</primary></indexterm>
	By definition, this means that users and groups will be created and controlled locally, and
	the identity of a network user must match a local UNIX/Linux user login. The IDMAP facility
	is therefore of little to no interest, winbind will not be necessary, and the IDMAP facility
	will not be relevant or of interest.
	</para>

	</sect2>

	<sect2>
	<title>Domain Member Server or Domain Member Client</title>

	<para>
	<indexterm significance="normal"><primary>PDC</primary></indexterm>
	<indexterm significance="normal"><primary>BDC</primary></indexterm>
	<indexterm significance="normal"><primary>NT4</primary></indexterm>
	<indexterm significance="normal"><primary>SID</primary></indexterm>
	<indexterm significance="normal"><primary>Active Directory</primary></indexterm>
	Samba-3 can act as a Windows NT4 PDC or BDC, thereby providing domain control protocols that
	are compatible with Windows NT4. Samba-3 file and print sharing protocols are compatible with
	all versions of MS Windows products. Windows NT4, as with MS Active Directory,
	extensively makes use of Windows SIDs.
	</para>

	<para>
	<indexterm significance="normal"><primary>MS Windows SID</primary></indexterm>
	<indexterm significance="normal"><primary>UID</primary></indexterm>
	<indexterm significance="normal"><primary>GID</primary></indexterm>
	Samba-3 domain member servers and clients must interact correctly with MS Windows SIDs. Incoming
	Windows SIDs must be translated to local UNIX UIDs and GIDs. Outgoing information from the Samba
	server must provide to MS Windows clients and servers appropriate SIDs.
	</para>

	<para>
	<indexterm significance="normal"><primary>ADS</primary></indexterm>
	<indexterm significance="normal"><primary>winbind</primary></indexterm>
	A Samba member of a Windows networking domain (NT4-style or ADS) can be configured to handle 
	identity mapping in a variety of ways. The mechanism it uses depends on whether or not
	the <literal>winbindd</literal> daemon is used and how the winbind functionality is configured.
	The configuration options are briefly described here:
	</para>

	<variablelist>
		<varlistentry><term>Winbind is not used; users and groups are local: </term>
			<listitem>
				<para>
				<indexterm significance="normal"><primary>winbindd</primary></indexterm>
				<indexterm significance="normal"><primary>smbd</primary></indexterm>
				<indexterm significance="normal"><primary>network traffic</primary></indexterm>
				<indexterm significance="normal"><primary>LoginID</primary></indexterm>
				<indexterm significance="normal"><primary>account name</primary></indexterm>
				<indexterm significance="normal"><primary>getpwnam</primary></indexterm>
				<indexterm significance="normal"><primary>NSS</primary></indexterm>
				<indexterm significance="normal"><primary>local users</primary></indexterm>
				<indexterm significance="normal"><primary>local groups</primary></indexterm>
				<indexterm significance="normal"><primary>/etc/passwd</primary></indexterm>
				<indexterm significance="normal"><primary>/etc/group</primary></indexterm>
				Where <literal>winbindd</literal> is not used Samba (<literal>smbd</literal>)
				uses the underlying UNIX/Linux mechanisms to resolve the identity of incoming
				network traffic. This is done using the LoginID (account name) in the
				session setup request and passing it to the getpwnam() system function call.
				This call is implemented using the name service switch (NSS) mechanism on
				modern UNIX/Linux systems. By saying "users and groups are local,"
				we are implying that they are stored only on the local system, in the
				<filename moreinfo="none">/etc/passwd</filename> and <filename moreinfo="none">/etc/group</filename> respectively.
				</para>

				<para>
				<indexterm significance="normal"><primary>SessionSetupAndX</primary></indexterm>
				<indexterm significance="normal"><primary>/etc/passwd</primary></indexterm>
				For example, when the user <literal moreinfo="none">BERYLIUM\WambatW</literal> tries to open a
				connection to a Samba server the incoming SessionSetupAndX request will make a 
				system call to look up the user <literal moreinfo="none">WambatW</literal> in the
				<filename moreinfo="none">/etc/passwd</filename> file.
				</para>

				<para>
				<indexterm significance="normal"><primary>standalone</primary></indexterm>
				<indexterm significance="normal"><primary>domain member server</primary></indexterm>
				<indexterm significance="normal"><primary>NT4</primary></indexterm>
				<indexterm significance="normal"><primary>ADS</primary></indexterm>
				<indexterm significance="normal"><primary>PDC</primary></indexterm>
				<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
				<indexterm significance="normal"><primary>tdbsam</primary></indexterm>
				<indexterm significance="normal"><primary>passdb backend</primary></indexterm>
				This configuration may be used with standalone Samba servers, domain member
				servers (NT4 or ADS), and for a PDC that uses either an smbpasswd
				or a tdbsam-based Samba passdb backend.
				</para>
			</listitem>
		</varlistentry>
	
		<varlistentry><term>Winbind is not used; users and groups resolved via NSS: </term>
			<listitem>
				<para>
				<indexterm significance="normal"><primary>user accounts</primary></indexterm>
				<indexterm significance="normal"><primary>group accounts</primary></indexterm>
				<indexterm significance="normal"><primary>local accounts</primary></indexterm>
				<indexterm significance="normal"><primary>repository</primary></indexterm>
				<indexterm significance="normal"><primary>NIS</primary></indexterm>
				<indexterm significance="normal"><primary>LDAP</primary></indexterm>
				In this situation user and group accounts are treated as if they are local
				accounts. The only way in which this differs from having local accounts is
				that the accounts are stored in a repository that can be shared. In practice
				this means that they will reside in either an NIS-type database or else in LDAP.
				</para>

				<para>
				<indexterm significance="normal"><primary>standalone</primary></indexterm>
				<indexterm significance="normal"><primary>domain member server</primary></indexterm>
				<indexterm significance="normal"><primary>NT4</primary></indexterm>
				<indexterm significance="normal"><primary>ADS</primary></indexterm>
				<indexterm significance="normal"><primary>PDC</primary></indexterm>
				<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
				<indexterm significance="normal"><primary>tdbsam</primary></indexterm>
				This configuration may be used with standalone Samba servers, domain member
				servers (NT4 or ADS), and for a PDC that uses either an smbpasswd
				or a tdbsam-based Samba passdb backend.
				</para>
			</listitem>
		</varlistentry>

		<varlistentry><term>Winbind/NSS with the default local IDMAP table: </term>
			<listitem>
				<para>
				<indexterm significance="normal"><primary>NT4 domain</primary></indexterm>
				<indexterm significance="normal"><primary>ADS domain</primary></indexterm>
				<indexterm significance="normal"><primary>winbind</primary></indexterm>
				<indexterm significance="normal"><primary>domain control</primary></indexterm>
				There are many sites that require only a simple Samba server or a single Samba
				server that is a member of a Windows NT4 domain or an ADS domain. A typical example
				is an appliance like file server on which no local accounts are configured and
				winbind is used to obtain account credentials from the domain controllers for the
				domain. The domain control can be provided by Samba-3, MS Windows NT4, or MS Windows
				Active Directory.
				</para>

				<para>
				<indexterm significance="normal"><primary>UID numbers</primary></indexterm>
				<indexterm significance="normal"><primary>GID numbers</primary></indexterm>
				<indexterm significance="normal"><primary>/etc/nsswitch.conf</primary></indexterm>
				<indexterm significance="normal"><primary>winbind</primary></indexterm>
				<indexterm significance="normal"><primary>SID</primary></indexterm>
				Winbind is a great convenience in this situation. All that is needed is a range of
				UID numbers and GID numbers that can be defined in the <filename moreinfo="none">smb.conf</filename> file. The
				<filename moreinfo="none">/etc/nsswitch.conf</filename> file is configured to use <literal>winbind</literal>,
				which does all the difficult work of mapping incoming SIDs to appropriate UIDs and GIDs.
				The SIDs are allocated a UID/GID in the order in which winbind receives them.
				</para>

				<para>
				<indexterm significance="normal"><primary>UID</primary></indexterm>
				<indexterm significance="normal"><primary>GID</primary></indexterm>
				<indexterm significance="normal"><primary>IDMAP</primary></indexterm>
				<indexterm significance="normal"><primary>corrupted file</primary></indexterm>
				This configuration is not convenient or practical in sites that have more than one
				Samba server and that require the same UID or GID for the same user or group across
				all servers. One of the hazards of this method is that in the event that the winbind
				IDMAP file becomes corrupted or lost, the repaired or rebuilt IDMAP file may allocate
				UIDs and GIDs to different users and groups from what was there previously with the
				result that MS Windows files that are stored on the Samba server may now not belong to
				the rightful owners.
				</para>
			</listitem>
		</varlistentry>

		<varlistentry><term>Winbind/NSS uses RID based IDMAP: </term>
			<listitem>
				<para>
				<indexterm significance="normal"><primary>RID</primary></indexterm>
				<indexterm significance="normal"><primary>idmap_rid</primary></indexterm>
				<indexterm significance="normal"><primary>ADS</primary></indexterm>
				<indexterm significance="normal"><primary>LDAP</primary></indexterm>
				The IDMAP_RID facility is new to Samba version 3.0.8. It was added to make life easier
				for a number of sites that are committed to use of MS ADS, that do not apply
				an ADS schema extension, and that do not have an installed an LDAP directory server just for
				the purpose of maintaining an IDMAP table. If you have a single ADS domain (not a forest of
				domains, and not multiple domain trees) and you want a simple cookie-cutter solution to the
				IDMAP table problem, then IDMAP_RID is an obvious choice.
				</para>

				<para>
				<indexterm significance="normal"><primary>idmap_rid</primary></indexterm>
				<indexterm significance="normal"><primary>idmap uid</primary></indexterm>
				<indexterm significance="normal"><primary>idmap gid</primary></indexterm>
				<indexterm significance="normal"><primary>RID</primary></indexterm>
				<indexterm significance="normal"><primary>SID</primary></indexterm>
				<indexterm significance="normal"><primary>UID</primary></indexterm>
				<indexterm significance="normal"><primary>idmap backend</primary></indexterm>
				<indexterm significance="normal"><primary>automatic mapping</primary></indexterm>
				This facility requires the allocation of the <parameter moreinfo="none">idmap uid</parameter> and the
				<parameter moreinfo="none">idmap gid</parameter> ranges, and within the <parameter moreinfo="none">idmap uid</parameter>
				it is possible to allocate a subset of this range for automatic mapping of the relative
				identifier (RID) portion of the SID directly to the base of the UID plus the RID value.
				For example, if the <parameter moreinfo="none">idmap uid</parameter> range is <constant>1000-100000000</constant>
				and the <parameter moreinfo="none">idmap backend = idmap_rid:DOMAIN_NAME=1000-50000000</parameter>, and
				a SID is encountered that has the value <constant>S-1-5-21-34567898-12529001-32973135-1234</constant>,
				the resulting UID will be <constant>1000 + 1234 = 2234</constant>.
				</para>
			</listitem>
		</varlistentry>

		<varlistentry><term>Winbind with an NSS/LDAP backend-based IDMAP facility: </term>
			<listitem>
				<para>
				<indexterm significance="normal"><primary>Domain Member</primary></indexterm>
				<indexterm significance="normal"><primary>winbind</primary></indexterm>
				<indexterm significance="normal"><primary>SID</primary></indexterm>
				<indexterm significance="normal"><primary>UID</primary></indexterm>
				<indexterm significance="normal"><primary>GID</primary></indexterm>
				<indexterm significance="normal"><primary>idmap gid</primary></indexterm>
				<indexterm significance="normal"><primary>idmap uid</primary></indexterm>
				<indexterm significance="normal"><primary>LDAP</primary></indexterm>
				In this configuration <literal>winbind</literal> resolved SIDs to UIDs and GIDs from
				the <parameter moreinfo="none">idmap uid</parameter> and <parameter moreinfo="none">idmap gid</parameter> ranges specified
				in the <filename moreinfo="none">smb.conf</filename> file, but instead of using a local winbind IDMAP table, it is stored
				in an LDAP directory so that all domain member machines (clients and servers) can share
				a common IDMAP table.
				</para>

				<para>
				<indexterm significance="normal"><primary>idmap backend</primary></indexterm>
				<indexterm significance="normal"><primary>LDAP server</primary></indexterm>
				<indexterm significance="normal"><primary>LDAP redirects</primary></indexterm>
				It is important that all LDAP IDMAP clients use only the master LDAP server because the
				<parameter moreinfo="none">idmap backend</parameter> facility in the <filename moreinfo="none">smb.conf</filename> file does not correctly
				handle LDAP redirects.
				</para>
			</listitem>
		</varlistentry>

		<varlistentry><term>Winbind with NSS to resolve UNIX/Linux user and group IDs: </term>
			<listitem>
				<para>
				The use of LDAP as the passdb backend is a smart solution for PDC, BDC, and
				domain member servers. It is a neat method for assuring that UIDs, GIDs, and the matching
				SIDs are consistent across all servers.
				</para>

				<para>
				<indexterm significance="normal"><primary>LDAP</primary></indexterm>
				<indexterm significance="normal"><primary>PADL</primary></indexterm>
				The use of the LDAP-based passdb backend requires use of the PADL nss_ldap utility or
				an equivalent. In this situation winbind is used to handle foreign SIDs, that is, SIDs from
				standalone Windows clients (i.e., not a member of our domain) as well as SIDs from 
				another domain. The foreign UID/GID is mapped from allocated ranges (idmap uid and idmap gid)
				in precisely the same manner as when using winbind with a local IDMAP table.
				</para>

				<para>
				<indexterm significance="normal"><primary>nss_ldap</primary></indexterm>
				<indexterm significance="normal"><primary>AD4UNIX</primary></indexterm>
				<indexterm significance="normal"><primary>MMC</primary></indexterm>
				The nss_ldap tool set can be used to access UIDs and GIDs via LDAP as well as via Active
				Directory. In order to use Active Directory, it is necessary to modify the ADS schema by
				installing either the AD4UNIX schema extension or using the Microsoft Services for UNIX
				version 3.5 or later to extend the ADS schema so it maintains UNIX account credentials.
				Where the ADS schema is extended, a Microsoft Management Console (MMC) snap-in is also
				installed to permit the UNIX credentials to be set and managed from the ADS User and Computer
				Management tool. Each account must be separately UNIX-enabled before the UID and GID data can
				be used by Samba.
				</para>
			</listitem>
		</varlistentry>

	</variablelist>

	</sect2>

	<sect2>
	<title>Primary Domain Controller</title>

	<para>
	<indexterm significance="normal"><primary>domain security</primary></indexterm>
	<indexterm significance="normal"><primary>SID</primary></indexterm>
	<indexterm significance="normal"><primary>RID</primary></indexterm>
	<indexterm significance="normal"><primary>algorithmic mapping</primary></indexterm>
	Microsoft Windows domain security systems generate the user and group SID as part
	of the process of creation of an account. Windows does not have a concept of the UNIX UID or a GID; rather,
	it has its own type of security descriptor. When Samba is used as a domain controller, it provides a method
	of producing a unique SID for each user and group. Samba generates a machine and a domain SID to which it
	adds an RID that is calculated algorithmically from a base value that can be specified
	in the <filename moreinfo="none">smb.conf</filename> file, plus twice (2x) the UID or GID. This method is called <quote>algorithmic mapping</quote>.
	</para>

	<para>
	<indexterm significance="normal"><primary>RID base</primary></indexterm>
	For example, if a user has a UID of 4321, and the algorithmic RID base has a value of 1000, the RID will
	be <literal moreinfo="none">1000 + (2 x 4321) = 9642</literal>. Thus, if the domain SID is
	<literal moreinfo="none">S-1-5-21-89238497-92787123-12341112</literal>, the resulting SID is
	<literal moreinfo="none">S-1-5-21-89238497-92787123-12341112-9642</literal>.
	</para>

	<para>
	<indexterm significance="normal"><primary>on-the-fly</primary></indexterm>
	<indexterm significance="normal"><primary>SID</primary></indexterm>
	<indexterm significance="normal"><primary>passdb backend</primary></indexterm>
	<indexterm significance="normal"><primary>ldapsam</primary></indexterm>
	The foregoing type of SID is produced by Samba as an automatic function and is either produced on the fly
	(as is the case when using a <parameter moreinfo="none">passdb backend = [tdbsam | smbpasswd]</parameter>), or may be stored
	as a permanent part of an account in an LDAP-based ldapsam.
	</para>

	<para>
	<indexterm significance="normal"><primary>SFU 3.5</primary></indexterm>
	<indexterm significance="normal"><primary>ADS</primary></indexterm>
	<indexterm significance="normal"><primary>directory schema</primary></indexterm>
	<indexterm significance="normal"><primary>account attributes</primary></indexterm>
	<indexterm significance="normal"><primary>UID</primary></indexterm>
	<indexterm significance="normal"><primary>GID</primary></indexterm>
	<indexterm significance="normal"><primary>ADS schema</primary></indexterm>
	<indexterm significance="normal"><primary>account management</primary></indexterm>
	<indexterm significance="normal"><primary>MMC</primary></indexterm>
	ADS uses a directory schema that can be extended to accommodate additional
	account attributes such as UIDs and GIDs. The installation of Microsoft Service for UNIX 3.5 will expand
	the normal ADS schema to include UNIX account attributes. These must of course be managed separately
	through a snap-in module to the normal ADS account management MMC interface.
	</para>

	<para>
	<indexterm significance="normal"><primary>PDC</primary></indexterm>
	<indexterm significance="normal"><primary>passdb backend</primary></indexterm>
	<indexterm significance="normal"><primary>BDC</primary></indexterm>
	<indexterm significance="normal"><primary>LDAP backend</primary></indexterm>
	Security identifiers used within a domain must be managed to avoid conflict and to preserve itegrity.
	In an NT4 domain context, the PDC manages the distribution of all security credentials to the backup
	domain controllers (BDCs). At this time the only passdb backend for a Samba domain controller that is suitable
	for such information is an LDAP backend.
	</para>

	</sect2>

	<sect2>
	<title>Backup Domain Controller</title>

	<para>
	<indexterm significance="normal"><primary>BDC</primary></indexterm>
	<indexterm significance="normal"><primary>read-only access</primary></indexterm>
	<indexterm significance="normal"><primary>security credentials</primary></indexterm>
	<indexterm significance="normal"><primary>LDAP</primary></indexterm>
	<indexterm significance="normal"><primary>group account</primary></indexterm>
	<indexterm significance="normal"><primary>write changes</primary></indexterm>
	<indexterm significance="normal"><primary>directory</primary></indexterm>
	BDCs have read-only access to security credentials that are stored in LDAP.
	Changes in user or group account information are passed by the BDC to the PDC. Only the PDC can write
	changes to the directory.
	</para>

	<para>
	IDMAP information can be written directly to the LDAP server so long as all domain controllers
	have access to the master (writable) LDAP server. Samba-3 at this time does not handle LDAP redirects
	in the IDMAP backend. This means that it is is unsafe to use a slave (replicate) LDAP server with
	the IDMAP facility.
	</para>

	</sect2>

</sect1>

<sect1>
<title>Examples of IDMAP Backend Usage</title>

<para>
<indexterm significance="normal"><primary>Domain Member Server</primary><see>DMS</see></indexterm>
<indexterm significance="normal"><primary>Domain Member Client</primary><see>DMC</see></indexterm>
<indexterm significance="normal"><primary>DMS</primary></indexterm>
<indexterm significance="normal"><primary>DMC</primary></indexterm>
<indexterm significance="normal"><primary>winbind</primary></indexterm>
Anyone who wishes to use <literal>winbind</literal> will find the following example configurations helpful.
Remember that in the majority of cases <literal>winbind</literal> is of primary interest for use with
domain member servers (DMSs) and domain member clients (DMCs).
</para>

	<sect2>
	<title>Default Winbind TDB</title>

	<para>
	Two common configurations are used:
	</para>

	<itemizedlist>
		<listitem><para>
		Networks that have an NT4 PDC (with or without BDCs) or a Samba PDC (with or without BDCs).
		</para></listitem>

		<listitem><para>
		Networks that use MS Windows 200x ADS.
		</para></listitem>
	</itemizedlist>

	<sect3>
	<title>NT4-Style Domains (Includes Samba Domains)</title>

	<para>
	<link linkend="idmapnt4dms">NT4 Domain Member Server smb.con</link> is a simple example of an NT4 DMS
	<filename moreinfo="none">smb.conf</filename> file that shows only the global section.
	</para>

<example id="idmapnt4dms">
<title>NT4 Domain Member Server smb.conf</title>
<simplelist>
<member># Global parameters</member>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>workgroup</indexterm><parameter>workgroup = MEGANET2</parameter></member>
<member><indexterm>security</indexterm><parameter>security = DOMAIN</parameter></member>
<member><indexterm>idmap uid</indexterm><parameter>idmap uid = 10000-20000</parameter></member>
<member><indexterm>idmap gid</indexterm><parameter>idmap gid = 10000-20000</parameter></member>
<member><indexterm>template primary group</indexterm><parameter>template primary group = "Domain Users"</parameter></member>
<member><indexterm>template shell</indexterm><parameter>template shell = /bin/bash</parameter></member>
</simplelist>
</example>

	<para>
	<indexterm significance="normal"><primary>winbind</primary></indexterm>
	<indexterm significance="normal"><primary>/etc/nsswitch.conf</primary></indexterm>
	The use of <literal>winbind</literal> requires configuration of NSS. Edit the <filename moreinfo="none">/etc/nsswitch.conf</filename>
	so it includes the following entries:
<screen format="linespecific">
...
passwd: files winbind
shadow: files winbind
group:  files winbind
...
hosts:  files [dns] wins
...
</screen>
	The use of DNS in the hosts entry should be made only if DNS is used on site.
	</para>

	<para>
	The creation of the DMS requires the following steps:
	</para>

	<procedure>
		<step performance="required"><para>
		Create or install an <filename moreinfo="none">smb.conf</filename> file with the above configuration.
		</para></step>

		<step performance="required"><para>
		Execute:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc join -UAdministrator%password
Joined domain MEGANET2.
</screen>
	<indexterm significance="normal"><primary>join</primary></indexterm>
	The success of the join can be confirmed with the following command:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc testjoin
Join to 'MIDEARTH' is OK
</screen>
		A failed join would report an error message like the following:
		<indexterm significance="normal"><primary>failed join</primary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc testjoin
[2004/11/05 16:34:12, 0] utils/net_rpc_join.c:net_rpc_join_ok(66)
Join to domain 'MEGANET2' is not valid
</screen>
		</para></step>

		<step performance="required"><para>
		<indexterm significance="normal"><primary>nmbd</primary></indexterm>
		<indexterm significance="normal"><primary>winbind</primary></indexterm>
		<indexterm significance="normal"><primary>smbd</primary></indexterm>
		Start the <literal>nmbd, winbind,</literal> and <literal>smbd</literal> daemons in the order shown.
		</para></step>
	</procedure>

	</sect3>

	<sect3>
	<title>ADS Domains</title>

	<para>
	<indexterm significance="normal"><primary>domain join</primary></indexterm>
	<indexterm significance="normal"><primary>ADS domain</primary></indexterm>
	The procedure for joining an ADS domain is similar to the NT4 domain join, except the <filename moreinfo="none">smb.conf</filename> file
	will have the contents shown in <link linkend="idmapadsdms">ADS Domain Member Server smb.conf</link>
	</para>

<example id="idmapadsdms">
<title>ADS Domain Member Server smb.conf</title>
<simplelist>
<member># Global parameters</member>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>workgroup</indexterm><parameter>workgroup = BUTTERNET</parameter></member>
<member><indexterm>netbios name</indexterm><parameter>netbios name = GARGOYLE</parameter></member>
<member><indexterm>realm</indexterm><parameter>realm = BUTTERNET.BIZ</parameter></member>
<member><indexterm>security</indexterm><parameter>security = ADS</parameter></member>
<member><indexterm>template shell</indexterm><parameter>template shell = /bin/bash</parameter></member>
<member><indexterm>idmap uid</indexterm><parameter>idmap uid = 500-10000000</parameter></member>
<member><indexterm>idmap gid</indexterm><parameter>idmap gid = 500-10000000</parameter></member>
<member><indexterm>winbind use default domain</indexterm><parameter>winbind use default domain = Yes</parameter></member>
<member><indexterm>winbind nested groups</indexterm><parameter>winbind nested groups = Yes</parameter></member>
<member><indexterm>printer admin</indexterm><parameter>printer admin = "BUTTERNET\Domain Admins"</parameter></member>
</simplelist>
</example>

	<para>
	<indexterm significance="normal"><primary>KRB</primary></indexterm>
	<indexterm significance="normal"><primary>kerberos</primary></indexterm>
	<indexterm significance="normal"><primary>/etc/krb5.conf</primary></indexterm>
	<indexterm significance="normal"><primary>MIT</primary></indexterm>
	<indexterm significance="normal"><primary>MIT kerberos</primary></indexterm>
	<indexterm significance="normal"><primary>Heimdal</primary></indexterm>
	<indexterm significance="normal"><primary>Heimdal kerberos</primary></indexterm>
	ADS DMS operation requires use of kerberos (KRB). For this to work, the <filename moreinfo="none">krb5.conf</filename>
	must be configured. The exact requirements depends on which version of MIT or Heimdal Kerberos is being
	used. It is sound advice to use only the latest version, which at this time are MIT Kerberos version
	1.3.5 and Heimdal 0.61.
	</para>

	<para>
	The creation of the DMS requires the following steps:
	</para>

	<procedure>
		<step performance="required"><para>
		Create or install an <filename moreinfo="none">smb.conf</filename> file with the above configuration.
		</para></step>

		<step performance="required"><para>
		Edit the <filename moreinfo="none">/etc/nsswitch.conf</filename> file as shown above.
		</para></step>

		<step performance="required"><para>
		Execute:
		<indexterm significance="normal"><primary>net</primary><secondary>ads</secondary><tertiary>join</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net ads join -UAdministrator%password
Joined domain BUTTERNET.
</screen>
	The success or failure of the join can be confirmed with the following command:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net ads testjoin
Using short domain name -- BUTTERNET
Joined 'GARGOYLE' to realm 'BUTTERNET.BIZ'
</screen>
	</para>

	<para>
	An invalid or failed join can be detected by executing:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net ads testjoin
GARGOYLE$@'s password:
[2004/11/05 16:53:03, 0] utils/net_ads.c:ads_startup(186)
  ads_connect: No results returned
Join to domain is not valid
</screen>
		<indexterm significance="normal"><primary>error message</primary></indexterm>
		<indexterm significance="normal"><primary>failure</primary></indexterm>
		<indexterm significance="normal"><primary>log level</primary></indexterm>
		<indexterm significance="normal"><primary>identify</primary></indexterm>
		The specific error message may differ from the above because it depends on the type of failure that
		may have occurred. Increase the <parameter moreinfo="none">log level</parameter> to 10, repeat the test,
		and then examine the log files produced to identify the nature of the failure.
		</para></step>

		<step performance="required"><para>
		Start the <literal>nmbd</literal>, <literal>winbind</literal>, and <literal>smbd</literal> daemons in the order shown.
		</para></step>

	</procedure>

	</sect3>
	</sect2>

	<sect2>
	<title>IDMAP_RID with Winbind</title>

	<para>
	<indexterm significance="normal"><primary>idmap_rid</primary></indexterm>
	<indexterm significance="normal"><primary>SID</primary></indexterm>
	<indexterm significance="normal"><primary>RID</primary></indexterm>
	<indexterm significance="normal"><primary>IDMAP</primary></indexterm>
	The <literal>idmap_rid</literal> facility is a new tool that, unlike native winbind, creates a
	predictable mapping of MS Windows SIDs to UNIX UIDs and GIDs. The key benefit of this method
	of implementing the Samba IDMAP facility is that it eliminates the need to store the IDMAP data
	in a central place. The downside is that it can be used only within a single ADS domain and
	is not compatible with trusted domain implementations.
	</para>

	<para>
	<indexterm significance="normal"><primary>SID</primary></indexterm>
	<indexterm significance="normal"><primary>allow trusted domains</primary></indexterm>
	<indexterm significance="normal"><primary>idmap uid</primary></indexterm>
	<indexterm significance="normal"><primary>idmap gid</primary></indexterm>
	This alternate method of SID to UID/GID  mapping can be achieved using the idmap_rid
        plug-in. This plug-in uses the RID of the user SID to derive the UID and GID by adding the
        RID to a base value specified. This utility requires that the parameter
        <quote>allow trusted domains = No</quote> be specified, as it is not compatible
        with multiple domain environments. The <parameter moreinfo="none">idmap uid</parameter> and 
	<parameter moreinfo="none">idmap gid</parameter> ranges must be specified.
	</para>

	<para>
	<indexterm significance="normal"><primary>idmap_rid</primary></indexterm>
	<indexterm significance="normal"><primary>realm</primary></indexterm>
	The idmap_rid facility can be used both for NT4/Samba-style domains and Active Directory.
	To use this with an NT4 domain, do not include the <parameter moreinfo="none">realm</parameter> parameter; additionally, the
	method used to join the domain uses the <constant>net rpc join</constant> process.
	</para>

	<para>
	An example <filename moreinfo="none">smb.conf</filename> file for and ADS domain environment is shown in <link linkend="idmapadsridDMS">ADS
	Domain Member smb.conf using idmap_rid</link>.
	</para>

<example id="idmapadsridDMS">
<title>ADS Domain Member smb.conf using idmap_rid</title>
<simplelist>
<member># Global parameters</member>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>workgroup</indexterm><parameter>workgroup = KPAK</parameter></member>
<member><indexterm>netbios name</indexterm><parameter>netbios name = BIGJOE</parameter></member>
<member><indexterm>realm</indexterm><parameter>realm = CORP.KPAK.COM</parameter></member>
<member><indexterm>server string</indexterm><parameter>server string = Office Server</parameter></member>
<member><indexterm>security</indexterm><parameter>security = ADS</parameter></member>
<member><indexterm>allow trusted domains</indexterm><parameter>allow trusted domains = No</parameter></member>
<member><indexterm>idmap backend</indexterm><parameter>idmap backend = idmap_rid:KPAK=500-100000000</parameter></member>
<member><indexterm>idmap uid</indexterm><parameter>idmap uid = 500-100000000</parameter></member>
<member><indexterm>idmap gid</indexterm><parameter>idmap gid = 500-100000000</parameter></member>
<member><indexterm>template shell</indexterm><parameter>template shell = /bin/bash</parameter></member>
<member><indexterm>winbind use default domain</indexterm><parameter>winbind use default domain = Yes</parameter></member>
<member><indexterm>winbind enum users</indexterm><parameter>winbind enum users = No</parameter></member>
<member><indexterm>winbind enum groups</indexterm><parameter>winbind enum groups = No</parameter></member>
<member><indexterm>winbind nested groups</indexterm><parameter>winbind nested groups = Yes</parameter></member>
<member><indexterm>printer admin</indexterm><parameter>printer admin = "Domain Admins"</parameter></member>
</simplelist>
</example>

	<para>
	<indexterm significance="normal"><primary>large domain</primary></indexterm>
	<indexterm significance="normal"><primary>Active Directory</primary></indexterm>
	<indexterm significance="normal"><primary>response</primary></indexterm>
	<indexterm significance="normal"><primary>getent</primary></indexterm>
	In a large domain with many users it is imperative to disable enumeration of users and groups.
	For example, at a site that has 22,000 users in Active Directory the winbind-based user and
	group resolution is unavailable for nearly 12 minutes following first startup of 
	<literal>winbind</literal>. Disabling enumeration resulted in instantaneous response.
	The disabling of user and group enumeration means that it will not be possible to list users
	or groups using the <literal>getent passwd</literal> and <literal>getent group</literal>
	commands. It will be possible to perform the lookup for individual users, as shown in the following procedure.
	</para>

	<para>
	<indexterm significance="normal"><primary>NSS</primary></indexterm>
	<indexterm significance="normal"><primary>/etc/nsswitch.conf</primary></indexterm>
	The use of this tool requires configuration of NSS as per the native use of winbind. Edit the
	<filename moreinfo="none">/etc/nsswitch.conf</filename> so it has the following parameters:
<screen format="linespecific">
...
passwd: files winbind
shadow: files winbind
group:  files winbind
...
hosts:  files wins
...
</screen>
	</para>

	<para>
	The following procedure can use the idmap_rid facility:
	</para>

	<procedure>
		<step performance="required"><para>
		Create or install an <filename moreinfo="none">smb.conf</filename> file with the above configuration.
		</para></step>

		<step performance="required"><para>
		Edit the <filename moreinfo="none">/etc/nsswitch.conf</filename> file as shown above.
		</para></step>

		<step performance="required"><para>
		Execute:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net ads join -UAdministrator%password
Using short domain name -- KPAK
Joined 'BIGJOE' to realm 'CORP.KPAK.COM'
</screen>
		</para>

		<para>
		<indexterm significance="normal"><primary>failed join</primary></indexterm>
		An invalid or failed join can be detected by executing:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net ads testjoin
BIGJOE$@'s password:
[2004/11/05 16:53:03, 0] utils/net_ads.c:ads_startup(186)
  ads_connect: No results returned
Join to domain is not valid
</screen>
		The specific error message may differ from the above because it depends on the type of failure that
		may have occurred. Increase the <parameter moreinfo="none">log level</parameter> to 10, repeat the test,
		and then examine the log files produced to identify the nature of the failure.
		</para></step>

		<step performance="required"><para>
		Start the <literal>nmbd</literal>, <literal>winbind</literal>, and <literal>smbd</literal> daemons in the order shown.
		</para></step>

		<step performance="required"><para>
		Validate the operation of this configuration by executing:
		<indexterm significance="normal"><primary/></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> getent passwd administrator
administrator:x:1000:1013:Administrator:/home/BE/administrator:/bin/bash
</screen>
		</para></step>
	</procedure>

	</sect2>

	<sect2>
	<title>IDMAP Storage in LDAP Using Winbind</title>

	<para>
	<indexterm significance="normal"><primary>ADAM</primary></indexterm>
	<indexterm significance="normal"><primary>ADS</primary></indexterm>
	The storage of IDMAP information in LDAP can be used with both NT4/Samba-3-style domains and
	ADS domains. OpenLDAP is a commonly used LDAP server for this purpose, although any
	standards-complying LDAP server can be used. It is therefore possible to deploy this IDMAP
	configuration using the Sun iPlanet LDAP server, Novell eDirectory, Microsoft ADS plus ADAM,
	and so on.
	</para>

	<para>
	An example is for an ADS domain is shown in <link linkend="idmapldapDMS">ADS Domain Member Server using
	LDAP</link>.
	</para>

<example id="idmapldapDMS">
<title>ADS Domain Member Server using LDAP</title>
<simplelist>
<member># Global parameters</member>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>workgroup</indexterm><parameter>workgroup = SNOWSHOW</parameter></member>
<member><indexterm>netbios name</indexterm><parameter>netbios name = GOODELF</parameter></member>
<member><indexterm>realm</indexterm><parameter>realm = SNOWSHOW.COM</parameter></member>
<member><indexterm>server string</indexterm><parameter>server string = Samba Server</parameter></member>
<member><indexterm>security</indexterm><parameter>security = ADS</parameter></member>
<member><indexterm>log level</indexterm><parameter>log level = 1 ads:10 auth:10 sam:10 rpc:10</parameter></member>
<member><indexterm>ldap admin dn</indexterm><parameter>ldap admin dn = cn=Manager,dc=SNOWSHOW,dc=COM</parameter></member>
<member><indexterm>ldap idmap suffix</indexterm><parameter>ldap idmap suffix = ou=Idmap</parameter></member>
<member><indexterm>ldap suffix</indexterm><parameter>ldap suffix = dc=SNOWSHOW,dc=COM</parameter></member>
<member><indexterm>idmap backend</indexterm><parameter>idmap backend = ldap:ldap://ldap.snowshow.com</parameter></member>
<member><indexterm>idmap uid</indexterm><parameter>idmap uid = 150000-550000</parameter></member>
<member><indexterm>idmap gid</indexterm><parameter>idmap gid = 150000-550000</parameter></member>
<member><indexterm>template shell</indexterm><parameter>template shell = /bin/bash</parameter></member>
<member><indexterm>winbind use default domain</indexterm><parameter>winbind use default domain = Yes</parameter></member>
</simplelist>
</example>

	<para>
	<indexterm significance="normal"><primary>realm</primary></indexterm>
	In the case of an NT4 or Samba-3-style domain the <parameter moreinfo="none">realm</parameter> is not used, and the
	command used to join the domain is <literal>net rpc join</literal>. The above example also demonstrates
	advanced error-reporting techniques that are documented in <link linkend="dbglvl">Reporting Bugs</link>.
	</para>

	<para>
	<indexterm significance="normal"><primary>MIT kerberos</primary></indexterm>
	<indexterm significance="normal"><primary>Heimdal kerberos</primary></indexterm>
	<indexterm significance="normal"><primary>/etc/krb5.conf</primary></indexterm>
	Where MIT kerberos is installed (version 1.3.4 or later), edit the <filename moreinfo="none">/etc/krb5.conf</filename> 
	file so it has the following contents:
<screen format="linespecific">
[logging]
 default = FILE:/var/log/krb5libs.log
 kdc = FILE:/var/log/krb5kdc.log
 admin_server = FILE:/var/log/kadmind.log

[libdefaults]
 default_realm = SNOWSHOW.COM
 dns_lookup_realm = false
 dns_lookup_kdc = true

[appdefaults]
 pam = {
   debug = false
   ticket_lifetime = 36000
   renew_lifetime = 36000
   forwardable = true
   krb4_convert = false
 }
</screen>
	</para>

	<para>
	Where Heimdal kerberos is installed, edit the <filename moreinfo="none">/etc/krb5.conf</filename>
        file so it is either empty (i.e., no contents) or it has the following contents:
<screen format="linespecific">
[libdefaults]
        default_realm = SNOWSHOW.COM
        clockskew = 300

[realms]
        SNOWSHOW.COM = {
                kdc = ADSDC.SHOWSHOW.COM
        }
        
[domain_realm]
        .snowshow.com = SNOWSHOW.COM
</screen>
	</para>

	<note><para>
	Samba cannot use the Heimdal libraries if there is no <filename moreinfo="none">/etc/krb5.conf</filename> file.
	So long as there is an empty file, the Heimdal kerberos libraries will be usable. There is no
	need to specify any settings because Samba, using the Heimdal libraries, can figure this out automatically.
	</para></note>

	<para>
	Edit the NSS control file <filename moreinfo="none">/etc/nsswitch.conf</filename> so it has the following entries:
<screen format="linespecific">
...
passwd: files ldap
shadow: files ldap
group:  files ldap
...
hosts:  files wins
...
</screen>
	</para>

	<para>
	<indexterm significance="normal"><primary>PADL</primary></indexterm>
	<indexterm significance="normal"><primary>/etc/ldap.conf</primary></indexterm>
	You will need the <ulink url="http://www.padl.com">PADL</ulink> <literal>nss_ldap</literal> 
	tool set for this solution. Configure the <filename moreinfo="none">/etc/ldap.conf</filename> file so it has 
	the information needed. The following is an example of a working file:
<screen format="linespecific">
host    192.168.2.1
base    dc=snowshow,dc=com
binddn  cn=Manager,dc=snowshow,dc=com
bindpw  not24get

pam_password exop

nss_base_passwd ou=People,dc=snowshow,dc=com?one
nss_base_shadow ou=People,dc=snowshow,dc=com?one
nss_base_group  ou=Groups,dc=snowshow,dc=com?one
ssl     no
</screen>
	</para>

	<para>
	The following procedure may be followed to effect a working configuration:
	</para>

	<procedure>
		<step performance="required"><para>
		Configure the <filename moreinfo="none">smb.conf</filename> file as shown above.
		</para></step>

		<step performance="required"><para>
		Create the <filename moreinfo="none">/etc/krb5.conf</filename> file as shown above.
		</para></step>

		<step performance="required"><para>
		Configure the <filename moreinfo="none">/etc/nsswitch.conf</filename> file as shown above.
		</para></step>

		<step performance="required"><para>
		Download, build, and install the PADL nss_ldap tool set. Configure the 
		<filename moreinfo="none">/etc/ldap.conf</filename> file as shown above.
		</para></step>

		<step performance="required"><para>
		Configure an LDAP server and initialize the directory with the top-level entries needed by IDMAP,
		shown in the following LDIF file:
<screen format="linespecific">
dn: dc=snowshow,dc=com
objectClass: dcObject
objectClass: organization
dc: snowshow
o: The Greatest Snow Show in Singapore.
description: Posix and Samba LDAP Identity Database

dn: cn=Manager,dc=snowshow,dc=com
objectClass: organizationalRole
cn: Manager
description: Directory Manager

dn: ou=Idmap,dc=snowshow,dc=com
objectClass: organizationalUnit
ou: idmap
</screen>
		</para></step>

		<step performance="required"><para>
		Execute the command to join the Samba DMS to the ADS domain as shown here:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net ads testjoin
Using short domain name -- SNOWSHOW
Joined 'GOODELF' to realm 'SNOWSHOW.COM'
</screen>
		</para></step>

		<step performance="required"><para>
		Store the LDAP server access password in the Samba <filename moreinfo="none">secrets.tdb</filename> file as follows:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> smbpasswd -w not24get
</screen>
		</para></step>

		<step performance="required"><para>
		Start the <literal>nmbd</literal>, <literal>winbind</literal>, and <literal>smbd</literal> daemons in the order shown.
		</para></step>
	</procedure>

	<para>
	<indexterm significance="normal"><primary>diagnostic</primary></indexterm>
	Follow the diagnositic procedures shown earlier in this chapter to identify success or failure of the join.
	In many cases a failure is indicated by a silent return to the command prompt with no indication of the
	reason for failure.
	</para>

	</sect2>

	<sect2>
	<title>IDMAP and NSS Using LDAP from ADS with RFC2307bis Schema Extension</title>

	<para>
	<indexterm significance="normal"><primary>rfc2307bis</primary></indexterm>
	<indexterm significance="normal"><primary>schema</primary></indexterm>
	The use of this method is messy. The information provided in the following is for guidance only
	and is very definitely not complete. This method does work; it is used in a number of large sites
	and has an acceptable level of performance.
	</para>

	<para>
	An example <filename moreinfo="none">smb.conf</filename> file is shown in <link linkend="idmaprfc2307">ADS Domain Member Server using
RFC2307bis Schema Extension Date via NSS</link>.
	</para>

<example id="idmaprfc2307">
<title>ADS Domain Member Server using RFC2307bis Schema Extension Date via NSS</title>
<simplelist>
<member># Global parameters</member>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>workgroup</indexterm><parameter>workgroup = BOBBY</parameter></member>
<member><indexterm>realm</indexterm><parameter>realm = BOBBY.COM</parameter></member>
<member><indexterm>security</indexterm><parameter>security = ADS</parameter></member>
<member><indexterm>idmap uid</indexterm><parameter>idmap uid = 150000-550000</parameter></member>
<member><indexterm>idmap gid</indexterm><parameter>idmap gid = 150000-550000</parameter></member>
<member><indexterm>template shell</indexterm><parameter>template shell = /bin/bash</parameter></member>
<member><indexterm>winbind cache time</indexterm><parameter>winbind cache time = 5</parameter></member>
<member><indexterm>winbind use default domain</indexterm><parameter>winbind use default domain = Yes</parameter></member>
<member><indexterm>winbind trusted domains only</indexterm><parameter>winbind trusted domains only = Yes</parameter></member>
<member><indexterm>winbind nested groups</indexterm><parameter>winbind nested groups = Yes</parameter></member>
</simplelist>
</example>

	<para>
	<indexterm significance="normal"><primary>nss_ldap</primary></indexterm>
	The DMS must be joined to the domain using the usual procedure. Additionally, it is necessary
	to build and install the PADL nss_ldap tool set. Be sure to build this tool set with the
	following:
<screen format="linespecific">
./configure --enable-rfc2307bis --enable-schema-mapping
make install
</screen> 
	</para>

	<para>
	<indexterm significance="normal"><primary>/etc/nsswitch.conf</primary></indexterm>
	The following <filename moreinfo="none">/etc/nsswitch.conf</filename> file contents are required:
<screen format="linespecific">
...
passwd: files ldap
shadow: files ldap
group:  files ldap
...
hosts:  files wins
...
</screen>
	</para>

	<para>
	<indexterm significance="normal"><primary>/etc/ldap.conf</primary></indexterm>
	<indexterm significance="normal"><primary>nss_ldap</primary></indexterm>
	The <filename moreinfo="none">/etc/ldap.conf</filename> file must be configured also. Refer to the PADL documentation
	and source code for nss_ldap to specific instructions.
	</para>

	<para>
	The next step involves preparation of the ADS schema. This is briefly discussed in the remaining
	part of this chapter.
	</para>

		<sect3>
		<title>IDMAP, Active Directory, and MS Services for UNIX 3.5</title>

		<para>
		<indexterm significance="normal"><primary>SFU</primary></indexterm>
		The Microsoft Windows Service for UNIX (SFU) version 3.5 is available for free 
		<ulink url="http://www.microsoft.com/windows/sfu/">download</ulink>
		from the Microsoft Web site. You will need to download this tool and install it following
		Microsoft instructions.
		</para>

		</sect3>

		<sect3>
		<title>IDMAP, Active Directory and AD4UNIX</title>

		<para>
		Instructions for obtaining and installing the AD4UNIX tool set can be found from the
		<ulink url="http://www.geekcomix.com/cgi-bin/classnotes/wiki.pl?LDAP01/An_Alternative_Approach">
		Geekcomix</ulink> Web site.
		</para>

		</sect3>

	</sect2>

</sect1>

</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="rights">
<chapterinfo>
        <author>
<firstname>Gerald</firstname><surname>Carter</surname><othername>(Jerry)</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jerry@samba.org</email></address>
</affiliation></author>
        <author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
</chapterinfo>

<title>User Rights and Privileges</title>

<para>
<indexterm significance="normal"><primary>Windows user</primary></indexterm>
<indexterm significance="normal"><primary>Windows group</primary></indexterm>
<indexterm significance="normal"><primary>machine accounts</primary></indexterm>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
The administration of Windows user, group, and machine accounts in the Samba
domain-controlled network necessitates interfacing between the MS Windows
networking environment and the UNIX operating system environment. The right
(permission) to add machines to the Windows security domain can be assigned
(set) to non-administrative users both in Windows NT4 domains and
Active Directory domains.
</para>

<para>
<indexterm significance="normal"><primary>Windows NT4/2kX/XPPro</primary></indexterm>
<indexterm significance="normal"><primary>machine account</primary></indexterm>
<indexterm significance="normal"><primary>trusted</primary></indexterm>
<indexterm significance="normal"><primary>user logons</primary></indexterm>
The addition of Windows NT4/2kX/XPPro machines to the domain necessitates the
creation of a machine account for each machine added. The machine account is
a necessity that is used to validate that the machine can be trusted to permit
user logons.
</para>

<para>
<indexterm significance="normal"><primary>user accounts</primary></indexterm>
<indexterm significance="normal"><primary>special account</primary></indexterm>
<indexterm significance="normal"><primary>account name</primary></indexterm>
<indexterm significance="normal"><primary>/bin/false</primary></indexterm>
<indexterm significance="normal"><primary>/dev/null</primary></indexterm>
<indexterm significance="normal"><primary>man-in-the-middle</primary></indexterm>
Machine accounts are analogous to user accounts, and thus in implementing them on a UNIX machine that is
hosting Samba (i.e., on which Samba is running), it is necessary to create a special type of user account.
Machine accounts differ from  normal user accounts in that the account name (login ID) is terminated with a
<literal moreinfo="none">$</literal> sign. An additional difference is that this type of account should not ever be able to
log into the UNIX environment as a system user and therefore is set to have a shell of
<literal>/bin/false</literal> and a home directory of <literal>/dev/null.</literal> The machine
account is used only to authenticate domain member machines during start-up. This security measure
is designed to block man-in-the-middle attempts to violate network integrity.
</para>

<note><para>
<indexterm significance="normal"><primary>computer accounts</primary></indexterm>
<indexterm significance="normal"><primary>domain member servers</primary></indexterm>
<indexterm significance="normal"><primary>domain controller</primary></indexterm>
<indexterm significance="normal"><primary>credentials</primary></indexterm>
<indexterm significance="normal"><primary>secure authentication</primary></indexterm>
Machine (computer) accounts are used in the Windows NT OS family to store security
credentials for domain member servers and workstations. When the domain member
starts up, it goes through a validation process that includes an exchange of
credentials with a domain controller. If the domain member fails to authenticate
using the credentials known for it by domain controllers, the machine will be refused
all access by domain users. The computer account is essential to the way that MS 
Windows secures authentication.
</para></note>

<para>
<indexterm significance="normal"><primary>UNIX system accounts</primary></indexterm>
<indexterm significance="normal"><primary>system administrator</primary></indexterm>
<indexterm significance="normal"><primary>root</primary></indexterm>
<indexterm significance="normal"><primary>UID</primary></indexterm>
The creation of UNIX system accounts has traditionally been the sole right of
the system administrator, better known as the <constant>root</constant> account.
It is possible in the UNIX environment to create multiple users who have the
same UID. Any UNIX user who has a UID=0 is inherently the same as the
<constant>root</constant> account user.
</para>

<para>
<indexterm significance="normal"><primary>system interface scripts</primary></indexterm>
<indexterm significance="normal"><primary>CIFS function calls</primary></indexterm>
<indexterm significance="normal"><primary>root account</primary></indexterm>
<indexterm significance="normal"><primary>UNIX host system</primary></indexterm>
All versions of Samba call system interface scripts that permit CIFS function
calls that are used to manage users, groups, and machine accounts
in the UNIX environment. All versions of Samba up to and including version 3.0.10
required the use of a Windows administrator account that unambiguously maps to
the UNIX <constant>root</constant> account to permit the execution of these
interface scripts. The requirement to do this has understandably met with some
disdain and consternation among Samba administrators, particularly where it became
necessary to permit people who should not possess <constant>root</constant>-level
access to the UNIX host system.
</para>

<sect1>
<title>Rights Management Capabilities</title>

<para>
<indexterm significance="normal"><primary>Windows privilege model</primary></indexterm>
<indexterm significance="normal"><primary>privilege model</primary></indexterm>
<indexterm significance="normal"><primary>rights assigned</primary></indexterm>
<indexterm significance="normal"><primary>SID</primary></indexterm>
Samba 3.0.11 introduced support for the Windows privilege model.  This model
allows certain rights to be assigned to a user or group SID.  In order to enable
this feature, <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="ENABLEPRIVILEGES" xlink:href="smb.conf.5.html#ENABLEPRIVILEGES">enable privileges = yes</link>
must be defined in the <parameter>global</parameter> section of the <filename moreinfo="none">smb.conf</filename> file.
</para>

<para>
<indexterm significance="normal"><primary>rights</primary></indexterm>
<indexterm significance="normal"><primary>privileges</primary></indexterm>
<indexterm significance="normal"><primary>manage privileges</primary></indexterm>
Currently, the rights supported in Samba-3 are listed in <link linkend="rp-privs"/>.
The remainder of this chapter explains how to manage and use these privileges on Samba servers.
</para>

<indexterm significance="normal"><primary>SeMachineAccountPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SePrintOperatorPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeAddUsersPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeRemoteShutdownPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeDiskOperatorPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeTakeOwnershipPrivilege</primary></indexterm>
<table id="rp-privs">
        <title>Current Privilege Capabilities</title>
        <tgroup cols="2">
                <colspec align="right"/>
                <colspec align="left"/>
                <thead>
                        <row>
                                <entry align="left">Privilege</entry>
                                <entry align="left">Description</entry>
                        </row>
                </thead>
                <tbody>
                        <row>
                                <entry><para>SeMachineAccountPrivilege</para></entry>
                                <entry><para>Add machines to domain</para></entry>
                        </row>
                        <row>
                                <entry><para>SePrintOperatorPrivilege</para></entry>
                                <entry><para>Manage printers</para></entry>
                        </row>
                        <row>
                                <entry><para>SeAddUsersPrivilege</para></entry>
                                <entry><para>Add users and groups to the domain</para></entry>
                        </row>
                        <row>
                                <entry><para>SeRemoteShutdownPrivilege</para></entry>
                                <entry><para>Force shutdown from a remote system</para></entry>
                        </row>
                        <row>
                                <entry><para>SeDiskOperatorPrivilege</para></entry>
                                <entry><para>Manage disk share</para></entry>
                        </row>
<!-- These are not used at this time - so void them from the docs.
                        <row>
                                <entry><para>SeBackupPrivilege</para></entry>
                                <entry><para>Back up files and directories</para></entry>
                        </row>
                        <row>
                                <entry><para>SeRestorePrivilege</para></entry>
                                <entry><para>Restore files and directories</para></entry>
                        </row>
**** End of commented out section **** -->    
                        <row>
                                <entry><para>SeTakeOwnershipPrivilege</para></entry>
                                <entry><para>Take ownership of files or other objects</para></entry>
                        </row>
            </tbody>
        </tgroup>
</table>

<sect2>
<title>Using the <quote>net rpc rights</quote> Utility</title>

<para>
<indexterm significance="normal"><primary>managing rights</primary></indexterm>
<indexterm significance="normal"><primary>rights assigned</primary></indexterm>
<indexterm significance="normal"><primary>NT4 User Manager for Domains</primary></indexterm>
<indexterm significance="normal"><primary>command-line utility</primary></indexterm>
<indexterm significance="normal"><primary>administrative actions</primary></indexterm>
There are two primary means of managing the rights assigned to users and groups
on a Samba server.  The <literal>NT4 User Manager for Domains</literal> may be
used from any Windows NT4, 2000, or XP Professional domain member client to
connect to a Samba domain controller and view/modify the rights assignments.
This application, however, appears to have bugs when run on a client running
Windows 2000 or later; therefore, Samba provides a command-line utility for
performing the necessary administrative actions.
</para>

<para>
The <literal>net rpc rights</literal> utility in Samba 3.0.11 has three new subcommands:
</para>

<variablelist>
	<varlistentry><term>list [name|accounts]</term>
		<listitem><para>
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>list</tertiary></indexterm>
<indexterm significance="normal"><primary>available rights</primary></indexterm>
<indexterm significance="normal"><primary>privileges assigned</primary></indexterm>
<indexterm significance="normal"><primary>privileged accounts</primary></indexterm>
		When called with no arguments, <literal>net rpc list</literal>
		simply lists the available rights on the server.  When passed
		a specific user or group name, the tool lists the privileges
		currently assigned to the specified account.  When invoked using
		the special string <constant>accounts</constant>,
		<literal>net rpc rights list</literal> returns a list of all
		privileged accounts on the server and the assigned rights.
		</para></listitem>
	</varlistentry>

	<varlistentry><term>grant &lt;user&gt; &lt;right [right ...]&gt;</term>
		<listitem><para>
<indexterm significance="normal"><primary>assign rights</primary></indexterm>
<indexterm significance="normal"><primary>grant rights</primary></indexterm>
<indexterm significance="normal"><primary>add client machines</primary></indexterm>
<indexterm significance="normal"><primary>user or group</primary></indexterm>
		When called with no arguments, this function is used to assign
		a list of rights to a specified user or group.  For example,
		to grant the members of the Domain Admins group on a Samba domain controller,
		the capability to add client machines to the domain, one would run:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net -S server -U domadmin rpc rights grant \
	 'DOMAIN\Domain Admins' SeMachineAccountPrivilege
</screen>
		The following syntax has the same result:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>rights grant</tertiary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc rights grant 'DOMAIN\Domain Admins' \
     SeMachineAccountPrivilege -S server -U domadmin
</screen>
		More than one privilege can be assigned by specifying a
		list of rights separated by spaces. The parameter 'Domain\Domain Admins'
		must be quoted with single ticks or using double-quotes to prevent
		the backslash and the space from being interpreted by the system shell.
		</para></listitem>
	</varlistentry>

	<varlistentry><term>revoke &lt;user&gt; &lt;right [right ...]&gt;</term>
		<listitem><para>
		This command is similar in format to <literal>net rpc rights grant</literal>.  Its
		effect is to remove an assigned right (or list of rights) from a user or group.
		</para></listitem>
	</varlistentry>

</variablelist>

<note><para>
<indexterm significance="normal"><primary>member</primary></indexterm>
<indexterm significance="normal"><primary>Domain Admins</primary></indexterm>
<indexterm significance="normal"><primary>revoke privileges</primary></indexterm>
You must be connected as a member of the Domain Admins group to be able to grant or revoke privileges assigned
to an account.  This capability is inherent to the Domain Admins group and is not configurable. There are no
default rights and privileges, except the ability for a member of the Domain Admins group to assign them.
This means that all administrative rights and privileges (other than the ability to assign them) must be 
explicitly assigned, even for the Domain Admins group.
</para></note>

<para>
<indexterm significance="normal"><primary>performed as root</primary></indexterm>
<indexterm significance="normal"><primary>necessary rights</primary></indexterm>
<indexterm significance="normal"><primary>add machine script</primary></indexterm>
<indexterm significance="normal"><primary/></indexterm>
By default, no privileges are initially assigned to any account because certain actions will be performed as
root once smbd determines that a user has the necessary rights.  For example, when joining a client to a
Windows domain, <parameter moreinfo="none">add machine script</parameter> must be executed with superuser rights in most 
cases.  For this reason, you should be very careful about handing out privileges to accounts.
</para>

<para>
<indexterm significance="normal"><primary>Access</primary></indexterm>
<indexterm significance="normal"><primary>root user</primary></indexterm>
<indexterm significance="normal"><primary>bypasses privilege</primary></indexterm>
Access as the root user (UID=0) bypasses all privilege checks.
</para>

</sect2>

<sect2>
<title>Description of Privileges</title>

<para>
<indexterm significance="normal"><primary>privileges</primary></indexterm>
<indexterm significance="normal"><primary>additional privileges</primary></indexterm>
<indexterm significance="normal"><primary>house-keeping</primary></indexterm>
The privileges that have been implemented in Samba-3.0.11 are shown below.  It is possible, and likely, that
additional privileges may be implemented in later releases of Samba. It is also likely that any privileges
currently implemented but not used may be removed from future releases as a housekeeping matter, so it is
important that the successful as well as unsuccessful use of these facilities should be reported on the Samba
mailing lists.
</para>

<variablelist>
	<varlistentry><term>SeAddUsersPrivilege</term>
		<listitem><para>
<indexterm significance="normal"><primary>SeAddUsersPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>smbd</primary></indexterm>
<indexterm significance="normal"><primary>net rpc user add</primary></indexterm>
		This right determines whether or not smbd will allow the
		user to create new user or group accounts via such tools
		as <literal>net rpc user add</literal> or 
		<literal>NT4 User Manager for Domains.</literal>
		</para></listitem>
	</varlistentry>

	<varlistentry><term>SeDiskOperatorPrivilege</term>
		<listitem><para>
<indexterm significance="normal"><primary>SeDiskOperatorPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>add/delete/change share</primary></indexterm>
<indexterm significance="normal"><primary>ACL</primary></indexterm>
		Accounts that possess this right will be able to execute
		scripts defined by the <literal>add/delete/change</literal>
		share command in <filename moreinfo="none">smb.conf</filename> file as root.  Such users will
		also be able to modify the ACL associated with file shares
		on the Samba server.
		</para></listitem>
	</varlistentry>

	<varlistentry><term>SeMachineAccountPrivilege</term>
		<listitem><para>
<indexterm significance="normal"><primary>SeMachineAccountPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>right to join domain</primary></indexterm>
<indexterm significance="normal"><primary>join client</primary></indexterm>
		This right controls whether or not the user can join client
		machines to a Samba-controlled domain.
		</para></listitem>
	</varlistentry>

	<varlistentry><term>SePrintOperatorPrivilege</term>
		<listitem><para>
<indexterm significance="normal"><primary>SePrintOperatorPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>privilege</primary></indexterm>
<indexterm significance="normal"><primary>global right</primary></indexterm>
<indexterm significance="normal"><primary>administrative rights</primary></indexterm>
<indexterm significance="normal"><primary>printers admin</primary></indexterm>
		This privilege operates identically to the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTERADMIN" xlink:href="smb.conf.5.html#PRINTERADMIN">printer admin</link>
		option in the <filename moreinfo="none">smb.conf</filename> file (see section 5 man page for <filename moreinfo="none">smb.conf</filename>)
		except that it is a global right (not on a per-printer basis). 
		Eventually the smb.conf option will be deprecated and administrative
		rights to printers will be controlled exclusively by this right and
		the security descriptor associated with the printer object in the
		<filename moreinfo="none">ntprinters.tdb</filename> file.
		</para></listitem>
	</varlistentry>

	<varlistentry><term>SeRemoteShutdownPrivilege</term>
		<listitem><para>
<indexterm significance="normal"><primary>SeRemoteShutdownPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>rebooting server</primary></indexterm>
<indexterm significance="normal"><primary>aborting shutdown</primary></indexterm>
		Samba provides two hooks for shutting down or rebooting
		the server and for aborting a previously issued shutdown
		command.  Since this is an operation normally limited by
		the operating system to the root user, an account must possess this
		right to be able to execute either of these hooks.
		</para></listitem>
	</varlistentry>

	<varlistentry><term>SeTakeOwnershipPrivilege</term>
		<listitem><para>
<indexterm significance="normal"><primary>SeTakeOwnershipPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>take ownership</primary></indexterm>
		This right permits users to take ownership of files and directories.
		</para></listitem>
	</varlistentry>

</variablelist>

</sect2>

<sect2>
<title>Privileges Suppored by Windows 2000 Domain Controllers</title>

<para>
    For reference purposes, a Windows NT4 Primary Domain Controller reports support for the following
	privileges:
<indexterm significance="normal"><primary>SeCreateTokenPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeAssignPrimaryTokenPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeLockMemoryPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeIncreaseQuotaPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeMachineAccountPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeTcbPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeSecurityPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeTakeOwnershipPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeLoadDriverPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeSystemProfilePrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeSystemtimePrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeProfileSingleProcessPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeIncreaseBasePriorityPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeCreatePagefilePrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeCreatePermanentPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeBackupPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeRestorePrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeShutdownPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeDebugPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeAuditPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeSystemEnvironmentPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeChangeNotifyPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeRemoteShutdownPrivilege</primary></indexterm>
<screen format="linespecific">
         SeCreateTokenPrivilege  Create a token object
  SeAssignPrimaryTokenPrivilege  Replace a process level token
          SeLockMemoryPrivilege  Lock pages in memory
       SeIncreaseQuotaPrivilege  Increase quotas
      SeMachineAccountPrivilege  Add workstations to domain
                 SeTcbPrivilege  Act as part of the operating system
            SeSecurityPrivilege  Manage auditing and security log
       SeTakeOwnershipPrivilege  Take ownership of files or other objects
          SeLoadDriverPrivilege  Load and unload device drivers
       SeSystemProfilePrivilege  Profile system performance
          SeSystemtimePrivilege  Change the system time
SeProfileSingleProcessPrivilege  Profile single process
SeIncreaseBasePriorityPrivilege  Increase scheduling priority
      SeCreatePagefilePrivilege  Create a pagefile
     SeCreatePermanentPrivilege  Create permanent shared objects
              SeBackupPrivilege  Back up files and directories
             SeRestorePrivilege  Restore files and directories
            SeShutdownPrivilege  Shut down the system
               SeDebugPrivilege  Debug programs
               SeAuditPrivilege  Generate security audits
   SeSystemEnvironmentPrivilege  Modify firmware environment values
        SeChangeNotifyPrivilege  Bypass traverse checking
      SeRemoteShutdownPrivilege  Force shutdown from a remote system
</screen>
	And Windows 200x/XP Domain Controllers and workstations reports to support the following privileges:
<indexterm significance="normal"><primary>SeCreateTokenPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeAssignPrimaryTokenPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeLockMemoryPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeIncreaseQuotaPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeMachineAccountPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeTcbPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeSecurityPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeTakeOwnershipPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeLoadDriverPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeSystemProfilePrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeSystemtimePrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeProfileSingleProcessPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeIncreaseBasePriorityPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeCreatePagefilePrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeCreatePermanentPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeBackupPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeRestorePrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeShutdownPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeDebugPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeAuditPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeSystemEnvironmentPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeChangeNotifyPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeRemoteShutdownPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeUndockPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeSyncAgentPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeEnableDelegationPrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeManageVolumePrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeImpersonatePrivilege</primary></indexterm>
<indexterm significance="normal"><primary>SeCreateGlobalPrivilege</primary></indexterm>
<screen format="linespecific">
         SeCreateTokenPrivilege  Create a token object
  SeAssignPrimaryTokenPrivilege  Replace a process level token
          SeLockMemoryPrivilege  Lock pages in memory
       SeIncreaseQuotaPrivilege  Increase quotas
      SeMachineAccountPrivilege  Add workstations to domain
                 SeTcbPrivilege  Act as part of the operating system
            SeSecurityPrivilege  Manage auditing and security log
       SeTakeOwnershipPrivilege  Take ownership of files or other objects
          SeLoadDriverPrivilege  Load and unload device drivers
       SeSystemProfilePrivilege  Profile system performance
          SeSystemtimePrivilege  Change the system time
SeProfileSingleProcessPrivilege  Profile single process
SeIncreaseBasePriorityPrivilege  Increase scheduling priority
      SeCreatePagefilePrivilege  Create a pagefile
     SeCreatePermanentPrivilege  Create permanent shared objects
              SeBackupPrivilege  Back up files and directories
             SeRestorePrivilege  Restore files and directories
            SeShutdownPrivilege  Shut down the system
               SeDebugPrivilege  Debug programs
               SeAuditPrivilege  Generate security audits
   SeSystemEnvironmentPrivilege  Modify firmware environment values
        SeChangeNotifyPrivilege  Bypass traverse checking
      SeRemoteShutdownPrivilege  Force shutdown from a remote system
              SeUndockPrivilege  Remove computer from docking station
           SeSyncAgentPrivilege  Synchronize directory service data
    SeEnableDelegationPrivilege  Enable computer and user accounts to
                                 be trusted for delegation
        SeManageVolumePrivilege  Perform volume maintenance tasks
         SeImpersonatePrivilege  Impersonate a client after authentication
        SeCreateGlobalPrivilege  Create global objects
</screen>
<indexterm significance="normal"><primary>equivalence</primary></indexterm>
    The Samba Team is implementing only those privileges that are logical and useful in the UNIX/Linux
    environment. Many of the Windows 200X/XP privileges have no direct equivalence in UNIX.
    </para>

</sect2>

</sect1>

<sect1>
<title>The Administrator Domain SID</title>

<para>
<indexterm significance="normal"><primary>domain Administrator</primary></indexterm>
<indexterm significance="normal"><primary>User Rights and Privileges</primary></indexterm>
<indexterm significance="normal"><primary>passdb backend</primary></indexterm>
<indexterm significance="normal"><primary>SID</primary></indexterm>
<indexterm significance="normal"><primary>net getlocalsid</primary></indexterm>
Please note that every Windows NT4 and later server requires a domain Administrator account. Samba versions
commencing with 3.0.11 permit Administrative duties to be performed via assigned rights and privileges
(see <link linkend="rights">User Rights and Privileges</link>).  An account in the server's passdb backend can
be set to the well-known RID of the default administrator account.  To obtain the domain SID on a Samba domain
controller, run the following command:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net getlocalsid
SID for domain FOO is: S-1-5-21-4294955119-3368514841-2087710299
</screen>
<indexterm significance="normal"><primary>RID</primary></indexterm>
You may assign the domain administrator RID to an account using the <literal>pdbedit</literal>
command as shown here:
<indexterm significance="normal"><primary>pdbedit</primary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> pdbedit -U S-1-5-21-4294955119-3368514841-2087710299-500 -u root -r
</screen>
</para>

<note><para>
<indexterm significance="normal"><primary>RID 500</primary></indexterm>
<indexterm significance="normal"><primary>well known RID</primary></indexterm>
<indexterm significance="normal"><primary>rights and privileges</primary></indexterm>
<indexterm significance="normal"><primary>root account</primary></indexterm>
The RID 500 is the well known standard value of the default Administrator account. It is the RID
that confers the rights and privileges that the Administrator account has on a Windows machine
or domain. Under UNIX/Linux the equivalent is UID=0 (the root account).
</para></note>

<para>
<indexterm significance="normal"><primary>without Administrator account</primary></indexterm>
<indexterm significance="normal"><primary>equivalent rights and privileges</primary></indexterm>
<indexterm significance="normal"><primary>Windows group account</primary></indexterm>
<indexterm significance="normal"><primary>3.0.11</primary></indexterm>
Releases of Samba version 3.0.11 and later make it possible to operate without an Administrator account
provided equivalent rights and privileges have been established for a Windows user or a Windows
group account. 
</para>

</sect1>

<sect1>
<title>Common Errors</title>

	<sect2>
	<title>What Rights and Privileges Will Permit Windows Client Administration?</title>

	<para>
<indexterm significance="normal"><primary>domain global</primary></indexterm>
<indexterm significance="normal"><primary>local group</primary></indexterm>
<indexterm significance="normal"><primary>administrative rights</primary></indexterm>
<indexterm significance="normal"><primary>Windows client</primary></indexterm>
	When a Windows NT4 (or later) client joins a domain, the domain global <literal moreinfo="none">Domain Admins</literal> group
	is added to the membership of the local <literal moreinfo="none">Administrators</literal> group on the client. Any user who is
	a member of the domain global <literal moreinfo="none">Domain Admins</literal> group will have administrative rights on the
	Windows client.
	</para>

	<para>
<indexterm significance="normal"><primary>desirable solution</primary></indexterm>
<indexterm significance="normal"><primary>administrative rights and privileges</primary></indexterm>
<indexterm significance="normal"><primary>Power Users</primary></indexterm>
<indexterm significance="normal"><primary>domain global user</primary></indexterm>
<indexterm significance="normal"><primary>domain global group</primary></indexterm>
	This is often not the most desirable solution because it means that the user will have administrative
	rights and privileges on domain servers also. The <literal moreinfo="none">Power Users</literal> group on Windows client
	workstations permits local administration of the workstation alone. Any domain global user or domain global
	group can be added to the membership of the local workstation group <literal moreinfo="none">Power Users</literal>.
	</para>

	<para>
<indexterm significance="normal"><primary>Nested Group Support</primary></indexterm>
<indexterm significance="normal"><primary>add domain users and groups to a local group</primary></indexterm>
<indexterm significance="normal"><primary>net</primary></indexterm>
<indexterm significance="normal"><primary>Windows workstation.</primary></indexterm>
	See <link linkend="nestedgrpmgmgt">Nested Group Support</link> for an example of how to add domain users
	and groups to a local group that is on a Windows workstation. The use of the <literal>net</literal>
	command permits this to be done from the Samba server.
	</para>

	<para>
<indexterm significance="normal"><primary>cmd</primary></indexterm>
<indexterm significance="normal"><primary>cmd shell</primary></indexterm>
<indexterm significance="normal"><primary>net</primary><secondary>localgroup</secondary></indexterm>
	Another way this can be done is to log onto the Windows workstation as the user
	<literal moreinfo="none">Administrator</literal>, then open a <literal>cmd</literal> shell, then execute:
<screen format="linespecific">
<prompt moreinfo="none">C:\&gt; </prompt> net localgroup administrators /add <userinput moreinfo="none">domain_name\entity</userinput>
</screen>
	where <literal moreinfo="none">entity</literal> is either a domain user or a domain group account name.
	</para>

	</sect2>

</sect1>

</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="AccessControls">
<chapterinfo>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
	
<author>
	<firstname>Jeremy</firstname><surname>Allison</surname>
	<affiliation>
		<orgname>Samba Team</orgname>
		<address format="linespecific"><email>jra@samba.org</email></address>
	</affiliation>
</author>
	<author>
<firstname>Jelmer</firstname><othername>R.</othername><surname>Vernooij</surname><othername>R.</othername>
<affiliation>
	<orgname>The Samba Team</orgname>
	<address format="linespecific"><email>jelmer@samba.org</email></address>
</affiliation><contrib>drawing</contrib></author>
	<pubdate>May 10, 2003</pubdate>
</chapterinfo>
<title>File, Directory, and Share Access Controls</title>

<para>
<indexterm significance="normal"><primary>ACLs</primary></indexterm>
<indexterm significance="normal"><primary>share</primary></indexterm>
<indexterm significance="normal"><primary>network access controls</primary></indexterm>
<indexterm significance="normal"><primary>unauthorized access</primary></indexterm>
Advanced MS Windows users are frequently perplexed when file, directory, and share manipulation of
resources shared via Samba do not behave in the manner they might expect. MS Windows network
administrators are often confused regarding network access controls and how to
provide users with the access they need while protecting resources from unauthorized access.
</para>

<para>
<indexterm significance="normal"><primary>file access permissions</primary></indexterm>
<indexterm significance="normal"><primary>directory access permissions</primary></indexterm>
Many UNIX administrators are unfamiliar with the MS Windows environment and in particular
have difficulty in visualizing what the MS Windows user wishes to achieve in attempts to set file
and directory access permissions. 
</para>

<para>
<indexterm significance="normal"><primary>bridge</primary></indexterm>
<indexterm significance="normal"><primary>directory controls</primary></indexterm>
<indexterm significance="normal"><primary>directory permissions</primary></indexterm>
<indexterm significance="normal"><primary/></indexterm>
The problem lies in the differences in how file and directory permissions and controls work
between the two environments. This difference is one that Samba cannot completely hide, even
though it does try to bridge the chasm to a degree.
</para>

<para>
<indexterm significance="normal"><primary>Extended Attributes</primary></indexterm>
<indexterm significance="normal"><primary>ACLs</primary><secondary>POSIX</secondary></indexterm>
<indexterm significance="normal"><primary>Access Control List</primary></indexterm>
<indexterm significance="normal"><primary>commercial Linux products</primary></indexterm>
POSIX Access Control List technology has been available (along with extended attributes)
for UNIX for many years, yet there is little evidence today of any significant use. This
explains to some extent the slow adoption of ACLs into commercial Linux products. MS Windows
administrators are astounded at this, given that ACLs were a foundational capability of the now
decade-old MS Windows NT operating system.
</para>

<para>
<indexterm significance="normal"><primary>network administrator</primary></indexterm>
The purpose of this chapter is to present each of the points of control that are possible with
Samba-3 in the hope that this will help the network administrator to find the optimum method
for delivering the best environment for MS Windows desktop users.
</para>

<para>
<indexterm significance="normal"><primary>interoperability</primary></indexterm>
<indexterm significance="normal"><primary>data interchange</primary></indexterm>
This is an opportune point to mention that Samba was created to provide a means of interoperability
and interchange of data between differing operating environments. Samba has no intent to change
UNIX/Linux into a platform like MS Windows. Instead the purpose was and is to provide a sufficient
level of exchange of data between the two environments. What is available today extends well
beyond early plans and expectations, yet the gap continues to shrink.
</para>

<sect1>
<title>Features and Benefits</title>

	<para>
	Samba offers much flexibility in file system access management. These are the key access control
	facilities present in Samba today:
	</para>

	<itemizedlist>
	<title>Samba Access Control Facilities</title>
		<listitem><para>
		<indexterm significance="normal"><primary>permissions</primary><secondary>UNIX file and directory</secondary></indexterm>
		<emphasis>UNIX File and Directory Permissions</emphasis>
		</para>

			<para>
<indexterm significance="normal"><primary>UNIX file system access controls</primary></indexterm>
<indexterm significance="normal"><primary>access controls</primary></indexterm>
<indexterm significance="normal"><primary>permissions and controls</primary></indexterm>
			Samba honors and implements UNIX file system access controls. Users
			who access a Samba server will do so as a particular MS Windows user.
			This information is passed to the Samba server as part of the logon or
			connection setup process. Samba uses this user identity to validate
			whether or not the user should be given access to file system resources
			(files and directories). This chapter provides an overview for those
			to whom the UNIX permissions and controls are a little strange or unknown.
			</para>
		</listitem>

		<listitem><para>
		<emphasis>Samba Share Definitions</emphasis>
		</para>

			<para>
<indexterm significance="normal"><primary>share settings</primary></indexterm>
			In configuring share settings and controls in the <filename moreinfo="none">smb.conf</filename> file,
			the network administrator can exercise overrides to native file
			system permissions and behaviors. This can be handy and convenient
			to effect behavior that is more like what MS Windows NT users expect,
			but it is seldom the <emphasis>best</emphasis> way to achieve this.
			The basic options and techniques are described herein.
			</para>
		</listitem>

		<listitem><para>
		<emphasis>Samba Share ACLs</emphasis>
		<indexterm significance="normal"><primary>ACLs</primary><secondary>share</secondary></indexterm>
		</para>

			<para>
<indexterm significance="normal"><primary>ACLs on shares</primary></indexterm>
			Just as it is possible in MS Windows NT to set ACLs on shares
			themselves, so it is possible to do in Samba.
			Few people make use of this facility, yet it remains one of the
			easiest ways to affect access controls (restrictions) and can often
			do so with minimum invasiveness compared with other methods.
			</para>
		</listitem>

		<listitem><para>
				<indexterm significance="normal"><primary>ACLs</primary><secondary>POSIX</secondary></indexterm>
				<indexterm significance="normal"><primary>ACLs</primary><secondary>Windows</secondary></indexterm>
		<emphasis>MS Windows ACLs through UNIX POSIX ACLs</emphasis>
		</para>

			<para>
<indexterm significance="normal"><primary>native ACLs</primary></indexterm>
			The use of POSIX ACLs on UNIX/Linux is possible only if the underlying
			operating system supports them. If not, then this option will not be
			available to you. Current UNIX technology platforms have native support
			for POSIX ACLs. There are patches for the Linux kernel that also provide
			this support. Sadly, few Linux platforms ship today with native ACLs and
			extended attributes enabled. This chapter has pertinent information
			for users of platforms that support them.
			</para>
		</listitem>
	</itemizedlist>

</sect1>

<sect1>
<title>File System Access Controls</title>

<para>
Perhaps the most important recognition to be made is the simple fact that MS Windows NT4/200x/XP
implement a totally divergent file system technology from what is provided in the UNIX operating system
environment. First we consider what the most significant differences are, then we look
at how Samba helps to bridge the differences.
</para>

	<sect2>
	<title>MS Windows NTFS Comparison with UNIX File Systems</title>

	<para>
	<indexterm significance="normal"><primary>NTFS</primary></indexterm>
	<indexterm significance="normal"><primary>File System</primary></indexterm>
	<indexterm significance="normal"><primary>File System</primary><secondary>UNIX</secondary></indexterm>
	<indexterm significance="normal"><primary>File System</primary><secondary>Windows</secondary></indexterm>
	Samba operates on top of the UNIX file system. This means it is subject to UNIX file system conventions
	and permissions. It also means that if the MS Windows networking environment requires file system
	behavior, that differs from UNIX file system behavior then somehow Samba is responsible for emulating
	that in a transparent and consistent manner.
	</para>

	<para>
	It is good news that Samba does this to a large extent, and on top of that, provides a high degree
	of optional configuration to override the default behavior. We look at some of these overrides,
	but for the greater part we stay within the bounds of default behavior. Those wishing to explore
	the depths of control ability should review the <filename moreinfo="none">smb.conf</filename> man page.
	</para>

	<para>The following compares file system features for UNIX with those of MS Windows NT/200x:
	<indexterm significance="normal"><primary>File System</primary><secondary>feature comparison</secondary></indexterm>
	
	</para>

	<variablelist>
		<varlistentry>
			<term>Name Space</term>
			<listitem>
		<para>
		MS Windows NT4/200x/XP file names may be up to 254 characters long, and UNIX file names
		may be 1023 characters long. In MS Windows, file extensions indicate particular file types;
		in UNIX this is not so rigorously observed because all names are considered arbitrary. 
		</para>
		<para>
		What MS Windows calls a folder, UNIX calls a directory.
		</para>
			</listitem>
		</varlistentry>

		<varlistentry>
			<term>Case Sensitivity</term>
			<listitem>
		<para>
		<indexterm significance="normal"><primary>8.3 file names</primary></indexterm>
		<indexterm significance="normal"><primary>File System</primary><secondary>case sensitivity</secondary></indexterm>
		MS Windows file names are generally uppercase if made up of 8.3 (8-character file name
		and 3 character extension. File names that are longer than 8.3 are case preserving and case
		insensitive.
		</para>

		<para>
		UNIX file and directory names are case sensitive and case preserving. Samba implements the
		MS Windows file name behavior, but it does so as a user application. The UNIX file system
		provides no mechanism to perform case-insensitive file name lookups. MS Windows does this
		by default. This means that Samba has to carry the processing overhead to provide features
		that are not native to the UNIX operating system environment.
		</para>

		<para>
		Consider the following. All are unique UNIX names but one single MS Windows file name:
		<screen format="linespecific">
				MYFILE.TXT
				MyFile.txt
				myfile.txt
		</screen></para>

		<para>
		So clearly, in an MS Windows file namespace these three files cannot co-exist, but in UNIX
		they can.
		</para>

		<para>
		So what should Samba do if all three are present? That which is lexically first will be
		accessible to MS Windows users; the others are invisible and unaccessible <?latex --- ?> any
		other solution would be suicidal. The Windows client will ask for a case-insensitive file
		lookup, and that is the reason for which Samba must offer a consistent selection in the
		event that the UNIX directory contains multiple files that would match a case insensitive
		file listing.
		</para></listitem>
		</varlistentry>

		<varlistentry>
		<term>Directory Separators</term>
		<listitem><para>
		<indexterm significance="normal"><primary>Directory Separators</primary></indexterm>
		MS Windows and DOS use the backslash <constant>\</constant> as a directory delimiter, and UNIX uses
		the forward-slash <constant>/</constant> as its directory delimiter. This is handled transparently by Samba.
		</para></listitem>
		</varlistentry>

		<varlistentry>
		<term>Drive Identification</term>
		<listitem><para>
		<indexterm significance="normal"><primary>Drive Identification</primary></indexterm>
		MS Windows products support a notion of drive letters, like <literal>C:</literal>, to represent
		disk partitions. UNIX has no concept of separate identifiers for file partitions; each
		such file system is mounted to become part of the overall directory tree.
		The UNIX directory tree begins at <constant>/</constant> just as the root of a DOS drive is specified as
		<constant>C:\</constant>.
		</para></listitem>
		</varlistentry>

		<varlistentry>
		<term>File Naming Conventions</term>
		<listitem><para>
		<indexterm significance="normal"><primary>File Naming Conventions</primary></indexterm>
		MS Windows generally never experiences file names that begin with a dot (<constant>.</constant>), while in UNIX these
		are commonly found in a user's home directory. Files that begin with a dot (<constant>.</constant>) are typically
		startup files for various UNIX applications, or they may be files that contain
		startup configuration data.
		</para></listitem>
		</varlistentry>

		<varlistentry>
		<term>Links and Short-Cuts</term>
		<listitem><para>
		<indexterm significance="normal"><primary>Links</primary><secondary>hard</secondary></indexterm>
		<indexterm significance="normal"><primary>Links</primary><secondary>soft</secondary></indexterm>
		<indexterm significance="normal"><primary>Shortcuts</primary></indexterm>
		MS Windows make use of <emphasis>links and shortcuts</emphasis> that are actually special types of files that will
		redirect an attempt to execute the file to the real location of the file. UNIX knows of file and directory
		links, but they are entirely different from what MS Windows users are used to.
		</para>

		<para>
		Symbolic links are files in UNIX that contain the actual location of the data (file or directory). An
		operation (like read or write) will operate directly on the file referenced. Symbolic links are also
		referred to as <quote>soft links.</quote> A hard link is something that MS Windows is not familiar with. It allows
		one physical file to be known simultaneously by more than one file name.
		</para></listitem>
		</varlistentry>
	</variablelist>

	<para>
	There are many other subtle differences that may cause the MS Windows administrator some temporary discomfort
	in the process of becoming familiar with UNIX/Linux. These are best left for a text that is dedicated to the
	purpose of UNIX/Linux training and education.
	</para>

	</sect2>

	<sect2>
	<title>Managing Directories</title>

	<para>
<indexterm significance="normal"><primary>create</primary></indexterm>
<indexterm significance="normal"><primary>delete</primary></indexterm>
<indexterm significance="normal"><primary>rename</primary></indexterm>
	There are three basic operations for managing directories: <literal>create</literal>, <literal>delete</literal>,
	<literal>rename</literal>. <link linkend="TOSH-Accesstbl">Managing Directories with UNIX and
	Windows</link> compares the commands in Windows and UNIX that implement these operations.
	</para>

	<table frame="all" id="TOSH-Accesstbl">
		<title>Managing Directories with UNIX and Windows</title>
		<tgroup align="center" cols="3">
		<thead>
		<row><entry>Action</entry><entry>MS Windows Command</entry><entry>UNIX Command</entry></row>
		</thead>
	
		<tbody>
			<row><entry>create</entry><entry>md folder</entry><entry>mkdir folder</entry></row>
			<row><entry>delete</entry><entry>rd folder</entry><entry>rmdir folder</entry></row>
			<row><entry>rename</entry><entry>rename oldname newname</entry><entry>mv oldname newname</entry></row>
		</tbody>
	</tgroup>
	</table>

	</sect2>

	<sect2>
	<title>File and Directory Access Control</title>

	<para>
	<indexterm significance="normal"><primary>ACLs</primary><secondary>File System</secondary></indexterm>
<indexterm significance="normal"><primary>POSIX ACLs</primary></indexterm>
<indexterm significance="normal"><primary>EAs</primary></indexterm>
	The network administrator is strongly advised to read basic UNIX training manuals and reference materials
	regarding file and directory permissions maintenance. Much can be achieved with the basic UNIX permissions
	without having to resort to more complex facilities like POSIX ACLs or extended attributes (EAs).
	</para>

	<para>
	UNIX/Linux file and directory access permissions involves setting three primary sets of data and one control set.
	A UNIX file listing looks as follows:
<screen format="linespecific">
<prompt moreinfo="none">$ </prompt><userinput moreinfo="none">ls -la</userinput>
total 632
drwxr-xr-x   13 maryo   gnomes      816 2003-05-12 22:56 .
drwxrwxr-x   37 maryo   gnomes     3800 2003-05-12 22:29 ..
dr-xr-xr-x    2 maryo   gnomes       48 2003-05-12 22:29 muchado02
drwxrwxrwx    2 maryo   gnomes       48 2003-05-12 22:29 muchado03
drw-rw-rw-    2 maryo   gnomes       48 2003-05-12 22:29 muchado04
d-w--w--w-    2 maryo   gnomes       48 2003-05-12 22:29 muchado05
dr--r--r--    2 maryo   gnomes       48 2003-05-12 22:29 muchado06
drwsrwsrwx    2 maryo   gnomes       48 2003-05-12 22:29 muchado08
----------    1 maryo   gnomes     1242 2003-05-12 22:31 mydata00.lst
--w--w--w-    1 maryo   gnomes     7754 2003-05-12 22:33 mydata02.lst
-r--r--r--    1 maryo   gnomes    21017 2003-05-12 22:32 mydata04.lst
-rw-rw-rw-    1 maryo   gnomes    41105 2003-05-12 22:32 mydata06.lst
<prompt moreinfo="none">$ </prompt>
</screen>
	</para>

	<para>
	The columns represent (from left to right) permissions, number of hard links to file, owner, group, size
	(bytes), access date, time of last modification, and file name.
	</para>

	<para>
	An overview of the permissions field is shown in <link linkend="access1">Overview of UNIX permissions
	field</link>.
	</para>

	<figure id="access1" float="0">
		<title>Overview of UNIX permissions field.</title>
		<mediaobject><imageobject role="html"><imagedata fileref="images/access1.png" scale="40" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/access1.png" scale="40" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/access1"/></imageobject></mediaobject>
	</figure>

	<para>
		Any bit flag may be unset. An unset bit flag is the equivalent of "cannot" and is represented
		as a <quote>-</quote> character (see <link linkend="access2"/>)
<indexterm significance="normal"><primary>read</primary></indexterm>
<indexterm significance="normal"><primary>write</primary></indexterm>
<indexterm significance="normal"><primary>execute</primary></indexterm>
<indexterm significance="normal"><primary>user</primary></indexterm>
<indexterm significance="normal"><primary>group</primary></indexterm>
<indexterm significance="normal"><primary>other</primary></indexterm>
	</para>

<example id="access2">
<title>Example File</title>
<programlisting format="linespecific">
-rwxr-x---   Means: 
 ^^^                The owner (user) can read, write, execute
    ^^^             the group can read and execute
       ^^^          everyone else cannot do anything with it.
</programlisting>
</example>


	<para>
<indexterm significance="normal"><primary>character device</primary></indexterm>
<indexterm significance="normal"><primary>block device</primary></indexterm>
<indexterm significance="normal"><primary>pipe device</primary></indexterm>
<indexterm significance="normal"><primary>UNIX Domain Socket</primary></indexterm>
	Additional possibilities in the [type] field are c = character device, b = block device, p = pipe device,
	s = UNIX Domain Socket.
	</para>

	<para>
<indexterm significance="normal"><primary>read</primary></indexterm>
<indexterm significance="normal"><primary>write</primary></indexterm>
<indexterm significance="normal"><primary>execute</primary></indexterm>
<indexterm significance="normal"><primary>SGID</primary></indexterm>
<indexterm significance="normal"><primary>SUID</primary></indexterm>
	The letters <constant>rwxXst</constant> set permissions for the user, group, and others as read (r), write (w),
	execute (or access for directories) (x), execute  only  if  the  file  is a directory or already has execute
	permission for some user (X), set user (SUID) or group ID (SGID) on execution (s), sticky (t).
	</para>

	<para>
<indexterm significance="normal"><primary>sticky bit</primary></indexterm>
<indexterm significance="normal"><primary>unlinked</primary></indexterm>
<indexterm significance="normal"><primary>/tmp</primary></indexterm>
<indexterm significance="normal"><primary>world-writable</primary></indexterm>
	When the sticky bit is set on a directory, files in that directory may be unlinked (deleted) or renamed only by root or their owner. 
	Without the sticky  bit, anyone able to write to the directory can delete or rename files. The sticky bit is commonly found on
	directories, such as <filename moreinfo="none">/tmp</filename>, that are world-writable.
	</para>

	<para>
<indexterm significance="normal"><primary>write</primary></indexterm>
<indexterm significance="normal"><primary>read</primary></indexterm>
<indexterm significance="normal"><primary>setting up directories</primary></indexterm>
<indexterm significance="normal"><primary>set user id</primary><see>SUID</see></indexterm>
<indexterm significance="normal"><primary>set group id</primary><see>SGID</see></indexterm>
	When the set user or group ID bit (s) is set on a directory, then all files created within it will be owned by the user and/or
	group whose `set user or group' bit is set. This can be helpful in setting up directories for which it is desired that
	all users who are in a group should be able to write to and read from a file, particularly when it is undesirable for that file
	to be exclusively owned by a user whose primary group is not the group that all such users belong to.
	</para>

	<para>
	When a directory is set <constant>d-wx--x---</constant>, the owner can read and create (write) files in it, but because
	the (r) read flags are not set, files cannot be listed (seen) in the directory by anyone. The group can read files in the
	directory but cannot create new files. If files in the directory are set to be readable and writable for the group, then
	group members will be able to write to (or delete) them.
	</para>

	<sect3>
	<title>Protecting Directories and Files from Deletion</title>

	<para>
<indexterm significance="normal"><primary>protect files</primary></indexterm>
<indexterm significance="normal"><primary>protect directories</primary></indexterm>
<indexterm significance="normal"><primary>access controls</primary></indexterm>
<indexterm significance="normal"><primary>capability to delete</primary></indexterm>
	People have asked on the Samba mailing list how is it possible to protect files or directories from deletion by users.
	For example, Windows NT/2K/XP provides the capacity to set access controls on a directory into which people can
	write files but not delete them. It is possible to set an ACL on a Windows file that permits the file to be written to
	but not deleted. Such concepts are foreign to the UNIX operating system file space. Within the UNIX file system
	anyone who has the ability to create a file can write to it. Anyone who has write permission on the
	directory that contains a file and has write permission for it has the capability to delete it.
	</para>

	<para>
<indexterm significance="normal"><primary>directory permissions</primary></indexterm>
<indexterm significance="normal"><primary>delete a file</primary></indexterm>
<indexterm significance="normal"><primary>write access</primary></indexterm>
	For the record, in the UNIX environment the ability to delete a file is controlled by the permissions on
	the directory that the file is in. In other words, a user can delete a file in a directory to which that
	user has write access, even if that user does not own the file.
	</para>
	
	<para>
<indexterm significance="normal"><primary>file system capabilities</primary></indexterm>
<indexterm significance="normal"><primary>inheritance</primary></indexterm>
<indexterm significance="normal"><primary>POSIX ACLs</primary></indexterm>
<indexterm significance="normal"><primary>extended attributes</primary></indexterm>
	Of necessity, Samba is subject to the file system semantics of the host operating system. Samba is therefore
	limited in the file system capabilities that can be made available through Windows ACLs, and therefore performs
	a "best fit" translation to POSIX ACLs. Some UNIX file systems do, however support, a feature known
	as extended attributes. Only the Windows concept of <emphasis>inheritance</emphasis> is implemented by Samba through
	the appropriate extended attribute.
	</para>	

	<para>
<indexterm significance="normal"><primary>extended attributes</primary></indexterm>
<indexterm significance="normal"><primary>immutible</primary></indexterm>
<indexterm significance="normal"><primary>chattr</primary></indexterm>
<indexterm significance="normal"><primary>CAP_LINUX_IMMUTABLE</primary></indexterm>
	The specific semantics of the extended attributes are not consistent across UNIX and UNIX-like systems such as Linux.
	For example, it is possible on some implementations of the extended attributes to set a flag that prevents the directory
	or file from being deleted. The extended attribute that may achieve this is called the <constant>immutible</constant> bit.
	Unfortunately, the implementation of the immutible flag is NOT consistent with published documentation. For example, the
	man page for the <literal>chattr</literal> on SUSE Linux 9.2 says:
<screen format="linespecific">
A file with the i attribute cannot be modified: it cannot be deleted
or renamed, no link can be created to this file and no data can be
written to the file. Only the superuser or a process possessing the
CAP_LINUX_IMMUTABLE capability can set or clear this attribute.
</screen>
	A simple test can be done to check if the immutible flag is supported on files in the file system of the Samba host
	server.
	</para>

	<procedure>
	<title>Test for File Immutibility Support</title>

	<step performance="required"><para>
	Create a file called <filename moreinfo="none">filename</filename>.
	</para></step>

	<step performance="required"><para>
	Login as the <constant>root</constant> user, then set the immutibile flag on a test file as follows:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> chattr +i `filename'
</screen>
	</para></step>

	<step performance="required"><para>
	Login as the user who owns the file (not root) and attempt to remove the file as follows:
<screen format="linespecific">
mystic:/home/hannibal &gt; rm filename
</screen>
	It will not be possible to delete the file if the immutible flag is correctly honored.
	</para></step>
	</procedure>

	<para>
	On operating systems and file system types that support the immutible bit, it is possible to create directories
	that cannot be deleted. Check the man page on your particular host system to determine whether or not
	immutable directories are writable. If they are not, then the entire directory and its contents will effectively
	be protected from writing (file creation also) and deletion.
	</para>

	</sect3>

	</sect2>

</sect1>

<sect1>
<title>Share Definition Access Controls</title>


	<para>
	<indexterm significance="normal"><primary>permissions</primary><secondary>share</secondary></indexterm>
	The following parameters in the <filename moreinfo="none">smb.conf</filename> file sections define a share control or affect access controls.
	Before using any of the following options, please refer to the man page for <filename moreinfo="none">smb.conf</filename>.
	</para>

	<sect2>
	<title>User- and Group-Based Controls</title>

	<para>
	User- and group-based controls can prove quite useful. In some situations it is distinctly desirable to
	force all file system operations as if a single user were doing so. The use of the
	<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="FORCEUSER" xlink:href="smb.conf.5.html#FORCEUSER">force user</link> and <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="FORCEGROUP" xlink:href="smb.conf.5.html#FORCEGROUP">force group</link> behavior will achieve this.
	In other situations it may be necessary to use a paranoia level of control to ensure that only particular
	authorized persons will be able to access a share or its contents. Here the use of the
	<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="VALIDUSERS" xlink:href="smb.conf.5.html#VALIDUSERS">valid users</link> or the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="INVALIDUSERS" xlink:href="smb.conf.5.html#INVALIDUSERS">invalid users</link> parameter may be useful.
	</para>

	<para>
	As always, it is highly advisable to use the easiest to maintain and the least ambiguous method for
	controlling access. Remember, when you leave the scene, someone else will need to provide assistance, and
	if he or she finds too great a mess or does not understand what you have done, there is risk of
	Samba being removed and an alternative solution being adopted.
	</para>

	<para>
	<link linkend="ugbc">User and Group Based Controls</link> enumerates these controls.
	</para>

	<table frame="all" pgwide="0" id="ugbc"><title>User- and Group-Based Controls</title>
	<tgroup cols="2">
		<colspec align="left"/>
		<colspec align="justify" colwidth="1*"/>
		<thead>
		<row>
			<entry align="center">Control Parameter</entry>
			<entry align="center">Description, Action, Notes</entry>
		</row>
		</thead>
		<tbody>
		<row>
			<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="ADMINUSERS" xlink:href="smb.conf.5.html#ADMINUSERS">admin users</link></entry>
			<entry><para>
			List of users who will be granted administrative privileges on the share.
			They will do all file operations as the superuser (root). 
			Users in this list will be able to do anything they like on the share,
			irrespective of file permissions.
			</para></entry>
		</row>
		<row>
			<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="FORCEGROUP" xlink:href="smb.conf.5.html#FORCEGROUP">force group</link></entry>
			<entry><para>
			Specifies a UNIX group name that will be assigned as the default primary group
			for all users connecting to this service.
			</para></entry>
		</row>
		<row>
			<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="FORCEUSER" xlink:href="smb.conf.5.html#FORCEUSER">force user</link></entry>
			<entry><para>
			Specifies a UNIX username that will be assigned as the default user for all users connecting to this service.
			This is useful for sharing files. Incorrect use can cause security problems.
			</para></entry>
		</row>
		<row>
			<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="GUESTOK" xlink:href="smb.conf.5.html#GUESTOK">guest ok</link></entry>
			<entry><para>
			If this parameter is set for a service, then no password is required to connect to the service. Privileges will be 
			those of the  guest account.
			</para></entry>
		</row>
		<row>
			<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="INVALIDUSERS" xlink:href="smb.conf.5.html#INVALIDUSERS">invalid users</link></entry>
			<entry><para>
			List of users that should not be allowed to login to this service.
			</para></entry>
		</row>
		<row>
			<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="ONLYUSER" xlink:href="smb.conf.5.html#ONLYUSER">only user</link></entry>
			<entry><para>
			Controls whether connections with usernames not in the user list will be allowed.
			</para></entry>
		</row>
		<row>
			<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="READLIST" xlink:href="smb.conf.5.html#READLIST">read list</link></entry>
			<entry><para>
			List of users that are given read-only access to a service. Users in this list
			will not be given write access, no matter what the  read-only  option is set to. 
			</para></entry>
		</row>
		<row>
			<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="USERNAME" xlink:href="smb.conf.5.html#USERNAME">username</link></entry>
			<entry><para>
			Refer to the <filename moreinfo="none">smb.conf</filename> man page for more information; this is a complex and potentially misused parameter.
			</para></entry>
		</row>
		<row>
			<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="VALIDUSERS" xlink:href="smb.conf.5.html#VALIDUSERS">valid users</link></entry>
			<entry><para>
			List of users that should be allowed to login to this service.
			</para></entry>
		</row>
		<row>
			<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="WRITELIST" xlink:href="smb.conf.5.html#WRITELIST">write list</link></entry>
			<entry><para>
			List of users that are given read-write access to a service.
			</para></entry>
		</row>
		</tbody>
	</tgroup>
	</table>

	</sect2>

	<sect2>
	<title>File and Directory Permissions-Based Controls</title>

	<para>
	Directory permission-based controls, if misused, can result in considerable difficulty in diagnosing the causes of 
	misconfiguration. Use them sparingly and carefully. By gradually introducing each, one at a time, undesirable side 
	effects may be detected. In the event of a problem, always comment all of them out and then gradually reintroduce 
	them in a controlled way.
	</para>

	<para>
	Refer to <link linkend="fdpbc">File and Directory Permission Based Controls</link> for information 
	regarding the parameters that may be used to set file and directory permission-based access controls.
	</para>

	<table frame="all" id="fdpbc"><title>File and Directory Permission-Based Controls</title>
		<tgroup cols="2">
			<colspec align="left"/>
			<colspec align="justify" colwidth="1*"/>
		<thead>
		<row>
			<entry align="center">Control Parameter</entry>
			<entry align="center">Description, Action, Notes</entry>
		</row>
		</thead>
		<tbody>
		<row>
			<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="CREATEMASK" xlink:href="smb.conf.5.html#CREATEMASK">create mask</link></entry>
			<entry><para>
			Refer to the <filename moreinfo="none">smb.conf</filename> man page.
			</para></entry>
		</row>
		<row>
			<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="DIRECTORYMASK" xlink:href="smb.conf.5.html#DIRECTORYMASK">directory mask</link></entry>
			<entry><para>
			The octal modes used when converting DOS modes to UNIX modes when creating UNIX directories.
			See also directory security mask.
			</para></entry></row>
		<row>
			<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="DOSFILEMODE" xlink:href="smb.conf.5.html#DOSFILEMODE">dos filemode</link></entry>
			<entry><para>
			Enabling this parameter allows a user who has write access to the file to modify the permissions on it.
			</para></entry>
		</row>
		<row>
			<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="FORCECREATEMODE" xlink:href="smb.conf.5.html#FORCECREATEMODE">force create mode</link></entry>
			<entry><para>
			This parameter specifies a set of UNIX-mode bit permissions that will always be set on a file created by Samba.
			</para></entry>
		</row>
		<row>
			<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="FORCEDIRECTORYMODE" xlink:href="smb.conf.5.html#FORCEDIRECTORYMODE">force directory mode</link></entry>
			<entry><para>
			This parameter specifies a set of UNIX-mode bit permissions that will always be set on a directory created by Samba.
			</para></entry>
		</row>
		<row>
			<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="FORCEDIRECTORYSECURITYMODE" xlink:href="smb.conf.5.html#FORCEDIRECTORYSECURITYMODE">force directory security mode</link></entry>
			<entry><para>
			Controls UNIX permission bits modified when a Windows NT client is manipulating UNIX permissions on a directory.
			</para></entry>
		</row>
		<row>
			<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="FORCESECURITYMODE" xlink:href="smb.conf.5.html#FORCESECURITYMODE">force security mode</link></entry>
			<entry><para>
			Controls UNIX permission bits modified when a Windows NT client manipulates UNIX permissions.
			</para></entry>
		</row>
		<row>
			<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="HIDEUNREADABLE" xlink:href="smb.conf.5.html#HIDEUNREADABLE">hide unreadable</link></entry>
			<entry><para>
			Prevents clients from seeing the existence of files that cannot be read.
			</para></entry>
		</row>
		<row>
			<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="HIDEUNWRITEABLEFILES" xlink:href="smb.conf.5.html#HIDEUNWRITEABLEFILES">hide unwriteable files</link></entry>
			<entry><para>
			Prevents clients from seeing the existence of files that cannot be written to. Unwritable directories are shown as usual. 
			</para></entry>
		</row>
		<row>
			<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="NTACLSUPPORT" xlink:href="smb.conf.5.html#NTACLSUPPORT">nt acl support</link></entry>
			<entry><para>
			This parameter controls whether smbd will attempt to map UNIX permissions into Windows NT ACLs.
			</para></entry>
		</row>
		<row>
			<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SECURITYMASK" xlink:href="smb.conf.5.html#SECURITYMASK">security mask</link></entry>
			<entry><para>
			Controls UNIX permission bits modified when a Windows NT client is manipulating the UNIX permissions on a file.
			</para></entry>
		</row>
		</tbody>
	</tgroup>
	</table>

	</sect2>

	<sect2>
	<title>Miscellaneous Controls</title>

	<para>
	The parameters documented in <link linkend="mcoc">Other Controls</link> are often used by administrators
	in ways that create inadvertent barriers to file access. Such are the consequences of not understanding the 
	full implications of <filename moreinfo="none">smb.conf</filename> file settings.
	</para>

	<table frame="all" id="mcoc"><title>Other Controls</title>
	<tgroup cols="2">
		<colspec align="justify" colwidth="1*"/>
		<colspec align="justify" colwidth="1*"/>
		<thead>
		<row>
			<entry align="center">Control Parameter</entry>
			<entry align="center">Description, Action, Notes</entry>
		</row>
		</thead>
		<tbody>
		<row>
			<entry>
			<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="CASESENSITIVE" xlink:href="smb.conf.5.html#CASESENSITIVE">case sensitive</link>,
			<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="DEFAULTCASE" xlink:href="smb.conf.5.html#DEFAULTCASE">default case</link>,
			<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SHORTPRESERVECASE" xlink:href="smb.conf.5.html#SHORTPRESERVECASE">short preserve case</link>
			</entry>
			<entry><para>
			This means that all file name lookup will be done in a case-sensitive manner. 
			Files will be created with the precise file name Samba received from the MS Windows client.
			</para></entry>
		</row>
		<row>
			<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="CSCPOLICY" xlink:href="smb.conf.5.html#CSCPOLICY">csc policy</link></entry>
			<entry><para>
			Client-side caching policy parallels MS Windows client-side file caching capabilities.
			</para></entry>
		</row>
		<row>
			<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="DONTDESCEND" xlink:href="smb.conf.5.html#DONTDESCEND">dont descend</link></entry>
			<entry><para>
			Allows specifying a comma-delimited list of directories that the server should always show as empty.
			</para></entry>
		</row>
		<row>
			<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="DOSFILETIMERESOLUTION" xlink:href="smb.conf.5.html#DOSFILETIMERESOLUTION">dos filetime resolution</link></entry>
			<entry><para>
			This option is mainly used as a compatibility option for Visual C++ when used against Samba shares.
			</para></entry>
		</row>
		<row>
			<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="DOSFILETIMES" xlink:href="smb.conf.5.html#DOSFILETIMES">dos filetimes</link></entry>
			<entry><para>
			DOS and Windows allow users to change file timestamps if they can write to the file. POSIX semantics prevent this.
			This option allows DOS and Windows behavior.
			</para></entry>
		</row>
		<row>
			<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="FAKEOPLOCKS" xlink:href="smb.conf.5.html#FAKEOPLOCKS">fake oplocks</link></entry>
			<entry><para>
			Oplocks are the way that SMB clients get permission from a server to locally cache file operations. If a server grants an
			oplock, the client is free to assume that it is the only one accessing the file, and it will aggressively cache file data.
			</para></entry>
		</row>
		<row>
			<entry>
			<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="HIDEDOTFILES" xlink:href="smb.conf.5.html#HIDEDOTFILES">hide dot files</link>,
			<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="HIDEFILES" xlink:href="smb.conf.5.html#HIDEFILES">hide files</link>,
			<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="VETOFILES" xlink:href="smb.conf.5.html#VETOFILES">veto files</link>
			</entry>
			<entry><para>
			Note: MS Windows Explorer allows override of files marked as hidden so they will still be visible.
			</para></entry>
		</row>
		<row>
			<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="READONLY" xlink:href="smb.conf.5.html#READONLY">read only</link></entry>
			<entry><para>
			If this parameter is yes, then users of a service may not create or modify files in the service's directory.
			</para></entry>
		</row>
		<row>
			<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="VETOFILES" xlink:href="smb.conf.5.html#VETOFILES">veto files</link></entry>
			<entry><para>
			List of files and directories that are neither visible nor accessible.
			</para></entry>
		</row>
		</tbody>
	</tgroup>
	</table>

	</sect2>

</sect1>

<sect1>
<title>Access Controls on Shares</title>


	<para>
<indexterm significance="normal"><primary>per-share access control</primary></indexterm>
<indexterm significance="normal"><primary>Everyone - Full Control</primary></indexterm>
<indexterm significance="normal"><primary>specific restrictions</primary></indexterm>
<indexterm significance="normal"><primary>share access</primary></indexterm>
	<indexterm significance="normal"><primary>permissions</primary><secondary>share ACLs</secondary></indexterm>
	This section deals with how to configure Samba per-share access control restrictions.
	By default, Samba sets no restrictions on the share itself. Restrictions on the share itself
	can be set on MS Windows NT4/200x/XP shares. This can be an effective way to limit who can
	connect to a share. In the absence of specific restrictions, the default setting is to allow
	the global user <constant>Everyone - Full Control</constant> (full control, change and read).
	</para>

	<para>
<indexterm significance="normal"><primary>access control</primary></indexterm>
<indexterm significance="normal"><primary>MMC</primary></indexterm>
<indexterm significance="normal"><primary>Computer Management</primary></indexterm>
	At this time Samba does not provide a tool for configuring access control settings on the share
	itself the only way to create those settings is to use either the NT4 Server Manager or the Windows 200x
	Microsoft Management Console (MMC) for Computer Management. There are currently no plans to provide
	this capability in the Samba command-line tool set.
	</para>

	<para>
<indexterm significance="normal"><primary>share_info.tdb</primary></indexterm>
<indexterm significance="normal"><primary>/usr/local/samba/var</primary></indexterm>
<indexterm significance="normal"><primary>tdbdump</primary></indexterm>
<indexterm significance="normal"><primary>tdb files</primary></indexterm>
	Samba stores the per-share access control settings in a file called <filename moreinfo="none">share_info.tdb</filename>.
	The location of this file on your system will depend on how Samba was compiled. The default location
	for Samba's tdb files is under <filename moreinfo="none">/usr/local/samba/var</filename>. If the <filename moreinfo="none">tdbdump</filename>
	utility has been compiled and installed on your system, then you can examine the contents of this file
	by executing <literal>tdbdump share_info.tdb</literal> in the directory containing the tdb files.
	</para>

	<sect2>
	<title>Share Permissions Management</title>

		<para>
		The best tool for share permissions management is platform-dependent. Choose the best tool for your environment.
		</para>

			<sect3>
			<title>Windows NT4 Workstation/Server</title>
			<para>
<indexterm significance="normal"><primary>manage share permissions</primary></indexterm>
<indexterm significance="normal"><primary>share permissions</primary></indexterm>
<indexterm significance="normal"><primary>NT Server Manager</primary></indexterm>
<indexterm significance="normal"><primary>Windows NT4</primary></indexterm>
			The tool you need to manage share permissions on a Samba server from a Windows NT4 Workstation or Server
			is the NT Server Manager.  Server Manager is shipped with Windows NT4 Server products but not with Windows
			NT4 Workstation.  You can obtain the NT Server Manager for MS Windows NT4 Workstation from the Microsoft
			web site <ulink url="http://support.microsoft.com/default.aspx?scid=kb;en-us;173673">support</ulink> section.
			</para>

			<procedure>
			<title>Instructions</title>

			<step performance="required"><para>
			Launch the <application moreinfo="none">NT4 Server Manager</application> and click on the Samba server you want to
			administer. From the menu select <guimenu moreinfo="none">Computer</guimenu>, then click on
			<guimenuitem moreinfo="none">Shared Directories</guimenuitem>.
			</para></step>

			<step performance="required"><para>
			Click on the share that you wish to manage and click the <guilabel moreinfo="none">Properties</guilabel> tab, then click
			the <guilabel moreinfo="none">Permissions</guilabel> tab. Now you can add or change access control settings as you wish.
			</para></step>
			</procedure>

			</sect3>

			<sect3>
			<title>Windows 200x/XP</title>

			<para>
<indexterm significance="normal"><primary>Windows NT4/200x/XP</primary></indexterm>
<indexterm significance="normal"><primary>ACLs on share</primary></indexterm>
<indexterm significance="normal"><primary>Sharing</primary></indexterm>
<indexterm significance="normal"><primary>Permissions</primary></indexterm>
			On <application moreinfo="none">MS Windows NT4/200x/XP</application> systems, ACLs on the share itself are set using
			tools like the MS Explorer. For example, in Windows 200x, right-click on the shared folder,
			then select <guimenuitem moreinfo="none">Sharing</guimenuitem>, then click on <guilabel moreinfo="none">Permissions</guilabel>. The default 
			Windows NT4/200x permissions allow the group "Everyone" full control on the share.
			</para>

			<para>
<indexterm significance="normal"><primary>Computer Management</primary></indexterm>
<indexterm significance="normal"><primary>MMC</primary></indexterm>
<indexterm significance="normal"><primary>tool</primary></indexterm>
			MS Windows 200x and later versions come with a tool called the <application moreinfo="none">Computer Management</application>
			snap-in for the MMC. This tool can be accessed via <guimenu moreinfo="none">Control Panel -&gt;
			Administrative Tools -&gt; Computer Management</guimenu>.
			</para>

			<procedure>
			<title>Instructions</title>
			<step performance="required"><para>
			After launching the MMC with the Computer Management snap-in, click the menu item <guimenuitem moreinfo="none">Action</guimenuitem>
			and select <guilabel moreinfo="none">Connect to another computer</guilabel>. If you are not logged onto a domain you will be prompted
			to enter a domain login user identifier and a password. This will authenticate you to the domain.
			If you are already logged in with administrative privilege, this step is not offered.
			</para></step>

			<step performance="required"><para>
			If the Samba server is not shown in the <guilabel moreinfo="none">Select Computer</guilabel> box, type in the name of the target
			Samba server in the field <guilabel moreinfo="none">Name:</guilabel>. Now click the on <guibutton moreinfo="none">[+]</guibutton> next to 
			<guilabel moreinfo="none">System Tools</guilabel>, then on the <guibutton moreinfo="none">[+]</guibutton> next to
			<guilabel moreinfo="none">Shared Folders</guilabel> in the left panel.
			</para></step>

			<step performance="required"><para>
<indexterm significance="normal"><primary>Share Permissions</primary></indexterm>
			In the right panel, double-click on the share on which you wish to set access control permissions.
			Then click the tab <guilabel moreinfo="none">Share Permissions</guilabel>. It is now possible to add access control entities
			to the shared folder. Remember to set what type of access (full control, change, read) you
			wish to assign for each entry.
			</para></step>
			</procedure>

			<warning>
			<para>
			Be careful. If you take away all permissions from the <constant>Everyone</constant> user without removing
			this user, effectively no user will be able to access the share. This is a result of what is known as
			ACL precedence. Everyone with <emphasis>no access</emphasis> means that <constant>MaryK</constant> who is
			part of the group <constant>Everyone</constant> will have no access even if she is given explicit full
			control access.
			</para>
			</warning>

			</sect3>
		</sect2>

</sect1>

<sect1>
<title>MS Windows Access Control Lists and UNIX Interoperability</title>

	<sect2>
		<title>Managing UNIX Permissions Using NT Security Dialogs</title>


		<para>
		<indexterm significance="normal"><primary>permissions</primary><secondary>file/directory ACLs</secondary></indexterm>
		Windows NT clients can use their native security settings dialog box to view and modify the
		underlying UNIX permissions.
		</para>

		<para>
		This ability is careful not to compromise the security of the UNIX host on which Samba is running and 
		still obeys all the file permission rules that a Samba administrator can set.
		</para>

		<para>
		Samba does not attempt to go beyond POSIX ACLs, so the various finer-grained access control
		options provided in Windows are actually ignored.
		</para> 

		<note>
		<para>
		All access to UNIX/Linux system files via Samba is controlled by the operating system file access controls.
		When trying to figure out file access problems, it is vitally important to find the identity of the Windows
		user as it is presented by Samba at the point of file access. This can best be determined from the
		Samba log files.
		</para>
		</note>
	</sect2>

	<sect2>
		<title>Viewing File Security on a Samba Share</title>

		<para>
		From an NT4/2000/XP client, right-click on any file or directory in a Samba-mounted drive letter
		or UNC path. When the menu pops up, click on the <guilabel moreinfo="none">Properties</guilabel> entry at the bottom
		of the menu. This brings up the file <constant>Properties</constant> dialog box. Click on the 
		<guilabel moreinfo="none">Security</guilabel> tab and you will see three buttons: <guibutton moreinfo="none">Permissions</guibutton>,
		<guibutton moreinfo="none">Auditing</guibutton>, and <guibutton moreinfo="none">Ownership</guibutton>. The <guibutton moreinfo="none">Auditing</guibutton>
		button will cause either an error message <errorname>"A requested privilege is not held by the client"</errorname>
		to appear if the user is not the NT administrator, or a dialog intended to allow an administrator
		to add auditing requirements to a file if the user is logged on as the NT administrator. This dialog is
		nonfunctional with a Samba share at this time, because the only useful button, the <guibutton moreinfo="none">Add</guibutton>
		button, will not currently allow a list of users to be seen.
		</para>

	</sect2>

	<sect2>
		<title>Viewing File Ownership</title>

		<para>
		Clicking on the <guibutton moreinfo="none">Ownership</guibutton> button brings up a dialog box telling you who owns
		the given file. The owner name will be displayed like this:
		<screen format="linespecific">
		<constant>SERVER\user (Long name)</constant>
		</screen>
		<replaceable>SERVER</replaceable> is the NetBIOS name of the Samba server, <replaceable>user</replaceable>
		is the username of the UNIX user who owns the file, and <replaceable>(Long name)</replaceable> is the
		descriptive string identifying the user (normally found in the GECOS field of the UNIX password database).
		Click on the <guibutton moreinfo="none">Close</guibutton> button to remove this dialog.
		</para>

		<para>
		If the parameter <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="NTACLSUPPORT" xlink:href="smb.conf.5.html#NTACLSUPPORT">nt acl support</link> is set to <constant>false</constant>,
		the file owner will be shown as the NT user <emphasis>Everyone</emphasis>.
		</para>

		<para>
<indexterm significance="normal"><primary>Take Ownership</primary></indexterm>
		The <guibutton moreinfo="none">Take Ownership</guibutton> button will not allow you to change the ownership of this file to
		yourself (clicking it will display a dialog box complaining that the user as whom you are currently logged onto
		the NT client cannot be found). The reason for this is that changing the ownership of a file is a privileged
		operation in UNIX, available only to the <emphasis>root</emphasis> user. Because clicking on this button causes
		NT to attempt to change the ownership of a file to the current user logged into the NT client, this will
		not work with Samba at this time.
		</para>

		<para>
<indexterm significance="normal"><primary>chown</primary></indexterm>
<indexterm significance="normal"><primary>ownership</primary></indexterm>
<indexterm significance="normal"><primary>Seclib</primary></indexterm>
		There is an NT <literal>chown</literal> command that will work with Samba and allow a user with administrator
		privilege connected to a Samba server as root to change the ownership of files on both a local NTFS file system
		or remote mounted NTFS or Samba drive. This is available as part of the <application moreinfo="none">Seclib</application> NT
		security library written by Jeremy Allison of the Samba Team and is downloadable from the main Samba FTP site.
		</para>

	</sect2>

	<sect2>
		<title>Viewing File or Directory Permissions</title>

		<para>
		The third button is the <guibutton moreinfo="none">Permissions</guibutton> button. Clicking on it brings up a dialog box
		that shows both the permissions and the UNIX owner of the file or directory. The owner is displayed like this:
		</para>

		<para><literal><replaceable>SERVER</replaceable>\
				<replaceable>user</replaceable> 
				<replaceable>(Long name)</replaceable></literal></para>

		<para><replaceable>SERVER</replaceable> is the NetBIOS name of the Samba server,
		<replaceable>user</replaceable> is the username of the UNIX user who owns the file, and
		<replaceable>(Long name)</replaceable> is the descriptive string identifying the user (normally found in the
		GECOS field of the UNIX password database).</para>

		<para>
		If the parameter <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="NTACLSUPPORT" xlink:href="smb.conf.5.html#NTACLSUPPORT">nt acl support</link> is set to <constant>false</constant>,
		the file owner will be shown as the NT user <constant>Everyone</constant>, and the permissions will be
		shown as NT <emphasis>Full Control</emphasis>.
		</para>


		<para>
		The permissions field is displayed differently for files and directories. Both are discussed next.
		</para>

		<sect3>
		<title>File Permissions</title>

		<para>
		The standard UNIX user/group/world triplet and the corresponding <constant>read, write,
		execute</constant> permissions triplets are mapped by Samba into a three-element NT ACL with the
		<quote>r</quote>, <quote>w</quote>, and <quote>x</quote> bits mapped into the corresponding NT
		permissions. The UNIX world permissions are mapped into the global NT group <constant>Everyone</constant>, followed 
		by the list of permissions allowed for the UNIX world. The UNIX owner and group permissions are displayed as an NT 
		<guiicon moreinfo="none">user</guiicon> icon and an NT <guiicon moreinfo="none">local group</guiicon> icon, respectively, followed by the list 
		of permissions allowed for the UNIX user and group.
		</para>

		<para>
		Because many UNIX permission sets do not map into common NT names such as <constant>read</constant>,
		<constant>change</constant>, or <constant>full control</constant>, usually the permissions will be prefixed
		by the words <constant>Special Access</constant> in the NT display list.
		</para>

		<para>
		But what happens if the file has no permissions allowed for a particular UNIX user group or world component?
		In order to  allow <emphasis>no permissions</emphasis> to be seen and modified, Samba then overloads the NT
		<constant>Take Ownership</constant> ACL attribute (which has no meaning in UNIX) and reports a component with
		no permissions as having the NT <literal>O</literal> bit set.  This was chosen, of course, to make it look
		like a zero, meaning zero permissions. More details on the decision behind this action are given below.
		</para>

		</sect3>
		
		<sect3>
		<title>Directory Permissions</title>

		<para>
		Directories on an NT NTFS file system have two different sets of permissions. The first set is the ACL set on the
		directory itself, which is usually displayed in the first set of parentheses in the normal <constant>RW</constant> 
		NT style. This first set of permissions is created by Samba in exactly the same way as normal file permissions are, described 
		above, and is displayed in the same way.
		</para>

		<para>
		The second set of directory permissions has no real meaning in the UNIX permissions world and represents the <constant>
		inherited</constant> permissions that any file created within this directory would inherit.
		</para>

		<para>
		Samba synthesizes these inherited permissions for NT by returning as an NT ACL the UNIX permission mode that a new file 
		created by Samba on this share would receive.
		</para>

		</sect3>

	</sect2>
		
	<sect2>
	<title>Modifying File or Directory Permissions</title>

	<para>
	Modifying file and directory permissions is as simple as changing the displayed permissions in the dialog box
	and clicking on <guibutton moreinfo="none">OK</guibutton>. However, there are limitations that a user needs to be aware of,
	and also interactions with the standard Samba permission masks and mapping of DOS attributes that also need to
	be taken into account.
	</para>

	<para>
	If the parameter <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="NTACLSUPPORT" xlink:href="smb.conf.5.html#NTACLSUPPORT">nt acl support</link> is set to <constant>false</constant>, any attempt to
	set security permissions will fail with an <errorname>"Access Denied" </errorname> message.
	</para>

	<para>
	The first thing to note is that the <guibutton moreinfo="none">Add</guibutton> button will not return a list of users in Samba
	(it will give an error message saying <errorname>"The remote procedure call failed and did not
	execute"</errorname>). This means that you can only manipulate the current user/group/world permissions listed
	in the dialog box. This actually works quite well because these are the only permissions that UNIX actually
	has.
	</para>

	<para>
	If a permission triplet (either user, group, or world) is removed from the list of permissions in the NT
	dialog box, then when the <guibutton moreinfo="none">OK</guibutton> button is pressed, it will be applied as <emphasis>no
	permissions</emphasis> on the UNIX side. If you view the permissions again, the <emphasis>no
	permissions</emphasis> entry will appear as the NT <literal>O</literal> flag, as described above. This allows
	you to add permissions back to a file or directory once you have removed them from a triplet component.
	</para>

	<para>
	Because UNIX supports only the <quote>r</quote>, <quote>w</quote>, and <quote>x</quote> bits of an NT ACL, if
	other NT security attributes such as <constant>Delete Access</constant> are selected, they will be ignored
	when applied on the Samba server.
	</para>

	<para>
	When setting permissions on a directory, the second set of permissions (in the second set of parentheses) is
	by default applied to all files within that directory. If this is not what you want, you must uncheck the
	<guilabel moreinfo="none">Replace permissions on existing files</guilabel> checkbox in the NT dialog before clicking on
	<guibutton moreinfo="none">OK</guibutton>.
	</para>

	<para>
	If you wish to remove all permissions from a user/group/world  component, you may either highlight the
	component and click on the <guibutton moreinfo="none">Remove</guibutton> button or set the component to only have the special
	<constant>Take Ownership</constant> permission (displayed as <literal>O</literal>) highlighted.
	</para>

	</sect2>

<?latex \newpage ?>
	<sect2>
	<title>Interaction with the Standard Samba <quote>create mask</quote> Parameters</title>

	<para>There are four parameters that control interaction with the standard Samba <parameter moreinfo="none">create mask</parameter> parameters:
	

	<itemizedlist>
		<listitem><para><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SECURITYMASK" xlink:href="smb.conf.5.html#SECURITYMASK">security mask</link></para></listitem>
		<listitem><para><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="FORCESECURITYMODE" xlink:href="smb.conf.5.html#FORCESECURITYMODE">force security mode</link></para></listitem>
		<listitem><para><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="DIRECTORYSECURITYMASK" xlink:href="smb.conf.5.html#DIRECTORYSECURITYMASK">directory security mask</link></para></listitem>
		<listitem><para><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="FORCEDIRECTORYSECURITYMODE" xlink:href="smb.conf.5.html#FORCEDIRECTORYSECURITYMODE">force directory security mode</link></para></listitem>
	</itemizedlist>

	</para>

	<para>
	When a user clicks on <guibutton moreinfo="none">OK</guibutton> to apply the 
	permissions, Samba maps the given permissions into a user/group/world 
	r/w/x triplet set, and then checks the changed permissions for a 
	file against the bits set in the  
	<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SECURITYMASK" xlink:href="smb.conf.5.html#SECURITYMASK">security mask</link> parameter. Any bits that 
	were changed that are not set to <emphasis>1</emphasis> in this parameter are left alone 
	in the file permissions.</para>

	<para>
	Essentially, zero bits in the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SECURITYMASK" xlink:href="smb.conf.5.html#SECURITYMASK">security mask</link>
	may be treated as a set of bits the user is <emphasis>not</emphasis> 
	allowed to change, and one bits are those the user is allowed to change.
	</para>

	<para>
	If not explicitly set, this parameter defaults to the same value as 
	the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="CREATEMASK" xlink:href="smb.conf.5.html#CREATEMASK">create mask</link> parameter. To allow a user to modify all the
	user/group/world permissions on a file, set this parameter to 0777.
	</para>

	<para>
	Next Samba checks the changed permissions for a file against the bits set in the 
	<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="FORCESECURITYMODE" xlink:href="smb.conf.5.html#FORCESECURITYMODE">force security mode</link> parameter. Any bits 
	that were changed that correspond to bits set to <emphasis>1</emphasis> in this parameter 
	are forced to be set.</para>

	<para>
	Essentially, bits set in the <parameter moreinfo="none">force security mode</parameter> parameter
	may be treated as a set of bits that, when modifying security on a file, the user 
	has always set to be <emphasis>on</emphasis>.</para>

	<para>
	If not explicitly set, this parameter defaults to the same value 
	as the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="FORCECREATEMODE" xlink:href="smb.conf.5.html#FORCECREATEMODE">force create mode</link> parameter.
	To allow a user to modify all the user/group/world permissions on a file
	with no restrictions, set this parameter to 000. The
	<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SECURITYMASK" xlink:href="smb.conf.5.html#SECURITYMASK">security mask</link> and <parameter moreinfo="none">force 
	security mode</parameter> parameters are applied to the change 
	request in that order.</para>

	<para>
	For a directory, Samba performs the same operations as 
	described above for a file except it uses the parameter <parameter moreinfo="none">
	directory security mask</parameter> instead of <parameter moreinfo="none">security 
	mask</parameter>, and <parameter moreinfo="none">force directory security mode
	</parameter> parameter instead of <parameter moreinfo="none">force security mode
	</parameter>.</para>

	<para>
	The <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="DIRECTORYSECURITYMASK" xlink:href="smb.conf.5.html#DIRECTORYSECURITYMASK">directory security mask</link> parameter 
	by default is set to the same value as the <parameter moreinfo="none">directory mask
	</parameter> parameter and the <parameter moreinfo="none">force directory security 
	mode</parameter> parameter by default is set to the same value as 
	the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="FORCEDIRECTORYMODE" xlink:href="smb.conf.5.html#FORCEDIRECTORYMODE">force directory mode</link> parameter.
	In this way Samba enforces the permission restrictions that 
	an administrator can set on a Samba share, while still allowing users 
	to modify the permission bits within that restriction.</para>

	<para>
	If you want to set up a share that allows users full control
	in modifying the permission bits on their files and directories and
	does not force any particular bits to be set <emphasis>on</emphasis>,
	then set the following parameters in the <filename moreinfo="none">smb.conf</filename> file in that
	share-specific section:
	</para>

<?latex \newpage ?>
	<simplelist>
		<member><indexterm>security mask</indexterm><parameter>security mask = 0777</parameter></member>
		<member><indexterm>force security mode</indexterm><parameter>force security mode = 0</parameter></member>
		<member><indexterm>directory security mask</indexterm><parameter>directory security mask = 0777</parameter></member>
		<member><indexterm>force directory security mode</indexterm><parameter>force directory security mode = 0</parameter></member>
	</simplelist>

</sect2>

<sect2>
	<title>Interaction with the Standard Samba File Attribute Mapping</title>

	<note>
	<para>
	Samba maps some of the DOS attribute bits (such as <quote>read-only</quote>)
	into the UNIX permissions of a file. This means there can 
	be a conflict between the permission bits set via the security 
	dialog and the permission bits set by the file attribute mapping.
	</para>
	</note>

	<para>
	If a file has no UNIX read access for the owner, it will show up
	as <quote>read-only</quote> in the standard file attributes tabbed dialog.
	Unfortunately, this dialog is the same one that contains the security information
	in another tab.
	</para>

	<para>
	What this can mean is that if the owner changes the permissions
	to allow himself or herself read access using the security dialog, clicks on
	<guibutton moreinfo="none">OK</guibutton> to get back to the standard attributes tab 
	dialog, and clicks on <guibutton moreinfo="none">OK</guibutton> on that dialog, then 
	NT will set the file permissions back to read-only (as that is what 
	the attributes still say in the dialog). This means that after setting 
	permissions and clicking on <guibutton moreinfo="none">OK</guibutton> to get back to the 
	attributes dialog, you should always press <guibutton moreinfo="none">Cancel</guibutton> 
	rather than <guibutton moreinfo="none">OK</guibutton> to ensure that your changes 
	are not overridden.
	</para>

	</sect2>

	<sect2>
	<title>Windows NT/200X ACLs and POSIX ACLs Limitations</title>

	<para>
	Windows administrators are familiar with simple ACL controls, and they typically
	consider that UNIX user/group/other (ugo) permissions are inadequate and not
	sufficiently fine-grained.
	</para>

	<para>
	Competing SMB implementations differ in how they handle Windows ACLs. Samba handles
	Windows ACLs from the perspective of UNIX file system administration and thus adopts
	the limitations of POSIX ACLs. Therefore, where POSIX ACLs lack a capability of the
	Windows NT/200X ACLs, the POSIX semantics and limitations are imposed on the Windows
	administrator.
	</para>

	<para>
	POSIX ACLs present an interesting challenge to the UNIX administrator and therefore
	force a compromise to be applied to Windows ACLs administration. POSIX ACLs are not
	covered by an official standard; rather, the latest standard is a draft standard
	1003.1e revision 17. This is the POSIX document on which the Samba implementation has
	been implemented.
	</para>

	<para>
	UNIX vendors differ in the manner in which POSIX ACLs are implemented. There are a
	number of Linux file systems that support ACLs. Samba has to provide a way to make
	transparent all the differences between the various implementations of POSIX ACLs.
	The pressure for ACLs support in Samba has noticeably increased the pressure to
	standardize ACLs support in the UNIX world.
	</para>

	<para>
	Samba has to deal with the complicated matter of handling the challenge of the Windows
	ACL that implements <emphasis>inheritance</emphasis>, a concept not anticipated by POSIX
	ACLs as implemented in UNIX file systems. Samba provides support for <emphasis>masks</emphasis>
	that permit normal ugo and ACLs functionality to be overrided. This further complicates
	the way in which Windows ACLs must be implemented.
	</para>

	<sect3>
	<title>UNIX POSIX ACL Overview</title>

	<para>
	In examining POSIX ACLs we must consider the manner in which they operate for 
	both files and directories. File ACLs have the following significance:
<screen format="linespecific">
# file: testfile      &lt;- the file name
# owner: jeremy       &lt;-- the file owner
# group: users        &lt;-- the POSIX group owner
user::rwx             &lt;-- perms for the file owner (user)
user:tpot:r-x         &lt;-- perms for the additional user `tpot'
group::r--            &lt;-- perms for the file group owner (group)
group:engrs:r--       &lt;-- perms for the additonal group `engineers'
mask:rwx              &lt;-- the mask that is `ANDed' with groups
other::---            &lt;-- perms applied to everyone else (other)
</screen>
	Directory ACLs have the following signficance:
<screen format="linespecific">
# file: testdir       &lt;-- the directory name
# owner: jeremy       &lt;-- the directory owner
# group: jeremy       &lt;-- the POSIX group owner
user::rwx             &lt;-- directory perms for owner (user)
group::rwx            &lt;-- directory perms for owning group (group)
mask::rwx             &lt;-- the mask that is `ANDed' with group perms
other:r-x             &lt;-- perms applied to everyone else (other)
default:user::rwx     &lt;-- inherited owner perms
default:user:tpot:rwx &lt;-- inherited extra perms for user `tpot'
default:group::r-x    &lt;-- inherited group perms
default:mask:rwx      &lt;-- inherited default mask
default:other:---     &lt;-- inherited permissions for everyone (other)
</screen>
	</para>

	</sect3>

	<sect3>
	<title>Mapping of Windows File ACLs to UNIX POSIX ACLs</title>

	<para>
	Microsoft Windows NT4/200X ACLs must of necessity be mapped to POSIX ACLs.
	The mappings for file permissions are shown in <link linkend="fdsacls">How
	Windows File ACLs Map to UNIX POSIX File ACLs</link>.
	The # character means this flag is set only when the Windows administrator
	sets the <constant>Full Control</constant> flag on the file.
	</para>

	<table frame="all" pgwide="0" id="fdsacls"><title>How Windows File ACLs Map to UNIX POSIX File ACLs</title>
	<tgroup cols="2">
		<colspec align="left"/>
		<colspec align="center"/>
		<thead>
		<row>
			<entry align="left">Windows ACE</entry>
			<entry align="center">File Attribute Flag</entry>
		</row>
		</thead>
		<tbody>
		<row>
			<entry><para>Full Control</para></entry>
			<entry><para>#</para></entry>
		</row>
		<row>
			<entry><para>Traverse Folder/Execute File</para></entry>
			<entry><para>x</para></entry>
		</row>
		<row>
			<entry><para>List Folder/Read Data</para></entry>
			<entry><para>r</para></entry>
		</row>
		<row>
			<entry><para>Read Attributes</para></entry>
			<entry><para>r</para></entry>
		</row>
		<row>
			<entry><para>Read Extended Attribures</para></entry>
			<entry><para>r</para></entry>
		</row>
		<row>
			<entry><para>Create Files/Write Data</para></entry>
			<entry><para>w</para></entry>
		</row>
		<row>
			<entry><para>Create Folders/Append Data</para></entry>
			<entry><para>w</para></entry>
		</row>
		<row>
			<entry><para>Write Attributes</para></entry>
			<entry><para>w</para></entry>
		</row>
		<row>
			<entry><para>Write Extended Attributes</para></entry>
			<entry><para>w</para></entry>
		</row>
		<row>
			<entry><para>Delete Subfolders and Files</para></entry>
			<entry><para>w</para></entry>
		</row>
		<row>
			<entry><para>Delete</para></entry>
			<entry><para>#</para></entry>
		</row>
		<row>
			<entry><para>Read Permissions</para></entry>
			<entry><para>all</para></entry>
		</row>
		<row>
			<entry><para>Change Permissions</para></entry>
			<entry><para>#</para></entry>
		</row>
		<row>
			<entry><para>Take Ownership</para></entry>
			<entry><para>#</para></entry>
		</row>
		</tbody>
	</tgroup>
	</table>

	<para>
	As can be seen from the mapping table, there is no one-to-one mapping capability, and therefore
	Samba must make a logical mapping that will permit Windows to operate more-or-less the way
	that is intended by the administrator.
	</para>

	<para>
	In general the mapping of UNIX POSIX user/group/other permissions will be mapped to
	Windows ACLs. This has precedence over the creation of POSIX ACLs. POSIX ACLs are necessary
	to establish access controls for users and groups other than the user and group that
	own the file or directory.
	</para>

	<para>
	The UNIX administrator can set any directory permission from within the UNIX environment.
	The Windows administrator is more restricted in that it is not possible from within 
	Windows Explorer to remove read permission for the file owner.
	</para>

	</sect3>

	<sect3>
	<title>Mapping of Windows Directory ACLs to UNIX POSIX ACLs</title>

	<para>
	Interesting things happen in the mapping of UNIX POSIX directory permissions and
	UNIX POSIX ACLs to Windows ACEs (Access Control Entries, the discrete components of
	an ACL) are mapped to Windows directory ACLs.
	</para>

	<para>
	Directory permissions function in much the same way as shown for file permissions, but
	there are some notable exceptions and a few peculiarities that the astute administrator
	will want to take into account in the setting up of directory permissions.
	</para>

	</sect3>

	</sect2>
</sect1>

<sect1>
<title>Common Errors</title>

<para>
File, directory, and share access problems are common topics on the mailing list. The following
are examples recently taken from the mailing list.
</para>


	<sect2>
	<title>Users Cannot Write to a Public Share</title>

	<para>
	The following complaint has frequently been voiced on the Samba mailing list: 
	<quote>
	We are facing some troubles with file/directory permissions. I can log on the domain as admin user (root),
	and there's a public share on which everyone needs to have permission to create/modify files, but only
	root can change the file, no one else can. We need to constantly go to the server to
	<userinput moreinfo="none">chgrp -R users *</userinput> and <userinput moreinfo="none">chown -R nobody *</userinput> to allow
	other users to change the file.
	</quote>
	</para>

	<para>
	Here is one way the problem can be solved:
	</para>

	<procedure>
		<step performance="required">
			<para>
			Go to the top of the directory that is shared.
			</para>
		</step>

		<step performance="required">
			<para>
			Set the ownership to whatever public user and group you want
<screen format="linespecific">
<prompt moreinfo="none">$ </prompt>find `directory_name' -type d -exec chown user:group {}\;
<prompt moreinfo="none">$ </prompt>find `directory_name' -type d -exec chmod 2775 {}\;
<prompt moreinfo="none">$ </prompt>find `directory_name' -type f -exec chmod 0775 {}\;
<prompt moreinfo="none">$ </prompt>find `directory_name' -type f -exec chown user:group {}\;
</screen>
			</para>

			<note><para>
			The above will set the <constant>SGID bit</constant> on all directories. Read your
			UNIX/Linux man page on what that does. This ensures that all files and directories
			that are created in the directory tree will be owned by the current user and will
			be owned by the group that owns the directory in which it is created.
			</para></note>
		</step>
		<step performance="required">
			<para>
			Directory is <replaceable>/foodbar</replaceable>:
<screen format="linespecific">
<prompt moreinfo="none">$ </prompt><userinput moreinfo="none">chown jack:engr /foodbar</userinput>
</screen>
			</para>

			<note>
			<para>This is the same as doing:</para>
<screen format="linespecific">
<prompt moreinfo="none">$ </prompt><userinput moreinfo="none">chown jack /foodbar</userinput>
<prompt moreinfo="none">$ </prompt><userinput moreinfo="none">chgrp engr /foodbar</userinput>
</screen>
			</note>
		</step>
		<step performance="required">
			<para>Now type: 

<screen format="linespecific">
<prompt moreinfo="none">$ </prompt><userinput moreinfo="none">chmod 2775 /foodbar</userinput>
<prompt moreinfo="none">$ </prompt><userinput moreinfo="none">ls -al /foodbar/..</userinput>
</screen>
			</para>
		
			<para>You should see:
<screen format="linespecific">
drwxrwsr-x  2 jack  engr    48 2003-02-04 09:55 foodbar
</screen>
			</para>
		</step>
		<step performance="required">

		<para>Now type:
<screen format="linespecific">
<prompt moreinfo="none">$ </prompt><userinput moreinfo="none">su - jill</userinput>
<prompt moreinfo="none">$ </prompt><userinput moreinfo="none">cd /foodbar</userinput>
<prompt moreinfo="none">$ </prompt><userinput moreinfo="none">touch Afile</userinput>
<prompt moreinfo="none">$ </prompt><userinput moreinfo="none">ls -al</userinput>
</screen>
		</para>

		<para>
		You should see that the file <filename moreinfo="none">Afile</filename> created by Jill will have ownership
		and permissions of Jack, as follows:
<screen format="linespecific">
-rw-r--r--  1 jill  engr     0 2007-01-18 19:41 Afile
</screen>
		</para>
		</step>

		<step performance="required">
		<para>
		If the user that must have write permission in the directory is not a member of the group
		<emphasis>engr</emphasis> set in the <filename moreinfo="none">smb.conf</filename> entry for the share:
		<simplelist>
<member><indexterm>force group</indexterm><parameter>force group = engr</parameter></member>
		</simplelist>
		</para>
	</step>
	</procedure>
	</sect2>


	<sect2>
		<title>File Operations Done as <emphasis>root</emphasis> with <emphasis>force user</emphasis> Set</title>

		<para>
		When you have a user in <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="ADMINUSERS" xlink:href="smb.conf.5.html#ADMINUSERS">admin users</link>, Samba will always do file operations for
		this user as <emphasis>root</emphasis>, even if <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="FORCEUSER" xlink:href="smb.conf.5.html#FORCEUSER">force user</link> has been set.
		</para>
	</sect2>
	
	<sect2>
		<title>MS Word with Samba Changes Owner of File</title>

		<para>
		<emphasis>Question:</emphasis> <quote>When user B saves a word document that is owned by user A,
		the updated file is now owned by user B.  Why is Samba doing this? How do I fix this?</quote>
		</para>

		<para>
		<emphasis>Answer:</emphasis> Word does the following when you modify/change a Word document: MS Word creates a new document with
		a temporary name. Word then closes the old document and deletes it, then renames the new document to the original document name.
		There is no mechanism by which Samba can in any way know that the new document really should be owned by the owners
		of the original file. Samba has no way of knowing that the file will be renamed by MS Word. As far as Samba is able
		to tell, the file that gets created is a new file, not one that the application (Word) is updating.
		</para>

		<para>
		There is a workaround to solve the permissions problem. It involves understanding how you can manage file
		system behavior from within the <filename moreinfo="none">smb.conf</filename> file, as well as understanding how UNIX file systems work. Set on the directory
		in which you are changing Word documents: <literal>chmod g+s `directory_name'.</literal> This ensures that all files will
		be created with the group that owns the directory. In <filename moreinfo="none">smb.conf</filename> share declaration section set:
		</para>

		<para>
		<simplelist>
                <member><indexterm>force create mode</indexterm><parameter>force create mode = 0660</parameter></member>
                <member><indexterm>force directory mode</indexterm><parameter>force directory mode = 0770</parameter></member>
		</simplelist>
		</para>

		<para>
		These two settings will ensure that all directories and files that get created in the share will be readable/writable by the
		owner and group set on the directory itself.
		</para>
		
	</sect2>

</sect1>

</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="locking">
<chapterinfo>
	
<author>
	<firstname>Jeremy</firstname><surname>Allison</surname>
	<affiliation>
		<orgname>Samba Team</orgname>
		<address format="linespecific"><email>jra@samba.org</email></address>
	</affiliation>
</author>
	<author>
<firstname>Jelmer</firstname><othername>R.</othername><surname>Vernooij</surname><othername>R.</othername>
<affiliation>
	<orgname>The Samba Team</orgname>
	<address format="linespecific"><email>jelmer@samba.org</email></address>
</affiliation></author>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
	
<author>
	<firstname>Eric</firstname><surname>Roseme</surname>
	<affiliation>
		<orgname>HP Oplocks Usage Recommendations Whitepaper</orgname>
		<address format="linespecific"><email>eric.roseme@hp.com</email></address>
	</affiliation>
</author>
</chapterinfo>
<title>File and Record Locking</title>

<para>
<indexterm significance="normal"><primary>locking</primary></indexterm>
One area that causes trouble for many network administrators is locking.
The extent of the problem is readily evident from searches over the Internet.
</para>

<sect1>
<title>Features and Benefits</title>

<para>
<indexterm significance="normal"><primary>locking semantics</primary></indexterm>
Samba provides all the same locking semantics that MS Windows clients expect
and that MS Windows NT4/200x servers also provide.
</para>

<para>
<indexterm significance="normal"><primary>locking</primary></indexterm>
The term <emphasis>locking</emphasis> has exceptionally broad meaning and covers
a range of functions that are all categorized under this one term.
</para>

<para>
<indexterm significance="normal"><primary>opportunistic locking</primary></indexterm>
<indexterm significance="normal"><primary>locking protocol</primary></indexterm>
<indexterm significance="normal"><primary>performance advantage</primary></indexterm>
Opportunistic locking is a desirable feature when it can enhance the
perceived performance of applications on a networked client. However, the
opportunistic locking protocol is not robust and therefore can
encounter problems when invoked beyond a simplistic configuration or
on extended slow or faulty networks. In these cases, operating
system management of opportunistic locking and/or recovering from
repetitive errors can offset the perceived performance advantage that
it is intended to provide.
</para>

<para>
<indexterm significance="normal"><primary>registry</primary></indexterm>
The MS Windows network administrator needs to be aware that file and record
locking semantics (behavior) can be controlled either in Samba or by way of registry
settings on the MS Windows client.
</para>

<note>
<para>
<indexterm significance="normal"><primary>disable locking</primary></indexterm>
Sometimes it is necessary to disable locking control settings on the Samba
server as well as on each MS Windows client!
</para>
</note>

</sect1>

<sect1>
<title>Discussion</title>

<para>
<indexterm significance="normal"><primary>record locking</primary></indexterm>
<indexterm significance="normal"><primary>deny modes</primary></indexterm>
There are two types of locking that need to be performed by an SMB server.
The first is <emphasis>record locking</emphasis> that allows a client to lock
a range of bytes in an open file. The second is the <emphasis>deny modes</emphasis>
that are specified when a file is open.
</para>

<para>
<indexterm significance="normal"><primary>locking semantics</primary></indexterm>
<indexterm significance="normal"><primary>record locking</primary></indexterm>
<indexterm significance="normal"><primary>locking</primary></indexterm>
<indexterm significance="normal"><primary>byte ranges</primary></indexterm>
<indexterm significance="normal"><primary>UNIX locking</primary></indexterm>
Record locking semantics under UNIX are very different from record locking under
Windows. Versions of Samba before 2.2 have tried to use the native fcntl() UNIX
system call to implement proper record locking between different Samba clients.
This cannot be fully correct for several reasons. The simplest is
that a Windows client is allowed to lock a byte range up to 2^32 or 2^64,
depending on the client OS. The UNIX locking only supports byte ranges up to 2^31.
So it is not possible to correctly satisfy a lock request above 2^31. There are
many more differences, too many to be listed here.
</para>

<para>
<indexterm significance="normal"><primary>record locking</primary></indexterm>
<indexterm significance="normal"><primary>byte-range lock</primary></indexterm>
Samba 2.2 and above implement record locking completely independently of the
underlying UNIX system. If a byte-range lock that the client requests happens
to fall into the range of 0 to 2^31, Samba hands this request down to the UNIX system.
No other locks can be seen by UNIX, anyway.
</para>

<para>
<indexterm significance="normal"><primary>check for locks</primary></indexterm>
<indexterm significance="normal"><primary>rpc.lockd</primary></indexterm>
Strictly speaking, an SMB server should check for locks before every read and write call on
a file. Unfortunately, with the way fcntl() works, this can be slow and may overstress
the <literal>rpc.lockd</literal>. This is almost always unnecessary because clients are 
independently supposed to make locking calls before reads and writes if locking is
important to them. By default, Samba only makes locking calls when explicitly asked
to by a client, but if you set <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="STRICTLOCKING" xlink:href="smb.conf.5.html#STRICTLOCKING">strict locking = yes</link>, it
will make lock checking calls on <emphasis>every</emphasis> read and write call.
</para>

<para>
<indexterm significance="normal"><primary>byte-range locking</primary></indexterm>
You can also disable byte-range locking completely by using
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOCKING" xlink:href="smb.conf.5.html#LOCKING">locking = no</link>.
This is useful for those shares that do not support locking or do not need it
(such as CD-ROMs). In this case, Samba fakes the return codes of locking calls to
tell clients that everything is okay.
</para>

<para>
<indexterm significance="normal"><primary>deny modes</primary></indexterm>
<indexterm significance="normal"><primary>DENY_NONE</primary></indexterm>
<indexterm significance="normal"><primary>DENY_READ</primary></indexterm>
<indexterm significance="normal"><primary>DENY_WRITE</primary></indexterm>
<indexterm significance="normal"><primary>DENY_ALL</primary></indexterm>
<indexterm significance="normal"><primary>DENY_FCB</primary></indexterm>
<indexterm significance="normal"><primary>DENY_DOS</primary></indexterm>
The second class of locking is the <emphasis>deny modes</emphasis>. These 
are set by an application when it opens a file to determine what types of
access should be allowed simultaneously with its open. A client may ask for
<constant>DENY_NONE</constant>, <constant>DENY_READ</constant>, 
<constant>DENY_WRITE</constant>, or <constant>DENY_ALL</constant>. There are also special compatibility
modes called <constant>DENY_FCB</constant> and <constant>DENY_DOS</constant>.
</para>

<sect2>
<title>Opportunistic Locking Overview</title>

<para>
<indexterm significance="normal"><primary>opportunistic locking</primary></indexterm>
<indexterm significance="normal"><primary>oplocks</primary></indexterm>
<indexterm significance="normal"><primary>caching</primary></indexterm>
Opportunistic locking (oplocks) is invoked by the Windows file system
(as opposed to an API) via registry entries (on the server and the client)
for the purpose of enhancing network performance when accessing a file
residing on a server. Performance is enhanced by caching the file
locally on the client that allows the following:
</para>

<variablelist>
	<varlistentry><term>Read-ahead:</term>
		<listitem><para>
<indexterm significance="normal"><primary>Read-ahead</primary></indexterm>
		The client reads the local copy of the file, eliminating network latency.
		</para></listitem>
	</varlistentry>

	<varlistentry><term>Write caching:</term>
		<listitem><para>
<indexterm significance="normal"><primary>Write caching</primary></indexterm>
		The client writes to the local copy of the file, eliminating network latency.
		</para></listitem>
	</varlistentry>

        <varlistentry><term>Lock caching:</term>
        <listitem><para>
<indexterm significance="normal"><primary>Lock caching</primary></indexterm>
		The client caches application locks locally, eliminating network latency.
		</para></listitem>
        </varlistentry>
</variablelist>

<para>
<indexterm significance="normal"><primary>performance enhancement</primary></indexterm>
<indexterm significance="normal"><primary>oplocks</primary></indexterm>
<indexterm significance="normal"><primary>deny-none</primary></indexterm>
The performance enhancement of oplocks is due to the opportunity of
exclusive access to the file <?latex --- ?> even if it is opened with deny-none <?latex --- ?>
because Windows monitors the file's status for concurrent access from
other processes.
</para>

<variablelist>
<title>Windows Defines Four Kinds of Oplocks:</title>

		<varlistentry><term>Level1 Oplock</term>
			<listitem><para>
<indexterm significance="normal"><primary>Level1 Oplock</primary></indexterm>
<indexterm significance="normal"><primary>redirector</primary></indexterm>
<indexterm significance="normal"><primary>concurrent access</primary></indexterm>
<indexterm significance="normal"><primary>cached local file</primary></indexterm>
			The redirector sees that the file was opened with deny
			none (allowing concurrent access), verifies that no
			other process is accessing the file, checks that
			oplocks are enabled, then grants deny-all/read-write/exclusive
			access to the file. The client now performs
			operations on the cached local file.
			</para>

			<para>
<indexterm significance="normal"><primary>oplock break</primary></indexterm>
<indexterm significance="normal"><primary>flush local locks</primary></indexterm>
<indexterm significance="normal"><primary>deferred open</primary></indexterm>
<indexterm significance="normal"><primary>byte-range locking</primary></indexterm>
			If a second process attempts to open the file, the open
			is deferred while the redirector "breaks" the original
			oplock. The oplock break signals the caching client to
			write the local file back to the server, flush the
			local locks, and discard read-ahead data. The break is
			then complete, the deferred open is granted, and the
			multiple processes can enjoy concurrent file access as
			dictated by mandatory or byte-range locking options.
			However, if the original opening process opened the
			file with a share mode other than deny-none, then the
			second process is granted limited or no access, despite
			the oplock break.
			</para></listitem>
        </varlistentry>

        <varlistentry><term>Level2 Oplock</term>
                <listitem><para>
<indexterm significance="normal"><primary>Level2 Oplock</primary></indexterm>
<indexterm significance="normal"><primary>Level1 oplock</primary></indexterm>
<indexterm significance="normal"><primary>caching</primary></indexterm>
				Performs like a Level1 oplock, except caching is only
                operative for reads. All other operations are performed
                on the server disk copy of the file.
                </para></listitem>
        </varlistentry>

        <varlistentry><term>Filter Oplock</term>
                <listitem><para>
<indexterm significance="normal"><primary>Filter Oplock</primary></indexterm>
				Does not allow write or delete file access.
                </para></listitem>
        </varlistentry>

        <varlistentry><term>Batch Oplock</term>
                <listitem><para>
<indexterm significance="normal"><primary>Batch Oplock</primary></indexterm>
				Manipulates file openings and closings and allows caching
                of file attributes.
                </para></listitem>
        </varlistentry>
</variablelist>

<para>
<indexterm significance="normal"><primary>oplocks</primary></indexterm>
An important detail is that oplocks are invoked by the file system, not
an application API. Therefore, an application can close an oplocked
file, but the file system does not relinquish the oplock. When the
oplock break is issued, the file system then simply closes the file in
preparation for the subsequent open by the second process.
</para>

<para>
<indexterm significance="normal"><primary>Opportunistic locking</primary></indexterm>
<indexterm significance="normal"><primary>client-side data caching</primary></indexterm>
<indexterm significance="normal"><primary>data caching</primary></indexterm>
<indexterm significance="normal"><primary>oplock break</primary></indexterm>
<emphasis>Opportunistic locking</emphasis> is actually an improper name for this feature.
The true benefit of this feature is client-side data caching, and
oplocks is merely a notification mechanism for writing data back to the
networked storage disk. The limitation of oplocks is the
reliability of the mechanism to process an oplock break (notification)
between the server and the caching client. If this exchange is faulty
(usually due to timing out for any number of reasons), then the
client-side caching benefit is negated.
</para>

<para>
<indexterm significance="normal"><primary>client-side caching</primary></indexterm>
The actual decision that a user or administrator should consider is
whether it is sensible to share among multiple users data that will
be cached locally on a client. In many cases the answer is no.
Deciding when to cache or not cache data is the real question, and thus
oplocks should be treated as a toggle for client-side
caching. Turn it <quote>on</quote> when client-side caching is desirable and
reliable. Turn it <quote>off</quote> when client-side caching is redundant,
unreliable, or counterproductive.
</para>

<para>
<indexterm significance="normal"><primary>oplocks</primary></indexterm>
Oplocks is by default set to <quote>on</quote> by Samba on all
configured shares, so careful attention should be given to each case to
determine if the potential benefit is worth the potential for delays.
The following recommendations will help to characterize the environment
where oplocks may be effectively configured.
</para>

<para>
<indexterm significance="normal"><primary>oplocks</primary></indexterm>
<indexterm significance="normal"><primary>high-availability</primary></indexterm>
Windows oplocks is a lightweight performance-enhancing
feature. It is not a robust and reliable protocol. Every
implementation of oplocks should be evaluated as a
trade-off between perceived performance and reliability. Reliability
decreases as each successive rule above is not enforced. Consider a
share with oplocks enabled, over a wide-area network, to a client on a
South Pacific atoll, on a high-availability server, serving a
mission-critical multiuser corporate database during a tropical
storm. This configuration will likely encounter problems with oplocks.
</para>

<para>
<indexterm significance="normal"><primary>mission-critical</primary></indexterm>
Oplocks can be beneficial to perceived client performance when treated
as a configuration toggle for client-side data caching. If the data
caching is likely to be interrupted, then oplock usage should be
reviewed. Samba enables oplocks by default on all
shares. Careful attention should be given to the client usage of
shared data on the server, the server network reliability, and the
oplocks configuration of each share.
In mission-critical, high-availability environments, data integrity is
often a priority. Complex and expensive configurations are implemented
to ensure that if a client loses connectivity with a file server, a
failover replacement will be available immediately to provide
continuous data availability.
</para>

<para>
<indexterm significance="normal"><primary>Windows client failover</primary></indexterm>
<indexterm significance="normal"><primary>transport connection loss</primary></indexterm>
Windows client failover behavior is more at risk of application
interruption than other platforms because it is dependent upon an
established TCP transport connection. If the connection is interrupted
<?latex --- ?> as in a file server failover <?latex --- ?> a new session must be established.
It is rare for Windows client applications to be coded to recover
correctly from a transport connection loss; therefore, most applications
will experience some sort of interruption <?latex --- ?> at worst, abort and
require restarting.
</para>

<para>
<indexterm significance="normal"><primary>caching writes</primary></indexterm>
<indexterm significance="normal"><primary>caching reads</primary></indexterm>
<indexterm significance="normal"><primary>oplock break</primary></indexterm>
If a client session has been caching writes and reads locally due to
oplocks, it is likely that the data will be lost when the
application restarts or recovers from the TCP interrupt. When the TCP
connection drops, the client state is lost. When the file server
recovers, an oplock break is not sent to the client. In this case, the
work from the prior session is lost. Observing this scenario with
oplocks disabled and with the client writing data to the file server
real-time,  the failover will provide the data on disk as it
existed at the time of the disconnect.
</para>

<para>
In mission-critical, high-availability environments, careful attention
should be given to oplocks. Ideally, comprehensive
testing should be done with all affected applications with oplocks
enabled and disabled.
</para>

<sect3>
<title>Exclusively Accessed Shares</title>

<para>
Oplocks is most effective when it is confined to shares
that are exclusively accessed by a single user, or by only one user at
a time. Because the true value of oplocks is the local
client caching of data, any operation that interrupts the caching
mechanism will cause a delay.
</para>

<para>
Home directories are the most obvious examples of where the performance
benefit of oplocks can be safely realized.
</para>

</sect3>

<sect3>
<title>Multiple-Accessed Shares or Files</title>

<para>
As each additional user accesses a file in a share with oplocks
enabled, the potential for delays and resulting perceived poor
performance increases. When multiple users are accessing a file on a
share that has oplocks enabled, the management impact of sending and
receiving oplock breaks and the resulting latency while other clients
wait for the caching client to flush data offset the performance gains
of the caching user.
</para>

<para>
As each additional client attempts to access a file with oplocks set,
the potential performance improvement is negated and eventually results
in a performance bottleneck.
</para>

</sect3>

<sect3>
<title>UNIX or NFS Client-Accessed Files</title>

<para>
<indexterm significance="normal"><primary>NFS clients</primary></indexterm>
<indexterm significance="normal"><primary>data corruption</primary></indexterm>
Local UNIX and NFS clients access files without a mandatory
file-locking mechanism. Thus, these client platforms are incapable of
initiating an oplock break request from the server to a Windows client
that has a file cached. Local UNIX or NFS file access can therefore
write to a file that has been cached by a Windows client, which
exposes the file to likely data corruption.
</para>

<para>
If files are shared between Windows clients and either local UNIX 
or NFS users, turn oplocks off.
</para>

</sect3>

<sect3>
<title>Slow and/or Unreliable Networks</title>

<para>
<indexterm significance="normal"><primary>performance improvement</primary></indexterm>
<indexterm significance="normal"><primary>WAN</primary></indexterm>
<indexterm significance="normal"><primary>latency</primary></indexterm>
The biggest potential performance improvement for oplocks
occurs when the client-side caching of reads and writes delivers the
most differential over sending those reads and writes over the wire.
This is most likely to occur when the network is extremely slow,
congested, or distributed (as in a WAN). However, network latency also
has a high impact on the reliability of the oplock break
mechanism, and thus increases the likelihood of encountering oplock
problems that more than offset the potential perceived performance
gain. Of course, if an oplock break never has to be sent, then this is
the most advantageous scenario in which to utilize oplocks.
</para>

<para>
If the network is slow, unreliable, or a WAN, then do not configure
oplocks if there is any chance of multiple users
regularly opening the same file.
</para>

</sect3>

<sect3>
<title>Multiuser Databases</title>

<para>
<indexterm significance="normal"><primary>Multiuser databases</primary></indexterm>
<indexterm significance="normal"><primary>management bottleneck</primary></indexterm>
<indexterm significance="normal"><primary>oplocks disabled</primary></indexterm>
Multiuser databases clearly pose a risk due to their very nature <?latex --- ?> they are typically heavily
accessed by numerous users at random intervals. Placing a multiuser database on a share with oplocks enabled
will likely result in a locking management bottleneck on the Samba server. Whether the database application is
developed in-house or a commercially available product, ensure that the share has oplocks disabled.
</para>

</sect3>

<sect3>
<title>PDM Data Shares</title>

<para>
<indexterm significance="normal"><primary>PDM</primary></indexterm>
<indexterm significance="normal"><primary>Process data management</primary></indexterm>
<indexterm significance="normal"><primary>client-side data caching</primary></indexterm>
<indexterm significance="normal"><primary>oplocks management</primary></indexterm>
<indexterm significance="normal"><primary>disabling oplocks</primary></indexterm>
Process data management (PDM) applications such as IMAN, Enovia, and Clearcase are increasing in usage with
Windows client platforms and therefore with SMB datastores. PDM applications manage multiuser environments for
critical data security and access. The typical PDM environment is usually associated with sophisticated client
design applications that will load data locally as demanded. In addition, the PDM application will usually
monitor the data state of each client.  In this case, client-side data caching is best left to the local
application and PDM server to negotiate and maintain. It is appropriate to eliminate the client OS from any
caching tasks, and the server from any oplocks management, by disabling oplocks on the share.
</para>

</sect3>

<sect3>
<title>Beware of Force User</title>

<para>
<indexterm significance="normal"><primary>oplock break</primary></indexterm>
Samba includes an <filename moreinfo="none">smb.conf</filename> parameter called <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="FORCEUSER" xlink:href="smb.conf.5.html#FORCEUSER">force user</link> that changes the user
accessing a share from the incoming user to whatever user is defined by the <filename moreinfo="none">smb.conf</filename> variable. If oplocks is
enabled on a share, the change in user access causes an oplock break to be sent to the client, even if the
user has not explicitly loaded a file. In cases where the network is slow or unreliable, an oplock break can
become lost without the user even accessing a file. This can cause apparent performance degradation as the
client continually reconnects to overcome the lost oplock break.
</para>

<para>
Avoid the combination of the following: 
</para>

<itemizedlist>
	<listitem><para>
	<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="FORCEUSER" xlink:href="smb.conf.5.html#FORCEUSER">force user</link> in the <filename moreinfo="none">smb.conf</filename> share configuration.
	</para></listitem>

	<listitem><para>
	Slow or unreliable networks.
	</para></listitem>

	<listitem><para>
	Oplocks enabled.
	</para></listitem>
</itemizedlist>

</sect3>

<sect3>
<title>Advanced Samba Oplocks Parameters</title>

<para>
<indexterm significance="normal"><primary>oplock parameters</primary></indexterm>
<indexterm significance="normal"><primary>oplock mechanism</primary></indexterm>
<indexterm significance="normal"><primary>implementing oplocks</primary></indexterm>
Samba provides oplock parameters that allow the
administrator to adjust various properties of the oplock mechanism to
account for timing and usage levels. These parameters provide good
versatility for implementing oplocks in environments where they would
likely cause problems. The parameters are 
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="OPLOCKBREAKWAITTIME" xlink:href="smb.conf.5.html#OPLOCKBREAKWAITTIME">oplock break wait time</link>, and
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="OPLOCKCONTENTIONLIMIT" xlink:href="smb.conf.5.html#OPLOCKCONTENTIONLIMIT">oplock contention limit</link>.
</para>

<para>
<indexterm significance="normal"><primary>turn oplocks off</primary></indexterm>
For most users, administrators, and environments, if these parameters
are required, then the better option is simply to turn oplocks off.
The Samba SWAT help text for both parameters reads: <quote>Do not change
this parameter unless you have read and understood the Samba oplock code.</quote>
This is good advice.
</para>

</sect3>

<sect3>
<title>Mission-Critical, High-Availability</title>

<para>
In mission-critical, high-availability environments, data integrity is
often a priority. Complex and expensive configurations are implemented
to ensure that if a client loses connectivity with a file server, a
failover replacement will be available immediately to provide
continuous data availability.
</para>

<para>
Windows client failover behavior is more at risk of application
interruption than other platforms because it is dependent upon an
established TCP transport connection. If the connection is interrupted
<?latex --- ?> as in a file server failover <?latex --- ?> a new session must be established.
It is rare for Windows client applications to be coded to recover
correctly from a transport connection loss; therefore, most applications
will experience some sort of interruption <?latex --- ?> at worst, abort and
require restarting.
</para>

<para>
If a client session has been caching writes and reads locally due to
oplocks, it is likely that the data will be lost when the
application restarts or recovers from the TCP interrupt. When the TCP
connection drops, the client state is lost. When the file server
recovers, an oplock break is not sent to the client. In this case, the
work from the prior session is lost. Observing this scenario with
oplocks disabled, if the client was writing data to the file server
real-time, then the failover will provide the data on disk as it
existed at the time of the disconnect.
</para>

<para>
In mission-critical, high-availability environments, careful attention
should be given to oplocks. Ideally, comprehensive
testing should be done with all affected applications with oplocks
enabled and disabled.
</para>

</sect3>
</sect2>
</sect1>

<sect1>
<title>Samba Oplocks Control</title>

<para>
Oplocks is a unique Windows file locking feature. It is
not really file locking, but is included in most discussions of Windows
file locking, so is considered a de facto locking feature.
Oplocks is actually part of the Windows client file
caching mechanism. It is not a particularly robust or reliable feature
when implemented on the variety of customized networks that exist in
enterprise computing.
</para>

<para>
Like Windows, Samba implements oplocks as a server-side
component of the client caching mechanism. Because of the lightweight
nature of the Windows feature design, effective configuration of
oplocks requires a good understanding of its limitations,
and then applying that understanding when configuring data access for
each particular customized network and client usage state.
</para>

<para>
Oplocks essentially means that the client is allowed to download and cache
a file on its hard drive while making changes; if a second client wants to access the
file, the first client receives a break and must synchronize the file back to the server.
This can give significant performance gains in some cases; some programs insist on
synchronizing the contents of the entire file back to the server for a single change.
</para>

<para>
Level1 Oplocks (also known as just plain <quote>oplocks</quote>) is another term for opportunistic locking.
</para>

<para>
Level2 Oplocks provides opportunistic locking for a file that will be treated as
<emphasis>read only</emphasis>. Typically this is used on files that are read-only or
on files that the client has no initial intention to write to at time of opening the file.
</para>

<para>
Kernel Oplocks are essentially a method that allows the Linux kernel to co-exist with
Samba's oplocked files, although this has provided better integration of MS Windows network
file locking with the underlying OS. SGI IRIX and Linux are the only two OSs that are
oplock-aware at this time.
</para>

<para>
Unless your system supports kernel oplocks, you should disable oplocks if you are
accessing the same files from both UNIX/Linux and SMB clients. Regardless, oplocks should
always be disabled if you are sharing a database file (e.g., Microsoft Access) between
multiple clients, because any break the first client receives will affect synchronization of
the entire file (not just the single record), which will result in a noticeable performance
impairment and, more likely, problems accessing the database in the first place. Notably,
Microsoft Outlook's personal folders (*.pst) react quite badly to oplocks. If in doubt,
disable oplocks and tune your system from that point.
</para>

<para>
If client-side caching is desirable and reliable on your network, you will benefit from
turning on oplocks. If your network is slow and/or unreliable, or you are sharing your
files among other file sharing mechanisms (e.g., NFS) or across a WAN, or multiple people
will be accessing the same files frequently, you probably will not benefit from the overhead
of your client sending oplock breaks and will instead want to disable oplocks for the share.
</para>

<para>
Another factor to consider is the perceived performance of file access. If oplocks provide no
measurable speed benefit on your network, it might not be worth the hassle of dealing with them.
</para>

<sect2>
<title>Example Configuration</title>

<para>
In the following section we examine two distinct aspects of Samba locking controls.
</para>

<sect3>
<title>Disabling Oplocks</title>

<para>
You can disable oplocks on a per-share basis with the following:
</para>

<para>
<simplelist>
<member> </member><member><parameter>[acctdata]</parameter></member>
<member><indexterm>oplocks</indexterm><parameter>oplocks = False</parameter></member>
<member><indexterm>level2 oplocks</indexterm><parameter>level2 oplocks = False</parameter></member>
</simplelist>
</para>

<para>
The default oplock type is Level1. Level2 oplocks are enabled on a per-share basis
in the <filename moreinfo="none">smb.conf</filename> file.
</para>

<para>
Alternately, you could disable oplocks on a per-file basis within the share:
</para>

<para>
	<simplelist>
<member><indexterm>veto oplock files</indexterm><parameter>veto oplock files = /*.mdb/*.MDB/*.dbf/*.DBF/</parameter></member>
</simplelist>
</para>

<para>
If you are experiencing problems with oplocks, as apparent from Samba's log entries,
you may want to play it safe and disable oplocks and Level2 oplocks.
</para>

</sect3>

<sect3>
<title>Disabling Kernel Oplocks</title>

<para>
Kernel oplocks is an <filename moreinfo="none">smb.conf</filename> parameter that notifies Samba (if
the UNIX kernel has the capability to send a Windows client an oplock
break) when a UNIX process is attempting to open the file that is
cached. This parameter addresses sharing files between UNIX and
Windows with oplocks enabled on the Samba server: the UNIX process
can open the file that is Oplocked (cached) by the Windows client and
the smbd process will not send an oplock break, which exposes the file
to the risk of data corruption. If the UNIX kernel has the ability to
send an oplock break, then the kernel oplocks parameter enables Samba
to send the oplock break. Kernel oplocks are enabled on a per-server
basis in the <filename moreinfo="none">smb.conf</filename> file.
</para>

<para>
<simplelist>
<member><indexterm>kernel oplocks</indexterm><parameter>kernel oplocks = yes</parameter></member>
</simplelist>
The default is no.
</para>

<para>
<emphasis>Veto oplocks</emphasis> is an <filename moreinfo="none">smb.conf</filename> parameter that identifies specific files for
which oplocks are disabled. When a Windows client opens a file that
has been configured for veto oplocks, the client will not be granted
the oplock, and all operations will be executed on the original file on
disk instead of a client-cached file copy. By explicitly identifying
files that are shared with UNIX processes and disabling oplocks for
those files, the server-wide oplock configuration can be enabled to
allow Windows clients to utilize the performance benefit of file
caching without the risk of data corruption. Veto oplocks can be
enabled on a per-share basis, or globally for the entire server, in the
<filename moreinfo="none">smb.conf</filename> file as shown in <link linkend="far1"/>.
</para>

<para>
<example id="far1">
<title>Share with Some Files Oplocked</title>
<simplelist>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>veto oplock files</indexterm><parameter>veto oplock files = /filename.htm/*.txt/</parameter></member>

<member> </member><member><parameter>[share_name]</parameter></member>
<member><indexterm>veto oplock files</indexterm><parameter>veto oplock files = /*.exe/filename.ext/</parameter></member>
</simplelist>
</example>
</para>

<para>
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="OPLOCKBREAKWAITTIME" xlink:href="smb.conf.5.html#OPLOCKBREAKWAITTIME">oplock break wait time</link> is an <filename moreinfo="none">smb.conf</filename> parameter
that adjusts the time interval for Samba to reply to an oplock break request. Samba recommends:
<quote>Do not change this parameter unless you have read and understood the Samba oplock code.</quote>
Oplock break wait time can only be configured globally in the <filename moreinfo="none">smb.conf</filename> file as shown:
</para>

<para>
	<simplelist>
<member><indexterm>oplock break wait time</indexterm><parameter>oplock break wait time =  0 (default)</parameter></member>
</simplelist>
</para>

<para>
<emphasis>Oplock break contention limit</emphasis> is an <filename moreinfo="none">smb.conf</filename> parameter that limits the
response of the Samba server to grant an oplock if the configured
number of contending clients reaches the limit specified by the parameter. Samba recommends
<quote>Do not change this parameter unless you have read and understood the Samba oplock code.</quote>
Oplock break contention limit can be enabled on a per-share basis, or globally for
the entire server, in the <filename moreinfo="none">smb.conf</filename> file as shown in <link linkend="far3"/>.
</para>

<para>
<example id="far3">
<title>Configuration with Oplock Break Contention Limit</title>
<simplelist>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>oplock break contention limit</indexterm><parameter>oplock break contention limit =  2 (default)</parameter></member>

<member> </member><member><parameter>[share_name]</parameter></member>
<member><indexterm>oplock break contention limit</indexterm><parameter>oplock break contention limit =  2 (default)</parameter></member>
</simplelist>
</example>
</para>

</sect3>
</sect2>

</sect1>

<sect1>
<title>MS Windows Oplocks and Caching Controls</title>

<para>
There is a known issue when running applications (like Norton Antivirus) on a Windows 2000/ XP
workstation computer that can affect any application attempting to access shared database files
across a network. This is a result of a default setting configured in the Windows 2000/XP
operating system. When a workstation
attempts to access shared data files located on another Windows 2000/XP computer,
the Windows 2000/XP operating system will attempt to increase performance by locking the
files and caching information locally. When this occurs, the application is unable to
properly function, which results in an <quote>Access Denied</quote>
 error message being displayed during network operations.
</para>

<para>
All Windows operating systems in the NT family that act as database servers for data files
(meaning that data files are stored there and accessed by other Windows PCs) may need to
have oplocks disabled in order to minimize the risk of data file corruption.
This includes Windows 9x/Me, Windows NT, Windows 200x, and Windows XP.
<footnote><para>Microsoft has documented this in Knowledge Base article 300216.</para></footnote>
</para>

<para>
If you are using a Windows NT family workstation in place of a server, you must also
disable oplocks on that workstation. For example, if you use a
PC with the Windows NT Workstation operating system instead of Windows NT Server, and you
have data files located on it that are accessed from other Windows PCs, you may need to
disable oplocks on that system.
</para>

<para>
The major difference is the location in the Windows registry where the values for disabling
oplocks are entered. Instead of the LanManServer location, the LanManWorkstation location
may be used.
</para>

<para>
You can verify (change or add, if necessary) this registry value using the Windows
Registry Editor. When you change this registry value, you will have to reboot the PC
to ensure that the new setting goes into effect.
</para>

<para>
The location of the client registry entry for oplocks has changed in
Windows 2000 from the earlier location in Microsoft Windows NT.
</para>

<note><para>
Windows 2000 will still respect the EnableOplocks registry value used to disable oplocks
in earlier versions of Windows.
</para></note>

<para>
You can also deny the granting of oplocks by changing the following registry entries:
</para>

<para>
<programlisting format="linespecific">
	HKEY_LOCAL_MACHINE\System\
		CurrentControlSet\Services\MRXSmb\Parameters\

		OplocksDisabled REG_DWORD 0 or 1
		Default: 0 (not disabled)
</programlisting>
</para>

<note><para>
The OplocksDisabled registry value configures Windows clients to either request or not
request oplocks on a remote file. To disable oplocks, the value of
 OplocksDisabled must be set to 1.
</para></note>

<para>
<programlisting format="linespecific">
	HKEY_LOCAL_MACHINE\System\
		CurrentControlSet\Services\LanmanServer\Parameters

		EnableOplocks REG_DWORD 0 or 1
		Default: 1 (Enabled by Default)

		EnableOpLockForceClose REG_DWORD 0 or 1
		Default: 0 (Disabled by Default)
</programlisting>
</para>

<note><para>
The EnableOplocks value configures Windows-based servers (including Workstations sharing
files) to allow or deny oplocks on local files.
</para></note>

<para>
To force closure of open oplocks on close or program exit, EnableOpLockForceClose must be set to 1.
</para>

<para>
An illustration of how Level2 oplocks work follows:
</para>

<itemizedlist>
	<listitem><para>
	Station 1 opens the file requesting oplock.
	</para></listitem>
	<listitem><para>
	Since no other station has the file open, the server grants station 1 exclusive oplock.
	</para></listitem>
	<listitem><para>
	Station 2 opens the file requesting oplock.
	</para></listitem>
	<listitem><para>
	Since station 1 has not yet written to the file, the server asks station 1 to break
	to Level2 oplock.
	</para></listitem>
	<listitem><para>
	Station 1 complies by flushing locally buffered lock information to the server.
	</para></listitem>
	<listitem><para>
	Station 1 informs the server that it has broken to level2 Oplock (alternately,
	station 1 could have closed the file).
	</para></listitem>
	<listitem><para>
	The server responds to station 2's open request, granting it Level2 oplock.
	Other stations can likewise open the file and obtain Level2 oplock.
	</para></listitem>
	<listitem><para>
	Station 2 (or any station that has the file open) sends a write request SMB.
	The server returns the write response.
	</para></listitem>
	<listitem><para>
	The server asks all stations that have the file open to break to none, meaning no
	station holds any oplock on the file. Because the workstations can have no cached
	writes or locks at this point, they need not respond to the break-to-none advisory;
	all they need do is invalidate locally cashed read-ahead data.
	</para></listitem>
</itemizedlist>

<sect2>
<title>Workstation Service Entries</title>

<para><programlisting format="linespecific">
	\HKEY_LOCAL_MACHINE\System\
		CurrentControlSet\Services\LanmanWorkstation\Parameters

	UseOpportunisticLocking   REG_DWORD   0 or 1
	Default: 1 (true)
</programlisting></para>

<para>
This indicates whether the redirector should use oplocks performance
enhancement. This parameter should be disabled only to isolate problems.
</para>

</sect2>
<sect2>
<title>Server Service Entries</title>

<para><programlisting format="linespecific">
	\HKEY_LOCAL_MACHINE\System\
		CurrentControlSet\Services\LanmanServer\Parameters

	EnableOplocks   REG_DWORD   0 or 1
	Default: 1 (true)
</programlisting></para>

<para>
This specifies whether the server allows clients to use oplocks on files. Oplocks are a
significant performance enhancement, but have the potential to cause lost cached
data on some networks, particularly WANs.
</para>

<para><programlisting format="linespecific">
	MinLinkThroughput   REG_DWORD   0 to infinite bytes per second
	Default: 0
</programlisting></para>

<para>
This specifies the minimum link throughput allowed by the server before it disables
raw I/O and oplocks for this connection.
</para>

<para><programlisting format="linespecific">
	MaxLinkDelay   REG_DWORD   0 to 100,000 seconds
	Default: 60
</programlisting></para>

<para>
This specifies the maximum time allowed for a link delay. If delays exceed this number,
the server disables raw I/O and oplocks for this connection.
</para>

<para><programlisting format="linespecific">
	OplockBreakWait   REG_DWORD   10 to 180 seconds
	Default: 35
</programlisting></para>

<para>
This specifies the time that the server waits for a client to respond to an oplock break
request. Smaller values can allow detection of crashed clients more quickly but can
potentially cause loss of cached data.
</para>

</sect2>
</sect1>

<sect1>
<title>Persistent Data Corruption</title>

<para>
If you have applied all of the settings discussed in this chapter but data corruption problems
and other symptoms persist, here are some additional things to check out.
</para>

<para>
We have credible reports from developers that faulty network hardware, such as a single
faulty network card, can cause symptoms similar to read caching and data corruption.
If you see persistent data corruption even after repeated re-indexing, you may have to
rebuild the data files in question. This involves creating a new data file with the
same definition as the file to be rebuilt and transferring the data from the old file
to the new one. There are several known methods for doing this that can be found in
our knowledge base.
</para>

</sect1>

<sect1>
<title>Common Errors</title>

<para>
In some sites locking problems surface as soon as a server is installed; in other sites
locking problems may not surface for a long time. Almost without exception, when a locking
problem does surface, it will cause embarrassment and potential data corruption.
</para>

<para>
Over the past few years there have been a number of complaints on the Samba mailing lists
that have claimed that Samba caused data corruption. Three causes have been identified
so far:
</para>

<itemizedlist>
	<listitem><para>
	Incorrect configuration of oplocks (incompatible with the application
	being used). This is a common problem even where MS Windows NT4 or MS Windows
	200x-based servers were in use. It is imperative that the software application vendors'
	instructions for configuration of file locking should be followed. If in doubt,
	disable oplocks on both the server and the client. Disabling of all forms of file
	caching on the MS Windows client may be necessary also.
	</para></listitem>

	<listitem><para>
	Defective network cards, cables, or hubs/switches. This is generally a more
	prevalent factor with low-cost networking hardware, although occasionally there
	have also been problems with incompatibilities in more up-market hardware.
	</para></listitem>

	<listitem><para>
	There have been some random reports of Samba log files being written over data
	files. This has been reported by very few sites (about five in the past 3 years)
	and all attempts to reproduce the problem have failed. The Samba Team has been
	unable to catch this happening and thus unable to isolate any particular
	cause. Considering the millions of systems that use Samba, for the sites that have
	been affected by this as well as for the Samba Team, this is a frustrating and
	vexing challenge. If you see this type of thing happening, please create a bug
	report on Samba <ulink url="https://bugzilla.samba.org">Bugzilla</ulink> without delay.
	Make sure that you give as much information as you possibly can to help isolate the
	cause and to allow replication of the problem (an essential step in problem isolation and correction).
	</para></listitem>
</itemizedlist>

	<sect2>
	<title>locking.tdb Error Messages</title>

	<para>
		<quote>
			We are seeing lots of errors in the Samba logs, like:
		</quote>
<programlisting format="linespecific">
tdb(/usr/local/samba_2.2.7/var/locks/locking.tdb): rec_read bad magic
 0x4d6f4b61 at offset=36116
</programlisting>

		<quote>
			What do these mean?
		</quote>
	</para>

	<para>
	This error indicates a corrupted tdb. Stop all instances of smbd, delete locking.tdb, and restart smbd.
	</para>

	</sect2>

	<sect2>
		<title>Problems Saving Files in MS Office on Windows XP</title>

<indexterm significance="normal"><primary>KB 812937</primary></indexterm>
		<para>This is a bug in Windows XP. More information can be 
		found in <ulink url="http://support.microsoft.com/?id=812937">Microsoft Knowledge Base article 812937</ulink></para>.

	</sect2>

	<sect2>

		<title>Long Delays Deleting Files over Network with XP SP1</title>
		
		<para><quote>It sometimes takes approximately 35 seconds to delete files over the network after XP SP1 has been applied.</quote></para>

<indexterm significance="normal"><primary>KB 811492</primary></indexterm>
		<para>This is a bug in Windows XP. More information can be found in <ulink url="http://support.microsoft.com/?id=811492">
				Microsoft Knowledge Base article 811492</ulink></para>.
	</sect2>

</sect1>

<sect1>
<title>Additional Reading</title>

<para>
You may want to check for an updated documentation regarding file and record locking issues on the Microsoft
<ulink url="http://support.microsoft.com/">Support</ulink> web site. Additionally, search for the word
<literal moreinfo="none">locking</literal> on the Samba <ulink url="http://www.samba.org/">web</ulink> site.
</para>

<para>
Section of the Microsoft MSDN Library on opportunistic locking: 
</para>

<para>
<indexterm significance="normal"><primary>KB 224992</primary></indexterm>
Microsoft Knowledge Base, <quote>Maintaining Transactional Integrity with OPLOCKS</quote>,
Microsoft Corporation, April 1999, <ulink noescape="1" url="http://support.microsoft.com/?id=224992">Microsoft
KB Article 224992</ulink>.
</para>

<para>
<indexterm significance="normal"><primary>KB 296264</primary></indexterm>
Microsoft Knowledge Base, <quote>Configuring Opportunistic Locking in Windows 2000</quote>,
Microsoft Corporation, April 2001 <ulink noescape="1" url="http://support.microsoft.com/?id=296264">Microsoft KB Article 296264</ulink>.
</para>

<para>
<indexterm significance="normal"><primary>KB 129202</primary></indexterm>
Microsoft Knowledge Base, <quote>PC Ext: Explanation of Opportunistic Locking on Windows NT</quote>,
Microsoft Corporation, April 1995 <ulink noescape="1" url="http://support.microsoft.com/?id=129202">Microsoft
KB Article 129202</ulink>.
</para>

</sect1>
</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="securing-samba">

<chapterinfo>
	
<author>
	<firstname>Andrew</firstname><surname>Tridgell</surname>
	<affiliation>
		<orgname>Samba Team</orgname>
		<address format="linespecific"><email>tridge@samba.org</email></address>
	</affiliation>
</author>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
	<pubdate>May 26, 2003</pubdate>
</chapterinfo>

<title>Securing Samba</title>

<sect1>
<title>Introduction</title>

<para>
<indexterm significance="normal"><primary>security</primary></indexterm>
<indexterm significance="normal"><primary>direct internet access</primary></indexterm>
<indexterm significance="normal"><primary>firewall</primary></indexterm>
<indexterm significance="normal"><primary>private network</primary></indexterm>
<indexterm significance="normal"><primary>barriers</primary></indexterm>
<indexterm significance="normal"><primary>deterents</primary></indexterm>
<indexterm significance="normal"><primary>secured networks</primary></indexterm>
The information contained in this chapter applies in general to all Samba installations. Security is
everyone's concern in the information technology world. A surprising number of Samba servers are being
installed on machines that have direct internet access, thus security is made more critical than it would have been had the
server been located behind a firewall and on a private network. Paranoia regarding server security is causing
some  network administrators to insist on the installation of robust firewalls even on servers that are located
inside secured networks. This chapter provides information to assist the administrator who understands
how to create the needed barriers and deterents against <quote>the enemy</quote>, no matter where [s]he may
come from.
</para>

<blockquote>
<para>
A new apprentice reported for duty to the chief engineer of a boiler house. He said, <quote>Here I am,
if you will show me the boiler I'll start working on it.</quote> Then engineer replied, <quote>You're leaning
on it!</quote>
</para>
</blockquote>

<para>
Security concerns are just like that. You need to know a little about the subject to appreciate
how obvious most of it really is. The challenge for most of us is to discover that first morsel
of knowledge with which we may unlock the secrets of the masters.
</para>

</sect1>

<sect1>
<title>Features and Benefits</title>

<para>
<indexterm significance="normal"><primary>moderately secure</primary></indexterm>
<indexterm significance="normal"><primary>perimeter firewall</primary></indexterm>
<indexterm significance="normal"><primary>host security</primary></indexterm>
<indexterm significance="normal"><primary>Samba security</primary></indexterm>
There are three levels at which security principles must be observed in order to render a site
at least moderately secure. They are the perimeter firewall, the configuration of the host
server that is running Samba, and Samba itself.
</para>

<para>
Samba permits a most flexible approach to network security. As far as possible Samba implements
the latest protocols to permit more secure MS Windows file and print operations.
</para>

<para>
<indexterm significance="normal"><primary>host-based protection</primary></indexterm>
<indexterm significance="normal"><primary>interface-based exclusion</primary></indexterm>
<indexterm significance="normal"><primary>resource-based exclusion</primary></indexterm>
Samba can be secured from connections that originate from outside the local network. This can be done using
<emphasis>host-based protection</emphasis>, using Samba's implementation of a technology known as
<quote>tcpwrappers,</quote> or it may be done be using <emphasis>interface-based exclusion</emphasis> so
<application moreinfo="none">smbd</application> will bind only to specifically permitted interfaces. It is also possible to set specific share- or
resource-based exclusions, for example, on the <parameter>[IPC$]</parameter> autoshare. The <parameter>[IPC$]</parameter> share is used for browsing purposes as well as to establish TCP/IP connections.
</para>

<para>
<indexterm significance="normal"><primary>Access Control Entries</primary><see>ACE</see></indexterm>
<indexterm significance="normal"><primary>ACL</primary></indexterm>
<indexterm significance="normal">controls<primary/></indexterm>
Another method by which Samba may be secured is by setting Access Control Entries (ACEs) in an Access 
Control List (ACL) on the shares themselves. This is discussed in
<link linkend="AccessControls">File, Directory, and Share Access Controls</link>.
</para>

</sect1>

<sect1>
<title>Technical Discussion of Protective Measures and Issues</title>

<para>
The key challenge of security is that protective measures suffice at best
only to close the door on known exploits and breach techniques. Never assume that
because you have followed these few measures, the Samba server is now an impenetrable
fortress! Given the history of information systems so far, it is only a matter of time
before someone will find yet another vulnerability.
</para>

	<sect2>
	<title>Using Host-Based Protection</title>

	<para>
<indexterm significance="normal"><primary>outside threat</primary></indexterm>
<indexterm significance="normal"><primary>insecure</primary></indexterm>
<indexterm significance="normal"><primary>Internet</primary></indexterm>
	In many installations of Samba, the greatest threat comes from outside
	your immediate network. By default, Samba accepts connections from
	any host, which means that if you run an insecure version of Samba on
	a host that is directly connected to the Internet, you can be
	especially vulnerable.
	</para>

	<para>
<indexterm significance="normal"><primary>allow access</primary></indexterm>
<indexterm significance="normal"><primary>range of hosts</primary></indexterm>
	One of the simplest fixes in this case is to use the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="HOSTSALLOW" xlink:href="smb.conf.5.html#HOSTSALLOW">hosts allow</link> and
	<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="HOSTSDENY" xlink:href="smb.conf.5.html#HOSTSDENY">hosts deny</link> options in the Samba <filename moreinfo="none">smb.conf</filename> configuration file to
	allow access to your server only from a specific range of hosts. An example might be:
	<simplelist>
	<member><indexterm>hosts allow</indexterm><parameter>hosts allow = 127.0.0.1 192.168.2.0/24 192.168.3.0/24</parameter></member>
	<member><indexterm>hosts deny</indexterm><parameter>hosts deny = 0.0.0.0/0</parameter></member>
	</simplelist>
	</para>

	<para>
<indexterm significance="normal"><primary>localhost</primary></indexterm>
<indexterm significance="normal"><primary>private networks</primary></indexterm>
<indexterm significance="normal"><primary>called name</primary></indexterm>
	The above will allow SMB connections only from <constant>localhost</constant> (your own
	computer) and from the two private networks 192.168.2 and 192.168.3. All other
	connections will be refused as soon as the client sends its first packet. The refusal
	will be marked as <literal moreinfo="none">not listening on called name</literal> error.
	</para>

	</sect2>

	<sect2>
	<title>User-Based Protection</title>

	<para>
	If you want to restrict access to your server to valid users only, then the following
	method may be of use. In the <filename moreinfo="none">smb.conf</filename> <parameter>[global]</parameter> section put:
	<simplelist>
	<member><indexterm>valid users</indexterm><parameter>valid users = @smbusers, jacko</parameter></member>
	</simplelist>
	</para>

	<para>
<indexterm significance="normal"><primary>smbusers</primary></indexterm>
	This restricts all server access either to the user <emphasis>jacko</emphasis>
	or to members of the system group <emphasis>smbusers</emphasis>.
	</para>

	</sect2>

	<sect2>

	<title>Using Interface Protection</title>

	<para>
<indexterm significance="normal"><primary>network interface</primary></indexterm>
<indexterm significance="normal"><primary>accept connections</primary></indexterm>
<indexterm significance="normal"><primary>Internet</primary></indexterm>
	By default, Samba accepts connections on any network interface that
	it finds on your system. That means if you have an ISDN line or a PPP
	connection to the Internet then Samba will accept connections on those
	links. This may not be what you want.
	</para>

	<para>
	You can change this behavior using options like this:
	<simplelist>
	<member><indexterm>interfaces</indexterm><parameter>interfaces = eth* lo</parameter></member>
	<member><indexterm>bind interfaces only</indexterm><parameter>bind interfaces only = yes</parameter></member>
	</simplelist>
	</para>

	<para>
<indexterm significance="normal"><primary>interfaces</primary></indexterm>
<indexterm significance="normal"><primary>loopback interface</primary></indexterm>
<indexterm significance="normal"><primary>Ethernet adapters</primary></indexterm>
<indexterm significance="normal"><primary>listen for connections</primary></indexterm>
	This tells Samba to listen for connections only on interfaces with a name starting with
	<constant>eth</constant> such as <constant>eth0</constant> or <constant>eth1</constant>, plus on the loopback interface called
	<constant>lo</constant>. The name you will need to use depends on what OS you are using. In the above, I used
	the common name for Ethernet adapters on Linux.
	</para>

	<para>
<indexterm significance="normal"><primary>PPP</primary></indexterm>
<indexterm significance="normal"><primary>SMB</primary></indexterm>
<indexterm significance="normal"><primary>cracker</primary></indexterm>
<indexterm significance="normal"><primary>confirm address</primary></indexterm>
	If you use the above and someone tries to make an SMB connection to your host over a PPP interface called
	<constant>ppp0</constant>, then [s]he will get a TCP connection refused reply. In that case, no Samba code
	is run at all, because the operating system has been told not to pass connections from that interface to any
	Samba process. However, the refusal helps a would-be cracker by confirming that the IP address provides
	valid active services.
	</para>

	<para>
<indexterm significance="normal"><primary>ignore connection</primary></indexterm>
<indexterm significance="normal"><primary>refusing connection</primary></indexterm>
<indexterm significance="normal"><primary>exploitation</primary></indexterm>
<indexterm significance="normal"><primary>denial of service</primary></indexterm>
<indexterm significance="normal"><primary>firewall</primary></indexterm>
	A better response would be to ignore the connection (from, for example, ppp0) altogether. The
	advantage of ignoring the connection attempt, as compared with refusing it, is that it foils those who
	probe an interface with the sole intention of finding valid IP addresses for later use in exploitation
	or denial of service attacks. This method of dealing with potential malicious activity demands the
	use of appropriate firewall mechanisms.
	</para>

	</sect2>

	<sect2 id="firewallports">
	<title>Using a Firewall</title>

	<para>
<indexterm significance="normal"><primary>deny access</primary></indexterm>
<indexterm significance="normal"><primary>exposed</primary></indexterm>
<indexterm significance="normal"><primary>firewall active</primary></indexterm>
	Many people use a firewall to deny access to services they do not want exposed outside their network. This can
	be a good idea, although I recommend using it in conjunction with the above methods so you are protected even
	if your firewall is not active for some reason.
	</para>

	<para>
	If you are setting up a firewall, you need to know what TCP and UDP ports to allow and block. Samba uses
	the following:
<indexterm significance="normal"><primary>Port 135/TCP</primary></indexterm>
<indexterm significance="normal"><primary>Port 137/UDP</primary></indexterm>
<indexterm significance="normal"><primary>Port 138/UDP</primary></indexterm>
<indexterm significance="normal"><primary>Port 139/TCP</primary></indexterm>
<indexterm significance="normal"><primary>Port 445/TCP</primary></indexterm>
	</para>

	<simplelist type="vert">
		<member>Port 135/TCP - used by smbd</member>
		<member>Port 137/UDP - used by nmbd</member>
		<member>Port 138/UDP - used by nmbd</member>
		<member>Port 139/TCP - used by smbd</member>
		<member>Port 445/TCP - used by smbd</member>
	</simplelist>

	<para>
<indexterm significance="normal"><primary>firewall setups</primary></indexterm>
	The last one is important because many older firewall setups may not be aware of it, given that this port
	was only added to the protocol in recent years.
	</para>

	<para>
<indexterm significance="normal"><primary>configuring a firewall</primary></indexterm>
<indexterm significance="normal"><primary>high order ports</primary></indexterm>
<indexterm significance="normal"><primary>block incoming packets</primary></indexterm>
	When configuring a firewall, the high order ports (1024-65535) are often used for outgoing connections and
	therefore should be permitted through the firewall. It is prudent to block incoming packets on the high order
	ports except for established connections.
	</para>

	</sect2>

	<sect2>
	<title>Using IPC$ Share-Based Denials </title>

	<para>
<indexterm significance="normal"><primary>IPC$</primary></indexterm>
<indexterm significance="normal"><primary>deny</primary></indexterm>
<indexterm significance="normal"><primary>security hole</primary></indexterm>
	If the above methods are not suitable, then you could also place a more specific deny on the IPC$ share that
	is used in the recently discovered security hole. This allows you to offer access to other shares while
	denying access to IPC$ from potentially untrustworthy hosts.
	</para>

	<para>
	To do this you could use:
	<simplelist>
	<member> </member><member><parameter>[IPC$]</parameter></member>
	<member><indexterm>hosts allow</indexterm><parameter>hosts allow = 192.168.115.0/24 127.0.0.1</parameter></member>
	<member><indexterm>hosts deny</indexterm><parameter>hosts deny = 0.0.0.0/0</parameter></member>
	</simplelist>
	</para>

	<para>
<indexterm significance="normal"><primary>IPC$</primary></indexterm>
<indexterm significance="normal"><primary>protection against attackers</primary></indexterm>
<indexterm significance="normal"><primary>valid username/password</primary></indexterm>
	This instructs Samba that IPC$ connections are not allowed from anywhere except the two listed network
	addresses (localhost and the 192.168.115 subnet). Connections to other shares are still allowed. Because the
	IPC$ share is the only share that is always accessible anonymously, this provides some level of protection
	against attackers who do not know a valid username/password for your host.
	</para>

	<para>
<indexterm significance="normal"><primary>access denied</primary></indexterm>
<indexterm significance="normal"><primary>IPC$</primary></indexterm>
<indexterm significance="normal"><primary>browse shares</primary></indexterm>
	If you use this method, then clients will be given an <literal moreinfo="none">`access denied'</literal> reply when they try
	to access the IPC$ share. Those clients will not be able to browse shares and may also be unable to access
	some other resources.  This is not recommended unless for some reason you cannot use one of the other methods
	just discussed.
	</para>

	</sect2>

	<sect2>
	<title>NTLMv2 Security</title>

	<para>
<indexterm significance="normal"><primary>NTLMv2</primary></indexterm>
	To configure NTLMv2 authentication, the following registry keys are worth knowing about:
	</para>

	<para>
		<screen format="linespecific">
		[HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Lsa]
		"lmcompatibilitylevel"=dword:00000003
		</screen>
	</para>

	<para>
	The value 0x00000003 means to send NTLMv2 response only. Clients will use NTLMv2 authentication;
	use NTLMv2 session security if the server supports it. Domain controllers accept LM,
	NTLM, and NTLMv2 authentication.
	</para>

	<para>
		<screen format="linespecific">
		[HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Lsa\MSV1_0]
		"NtlmMinClientSec"=dword:00080000
		</screen>
	</para>

	<para>
	The value 0x00080000 means permit only NTLMv2 session security. If either NtlmMinClientSec or
	NtlmMinServerSec is set to 0x00080000, the connection will fail if NTLMv2
	session security is negotiated.
	</para>
	</sect2>
</sect1>

<sect1>
<title>Upgrading Samba</title>

<para>
<indexterm significance="normal"><primary>updates</primary></indexterm>
<indexterm significance="normal"><primary>important announcements</primary></indexterm>
<indexterm significance="normal"><primary>security vulnerability</primary></indexterm>
Please check regularly on <ulink noescape="1" url="http://www.samba.org/">http://www.samba.org/</ulink> for
updates and important announcements. Occasionally security releases are made, and it is highly recommended to
upgrade Samba promptly when a security vulnerability is discovered. Check with your OS vendor for OS-specific
upgrades.
</para>

</sect1>

<sect1>
<title>Common Errors</title>

<para>
If all Samba and host platform configurations were really as intuitive as one might like them to be, this
chapter would not be necessary. Security issues are often vexing for a support person to resolve, not because
of the complexity of the problem, but because most administrators who post what turns out to be a security
problem request are totally convinced that the problem is with Samba.
</para>

	<sect2>
	<title>Smbclient Works on Localhost, but the Network Is Dead</title>

	<para>
	This is a common problem. Linux vendors tend to install a default firewall.
	With the default firewall in place, only traffic on the loopback adapter (IP address 127.0.0.1)
	is allowed through the firewall.
	</para>

	<para>
	The solution is either to remove the firewall (stop it) or modify the firewall script to
	allow SMB networking traffic through. See <link linkend="firewallports">the Using a 
	Firewall</link> section.
	</para>

	</sect2>

	<sect2>
	<title>Why Can Users Access Other Users' Home Directories?</title>

	<para>
	<quote>
<indexterm significance="normal"><primary>mapping home directory</primary></indexterm>
<indexterm significance="normal"><primary>own home directory</primary></indexterm>
	We are unable to keep individual users from mapping to any other user's home directory once they have
	supplied a valid password! They only need to enter their own password. I have not found any method to
	configure Samba so that users may map only their own home directory.
	</quote>
	</para>

	<para><quote>
	User xyzzy can map his home directory. Once mapped, user xyzzy can also map anyone else's home directory.
	</quote></para>

	<para>
<indexterm significance="normal"><primary>security flaw</primary></indexterm>
<indexterm significance="normal"><primary>defined shares</primary></indexterm>
	This is not a security flaw, it is by design. Samba allows users to have exactly the same access to the UNIX
	file system as when they were logged on to the UNIX box, except that it only allows such views onto the file
	system as are allowed by the defined shares.
	</para>

	<para>
<indexterm significance="normal"><primary>UNIX home directories</primary></indexterm>
<indexterm significance="normal"><primary>permissions</primary></indexterm>
	If your UNIX home directories are set up so that one user can happily <literal>cd</literal>
	into another user's directory and execute <literal>ls</literal>, the UNIX security solution is to change file
	permissions on the user's home directories so that the <literal>cd</literal> and <literal>ls</literal> are denied.
	</para>

	<para>
<indexterm significance="normal"><primary>security policies</primary></indexterm>
<indexterm significance="normal"><primary>permissions</primary></indexterm>
	Samba tries very hard not to second guess the UNIX administrator's security policies and
	trusts the UNIX admin to set the policies and permissions he or she desires.
	</para>

	<para>
	Samba allows the behavior you require. Simply put the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="ONLYUSER" xlink:href="smb.conf.5.html#ONLYUSER">only user = %S</link>
	option in the <parameter>[homes]</parameter> share definition.
	</para>

	<para>
	The <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="ONLYUSER" xlink:href="smb.conf.5.html#ONLYUSER">only user</link> works in conjunction with the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="USERS" xlink:href="smb.conf.5.html#USERS">users = list</link>,
	so to get the behavior you require, add the line:
	<simplelist>
	<member><indexterm>users</indexterm><parameter>users = %S</parameter></member>
	</simplelist>
	This is equivalent to adding
	<simplelist>
	<member><indexterm>valid users</indexterm><parameter>valid users = %S</parameter></member>
	</simplelist>
	to the definition of the <parameter>[homes]</parameter> share, as recommended in
	the <filename moreinfo="none">smb.conf</filename> man page.
	</para>
	</sect2>

</sect1>
</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="InterdomainTrusts">
<chapterinfo>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
	
<author>
	<firstname>Rafal</firstname><surname>Szczesniak</surname>
	<affiliation>
		<orgname>Samba Team</orgname>
		<address format="linespecific"><email>mimir@samba.org</email></address>
	</affiliation>
</author>
	<author>
<firstname>Jelmer</firstname><othername>R.</othername><surname>Vernooij</surname><othername>R.</othername>
<affiliation>
	<orgname>The Samba Team</orgname>
	<address format="linespecific"><email>jelmer@samba.org</email></address>
</affiliation><contrib>drawing</contrib></author>
        <author>
                <firstname>Stephen</firstname><surname>Langasek</surname>
                <affiliation>
                        <address format="linespecific"><email>vorlon@netexpress.net</email></address>
                </affiliation>
        </author>
	<pubdate>April 3, 2003</pubdate>
</chapterinfo>

<title>Interdomain Trust Relationships</title>


<para>
<indexterm significance="normal"><primary>Interdomain Trusts</primary></indexterm>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
<indexterm significance="normal"><primary>trusts</primary></indexterm>
<indexterm significance="normal"><primary>samba-to-samba trusts</primary></indexterm>
<indexterm significance="normal"><primary>Active Directory</primary></indexterm>
<indexterm significance="normal"><primary>NT4-style domain</primary></indexterm>
<indexterm significance="normal"><primary>trust relationships</primary></indexterm>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
<indexterm significance="normal"><primary>LDAP-based</primary></indexterm>
Samba-3 supports NT4-style domain trust relationships. This is a feature that many sites
will want to use if they migrate to Samba-3 from an NT4-style domain and do not want to
adopt Active Directory or an LDAP-based authentication backend. This chapter explains
some background information regarding trust relationships and how to create them. It is now
possible for Samba-3 to trust NT4 (and vice versa), as well as to create Samba-to-Samba 
trusts.
</para>

<para>
<indexterm significance="normal"><primary>winbind</primary></indexterm>
<indexterm significance="normal"><primary>UID range</primary></indexterm>
<indexterm significance="normal"><primary>GID range</primary></indexterm>
<indexterm significance="normal"><primary>daemon</primary></indexterm>
<indexterm significance="normal"><primary>winbindd</primary></indexterm>
The use of interdomain trusts requires use of <literal>winbind</literal>, so the
<literal>winbindd</literal> daemon must be running. Winbind operation in this mode is
dependent on the specification of a valid UID range and a valid GID range in the <filename moreinfo="none">smb.conf</filename> file.
These are specified respectively using:
<simplelist>
<member><indexterm>idmap uid</indexterm><parameter>idmap uid = 10000-20000</parameter></member>
<member><indexterm>idmap gid</indexterm><parameter>idmap gid = 10000-20000</parameter></member>
</simplelist>
<indexterm significance="normal"><primary>passdb backend</primary></indexterm>
<indexterm significance="normal"><primary>POSIX user accounts</primary></indexterm>
<indexterm significance="normal"><primary>maximum value</primary></indexterm>
<indexterm significance="normal"><primary>4294967295</primary></indexterm>
The range of values specified must not overlap values used by the host operating system and must
not overlap values used in the passdb backend for POSIX user accounts. The maximum value is
limited by the upper-most value permitted by the host operating system. This is a UNIX kernel
limited parameter. Linux kernel 2.6-based systems support a maximum value of 4294967295 
(32-bit unsigned variable).
</para>

<note><para>
<indexterm significance="normal"><primary>winbind</primary></indexterm>
<indexterm significance="normal"><primary>trusting domain</primary></indexterm>
<indexterm significance="normal"><primary>trusted domain</primary></indexterm>
The use of winbind is necessary only when Samba is the trusting domain, not when it is the
trusted domain.
</para></note>

<sect1>
<title>Features and Benefits</title>

<para>
<indexterm significance="normal"><primary>scalability</primary></indexterm>
<indexterm significance="normal"><primary>trust relationships</primary></indexterm>
Samba-3 can participate in Samba-to-Samba as well as in Samba-to-MS Windows NT4-style
trust relationships. This imparts to Samba scalability similar to that with MS Windows NT4.
</para>

<para>
<indexterm significance="normal"><primary>scalable backend</primary></indexterm>
<indexterm significance="normal"><primary>authentication database</primary></indexterm>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
<indexterm significance="normal"><primary>interdomain trusts</primary></indexterm>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
Given that Samba-3 can function with a scalable backend authentication database such as LDAP, and given its
ability to run in primary as well as backup domain control modes, the administrator would be well-advised to
consider alternatives to the use of interdomain trusts simply because, by the very nature of how trusts
function, this system is fragile.  That was, after all, a key reason for the development and adoption of
Microsoft Active Directory.
</para>

</sect1>

<sect1>
<title>Trust Relationship Background</title>

<para>
<indexterm significance="normal"><primary>security domains</primary></indexterm>
<indexterm significance="normal"><primary>nonhierarchical</primary></indexterm>
<indexterm significance="normal"><primary>security structure</primary></indexterm>
<indexterm significance="normal"><primary>large organizations</primary></indexterm>
<indexterm significance="normal"><primary>delegation</primary></indexterm>
<indexterm significance="normal"><primary>administrative responsibilities</primary></indexterm>
MS Windows NT3/4-type security domains employ a nonhierarchical security structure.
The limitations of this architecture as it effects the scalability of MS Windows networking
in large organizations is well known. Additionally, the flat namespace that results from
this design significantly impacts the delegation of administrative responsibilities in
large and diverse organizations.
</para>

<para>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
<indexterm significance="normal"><primary>Kerberos</primary></indexterm>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
<indexterm significance="normal"><primary>limitations</primary></indexterm>
<indexterm significance="normal"><primary>domain security</primary></indexterm>
Microsoft developed Active Directory Service (ADS), based on Kerberos and LDAP, as a means
of circumventing the limitations of the older technologies. Not every organization is ready
or willing to embrace ADS. For small companies the older NT4-style domain security paradigm
is quite adequate, and so there remains an entrenched user base for whom there is no direct
desire to go through a disruptive change to adopt ADS.
</para>

<para>
<indexterm significance="normal"><primary>security domains</primary></indexterm>
<indexterm significance="normal"><primary>access rights</primary></indexterm>
<indexterm significance="normal"><primary>privileges</primary></indexterm>
<indexterm significance="normal"><primary>trusts</primary></indexterm>
<indexterm significance="normal"><primary>trusted domain</primary></indexterm>
<indexterm significance="normal"><primary>trusting domain</primary></indexterm>
<indexterm significance="normal"><primary>one direction</primary></indexterm>
With Windows NT, Microsoft introduced the ability to allow different security domains
to effect a mechanism so users from one domain may be given access rights and privileges
in another domain. The language that describes this capability is couched in terms of
<emphasis>trusts</emphasis>. Specifically, one domain will <emphasis>trust</emphasis> the users
from another domain. The domain from which users can access another security domain is
said to be a trusted domain. The domain in which those users have assigned rights and privileges
is the trusting domain. With NT3.x/4.0 all trust relationships are always in one direction only,
so if users in both domains are to have privileges and rights in each others' domain, then it is
necessary to establish two relationships, one in each direction.
</para>

<para>
<indexterm significance="normal"><primary>security domain</primary></indexterm>
<indexterm significance="normal"><primary>nontransitive</primary></indexterm>
<indexterm significance="normal"><primary>trust relationship</primary></indexterm>
<indexterm significance="normal"><primary>transitive</primary></indexterm>
<indexterm significance="normal"><primary>explicit trust</primary></indexterm>
Further, in an NT4-style MS security domain, all trusts are nontransitive. This means that if there are three
domains (let's call them red, white, and blue), where red and white have a trust relationship, and white and
blue have a trust relationship, then it holds that there is no implied trust between the red and blue domains.
Relationships are explicit and not transitive.
</para>

<para>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
<indexterm significance="normal"><primary>security contexts</primary></indexterm>
<indexterm significance="normal"><primary>trust relationships</primary></indexterm>
<indexterm significance="normal"><primary>two-way trust</primary></indexterm>
<indexterm significance="normal"><primary>Windows 2000</primary></indexterm>
<indexterm significance="normal"><primary>security domains</primary></indexterm>
<indexterm significance="normal"><primary>NT4-style domains</primary></indexterm>
New to MS Windows 2000 ADS security contexts is the fact that trust relationships are two-way by default.
Also, all inter-ADS domain trusts are transitive. In the case of the red, white, and blue domains, with
Windows 2000 and ADS, the red and blue domains can trust each other. This is an inherent feature of ADS
domains. Samba-3 implements MS Windows NT4-style interdomain trusts and interoperates with MS Windows 200x ADS
security domains in similar manner to MS Windows NT4-style domains.
</para>

</sect1>

<sect1>
<title>Native MS Windows NT4 Trusts Configuration</title>

<para>
<indexterm significance="normal"><primary>Interdomain Trusts</primary><secondary>creating</secondary></indexterm>
<indexterm significance="normal"><primary>two-way trust</primary></indexterm>
<indexterm significance="normal"><primary>security credentials</primary></indexterm>
There are two steps to creating an interdomain trust relationship. To effect a two-way trust
relationship, it is necessary for each domain administrator to create a trust account for the 
other domain to use in verifying security credentials.
</para>


<sect2>
<title>Creating an NT4 Domain Trust</title>

<para>
<indexterm significance="normal"><primary>domain trust</primary></indexterm>
<indexterm significance="normal"><primary>trust relationships</primary></indexterm>
<indexterm significance="normal"><primary>&gt;Domain User Manager</primary></indexterm>
<indexterm significance="normal"><primary>remote domain</primary></indexterm>
<indexterm significance="normal"><primary>standard confirmation</primary></indexterm>
For MS Windows NT4, all domain trust relationships are configured using the 
<application moreinfo="none">Domain User Manager</application>. This is done from the Domain User Manager Policies
entry on the menu bar. From the <guimenu moreinfo="none">Policy</guimenu> menu, select
<guimenuitem moreinfo="none">Trust Relationships</guimenuitem>. Next to the lower box labeled
<guilabel moreinfo="none">Permitted to Trust this Domain</guilabel> are two buttons, <guibutton moreinfo="none">Add</guibutton>
and <guibutton moreinfo="none">Remove</guibutton>. The <guibutton moreinfo="none">Add</guibutton> button will open a panel in which
to enter the name of the remote domain that will be able to assign access rights to users in 
your domain. You will also need to enter a password for this trust relationship, which the 
trusting domain will use when authenticating users from the trusted domain.
The password needs to be typed twice (for standard confirmation).
</para>

</sect2>


<sect2>
<title>Completing an NT4 Domain Trust</title>

<para>
<indexterm significance="normal"><primary>trust relationship</primary></indexterm>
<indexterm significance="normal"><primary>trusting domain</primary></indexterm>
<indexterm significance="normal"><primary>trusted domain</primary></indexterm>
<indexterm significance="normal"><primary>remote domain</primary></indexterm>
<indexterm significance="normal"><primary>password assigned</primary></indexterm>
<indexterm significance="normal"><primary>Interdomain Trusts</primary><secondary>Completing</secondary></indexterm>
A trust relationship will work only when the other (trusting) domain makes the appropriate connections
with the trusted domain. To consummate the trust relationship, the administrator launches the
Domain User Manager from the menu selects <guilabel moreinfo="none">Policies</guilabel>, then select
<guilabel moreinfo="none">Trust Relationships</guilabel>, and clicks on the <guibutton moreinfo="none">Add</guibutton> button
next to the box that is labeled <guilabel moreinfo="none">Trusted Domains</guilabel>. A panel opens in which
must be entered the name of the remote domain as well as the password assigned to that trust.
</para>

</sect2>

<sect2>
<title>Interdomain Trust Facilities</title>


<para>
<indexterm significance="normal"><primary>two-way trust</primary></indexterm>
<indexterm significance="normal"><primary>trust relationship</primary></indexterm>
<indexterm significance="normal"><primary>trust established</primary></indexterm>
<indexterm significance="normal"><primary>one-way trust</primary></indexterm>
<indexterm significance="normal"><primary>Windows NT4 domains</primary></indexterm>
<indexterm significance="normal"><primary>Interdomain Trusts</primary><secondary>Facilities</secondary></indexterm>
A two-way trust relationship is created when two one-way trusts are created, one in each direction.
Where a one-way trust has been established between two MS Windows NT4 domains (let's call them
DomA and DomB), the following facilities are created:
</para>

<figure id="trusts1" float="0">
	<title>Trusts overview.</title>
	<mediaobject><imageobject role="html"><imagedata fileref="images/trusts1.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/trusts1.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/trusts1"/></imageobject></mediaobject>
</figure>

<itemizedlist>
	<listitem><para>
	DomA (completes the trust connection) <parameter moreinfo="none">Trusts</parameter> DomB.
	</para></listitem>

	<listitem><para>
	DomA is the <parameter moreinfo="none">Trusting</parameter> domain.
	</para></listitem>

	<listitem><para>
	DomB is the <parameter moreinfo="none">Trusted</parameter> domain (originates the trust account).
	</para></listitem>

	<listitem><para>
	Users in DomB can access resources in DomA.
	</para></listitem>

	<listitem><para>
	Users in DomA cannot access resources in DomB.
	</para></listitem>

	<listitem><para>
	Global groups from DomB can be used in DomA.
	</para></listitem>

	<listitem><para>
	Global groups from DomA cannot be used in DomB.
	</para></listitem>

	<listitem><para>
	DomB does appear in the logon dialog box on client workstations in DomA.
	</para></listitem>

	<listitem><para>
	DomA does not appear in the logon dialog box on client workstations in DomB.
	</para></listitem>
</itemizedlist>

<itemizedlist>
	<listitem><para>
	Users and groups in a trusting domain cannot be granted rights, permissions, or access
	to a trusted domain.
	</para></listitem>

	<listitem><para>
	The trusting domain can access and use accounts (users/global groups) in the
	trusted domain.
	</para></listitem>

	<listitem><para>
	Administrators of the trusted domain can be granted administrative rights in the 
	trusting domain.
	</para></listitem>

	<listitem><para>
	Users in a trusted domain can be given rights and privileges in the trusting
	domain.
	</para></listitem>

	<listitem><para>
	Trusted domain global groups can be given rights and permissions in the trusting
	domain.
	</para></listitem>

	<listitem><para>
	Global groups from the trusted domain can be made members in local groups on
	MS Windows domain member machines.
	</para></listitem>
</itemizedlist>

</sect2>

</sect1>

<sect1>
<title>Configuring Samba NT-Style Domain Trusts</title>

<para>
<indexterm significance="normal"><primary>interdomain trust</primary></indexterm>
This description is meant to be a fairly short introduction about how to set up a Samba server so
that it can participate in interdomain trust relationships. Trust relationship support in Samba
is at an early stage, so do not be surprised if something does not function as it should.
</para>

<para>
<indexterm significance="normal"><primary>peer domain</primary></indexterm>
<indexterm significance="normal"><primary>trust relationship</primary></indexterm>
<indexterm significance="normal"><primary>Windows NT4 Server</primary></indexterm>
<indexterm significance="normal"><primary>between domains</primary></indexterm>
Each of the procedures described next assumes the peer domain in the trust relationship is controlled by a
Windows NT4 server. However, the remote end could just as well be another Samba-3  domain. It can be clearly
seen, after reading this document, that combining Samba-specific parts of what's written in the following
sections leads to trust between domains in a purely Samba environment.
</para>

<sect2 id="samba-trusted-domain">
<title>Samba as the Trusted Domain</title>

<para>
<indexterm significance="normal"><primary>trusted party</primary></indexterm>
<indexterm significance="normal"><primary>special account</primary></indexterm>
<indexterm significance="normal"><primary>trusting party</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
In order to set the Samba PDC to be the trusted party of the relationship, you first need
to create a special account for the domain that will be the trusting party. To do that,
you can use the <literal>smbpasswd</literal> utility. Creating the trusted domain account is 
similar to creating a trusted machine account. Suppose, your domain is
called SAMBA, and the remote domain is called RUMBA. The first step
will be to issue this command from your favorite shell:
</para>

<para>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> <userinput moreinfo="none">smbpasswd -a -i rumba</userinput>
New SMB password: <userinput moreinfo="none">XXXXXXXX</userinput>
Retype SMB password: <userinput moreinfo="none">XXXXXXXX</userinput>
Added user rumba$
</screen>

where <option>-a</option> means to add a new account into the
passdb database and <option>-i</option> means to <quote>create this
account with the Interdomain trust flag</quote>.
</para>

<para>
<indexterm significance="normal"><primary>account name</primary></indexterm>
<indexterm significance="normal"><primary>remote domain</primary></indexterm>
<indexterm significance="normal"><primary>password database</primary></indexterm>
<indexterm significance="normal"><primary>/etc/passwd</primary></indexterm>
The account name will be <quote>rumba$</quote> (the name of the remote domain).
If this fails, you should check that the trust account has been added to the system
password database (<filename moreinfo="none">/etc/passwd</filename>). If it has not been added, you
can add it manually and then repeat the previous step.
</para>

<para>
<indexterm significance="normal"><primary>password</primary></indexterm>
<indexterm significance="normal"><primary>new account</primary></indexterm>
<indexterm significance="normal"><primary>confirm the trust</primary></indexterm>
<indexterm significance="normal"><primary>Windows NT Server</primary></indexterm>
After issuing this command, you will be asked to enter the password for the account. You can use any password
you want, but be aware that Windows NT will not change this password until 7 days following account creation.
After the command returns successfully, you can look at the entry for the new account (in the standard way as
appropriate for your configuration) and see that the account's name is really RUMBA$ and it has the
<quote>I</quote> flag set in the flags field. Now you are ready to confirm the trust by establishing it from
Windows NT Server.
</para>


<para>
<indexterm significance="normal"><primary>User Manager</primary></indexterm>
<indexterm significance="normal"><primary>trusted domain name</primary></indexterm>
<indexterm significance="normal"><primary>relationship password</primary></indexterm>
<indexterm significance="normal"><primary>remote domain</primary></indexterm>
<indexterm significance="normal"><primary>established</primary></indexterm>
Open <application moreinfo="none">User Manager for Domains</application> and from the <guimenu moreinfo="none">Policies</guimenu> menu, select
<guimenuitem moreinfo="none">Trust Relationships...</guimenuitem>.  Beside the <guilabel moreinfo="none">Trusted domains</guilabel> list box,
click the <guimenu moreinfo="none">Add...</guimenu> button. You will be prompted for the trusted domain name and the
relationship password. Type in SAMBA, as this is the name of the remote domain and the password used at the
time of account creation.  Click on <guibutton moreinfo="none">OK</guibutton> and, if everything went without incident, you
will see the <computeroutput moreinfo="none">Trusted domain relationship successfully established</computeroutput> message.
</para>

</sect2>
<sect2>
<title>Samba as the Trusting Domain</title>

<para>
<indexterm significance="normal"><primary>NT-controlled domain</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
This time activities are somewhat reversed. Again, we'll assume that your domain
controlled by the Samba PDC is called SAMBA and the NT-controlled domain is called RUMBA.
</para>

<para>
The very first step is to add an account for the SAMBA domain on RUMBA's PDC.
</para>


<para>
<indexterm significance="normal"><primary>User Manager</primary></indexterm>
<indexterm significance="normal"><primary>trusted domain</primary></indexterm>
<indexterm significance="normal"><primary>password</primary></indexterm>
Launch the <application moreinfo="none">Domain User Manager</application>, then from the menu select
<guimenu moreinfo="none">Policies</guimenu>, <guimenuitem moreinfo="none">Trust Relationships</guimenuitem>.
Now, next to the <guilabel moreinfo="none">Trusting Domains</guilabel> box, press the <guibutton moreinfo="none">Add</guibutton>
button and type in the name of the trusted domain (SAMBA) and the password to use in securing
the relationship.
</para>

<para>
<indexterm significance="normal"><primary>password</primary></indexterm>
<indexterm significance="normal"><primary>confirm the password</primary></indexterm>
The password can be arbitrarily chosen. It is easy to change the password from the Samba server whenever you
want. After you confirm the password, your account is ready for use. Now its Samba's turn.
</para>

<para>
Using your favorite shell while logged in as root, issue this command:
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>trustdom establish</tertiary></indexterm>
</para>

<para>
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">net rpc trustdom establish rumba</userinput>
</para>

<para>
<indexterm significance="normal"><primary>password</primary></indexterm>
<indexterm significance="normal"><primary>interdomain connection</primary></indexterm>
<indexterm significance="normal"><primary>ordinary connection</primary></indexterm>
You will be prompted for the password you just typed on your Windows NT4 Server box.
An error message, <literal moreinfo="none">"NT_STATUS_NOLOGON_INTERDOMAIN_TRUST_ACCOUNT,"</literal>
that may be reported periodically is of no concern and may safely be ignored.
It means the password you gave is correct and the NT4 server says the account is ready for
interdomain connection and not for ordinary connection.  After that, be patient;
it can take a while (especially in large networks), but eventually you should see
the <literal moreinfo="none">Success</literal> message. Congratulations! Your trust
relationship has just been established.
</para>

<note><para>
You have to run this command as root because you must have write access to
the <filename moreinfo="none">secrets.tdb</filename> file.
</para></note>

</sect2>
</sect1>

<sect1>
<title>NT4-Style Domain Trusts with Windows 2000</title>

<para>
<indexterm significance="normal"><primary>trust relationship</primary></indexterm>
<indexterm significance="normal"><primary>Windows 2000 server</primary></indexterm>
<indexterm significance="normal"><primary>NT4-style</primary></indexterm>
<indexterm significance="normal"><primary>mixed mode</primary></indexterm>
Although <application moreinfo="none">Domain User Manager</application> is not present in Windows 2000, it is 
also possible to establish an NT4-style trust relationship with a Windows 2000 domain 
controller running in mixed mode as the trusting server. It should also be possible for 
Samba to trust a Windows 2000 server; however, more testing is still needed in this area.
</para>

<para>
<indexterm significance="normal"><primary>interdomain trust</primary></indexterm>
<indexterm significance="normal"><primary>trust account</primary></indexterm>
<indexterm significance="normal"><primary>not transitive</primary></indexterm>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
After <link linkend="samba-trusted-domain">creating the interdomain trust account on the Samba server</link>
as described previously, open <application moreinfo="none">Active Directory Domains and Trusts</application> on the AD
controller of the domain whose resources you wish Samba users to have access to. Remember that since NT4-style
trusts are not transitive, if you want your users to have access to multiple mixed-mode domains in your AD
forest, you will need to repeat this process for each of those domains. With <application moreinfo="none">Active Directory
domains and trusts</application> open, right-click on the name of the Active Directory domain that will trust
our Samba domain and choose <guimenuitem moreinfo="none">Properties</guimenuitem>, then click on the
<guilabel moreinfo="none">Trusts</guilabel> tab. In the upper part of the panel, you will see a list box labeled
<guilabel moreinfo="none">Domains trusted by this domain:</guilabel> and an <guilabel moreinfo="none">Add...</guilabel> button next to it.
Press this button and, just as with NT4, you will be prompted for the trusted domain name and the relationship
password. Press <emphasis>OK</emphasis> and after a moment, Active Directory will respond with
<computeroutput moreinfo="none">The trusted domain has been added and the trust has been verified.</computeroutput> Your
Samba users can now be granted access to resources in the AD domain.
</para>
</sect1>

<sect1>
<title>Common Errors</title>

<para>
Interdomain trust relationships should not be attempted on networks that are unstable
or that suffer regular outages. Network stability and integrity are key concerns with
distributed trusted domains.
</para>

<sect2>
<title>Browsing of Trusted Domain Fails</title>

<para>
<emphasis>Browsing from a machine in a trusted Windows 200x domain to a Windows 200x member of
a trusting Samba domain, I get the following error:</emphasis>
<screen format="linespecific">
The system detected a possible attempt to compromise security. Please
ensure that you can contact the server that authenticated you.
</screen>
</para>

<para>
<emphasis>The event logs on the box I'm trying to connect to have entries regarding group
policy not being applied because it is a member of a down-level domain.</emphasis>
</para>

<para>If there is a computer account in the Windows
200x domain for the machine in question, and it is disabled, this problem can
occur.  If there is no computer account (removed or never existed), or if that 
account is still intact (i.e., you just joined it to another domain), everything 
seems to be fine. By default, when you unjoin a domain (the Windows 200x 
domain), the computer tries to automatically disable the computer account in 
the domain.  If you are running as an account that has privileges to do this 
when you unjoin the machine, it is done; otherwise it is not done.
</para>

</sect2>

<sect2>
<title>Problems with LDAP ldapsam and Older Versions of smbldap-tools</title>

<para>
If you use the <literal>smbldap-useradd</literal> script to create a trust
account to set up interdomain trusts, the process of setting up the trust will
fail. The account that was created in the LDAP database will have an account
flags field that has <literal moreinfo="none">[W          ]</literal>, when it must have
<literal moreinfo="none">[I          ]</literal> for interdomain trusts to work.
</para>

<para>Here is a simple solution.
Create a machine account as follows:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> smbldap-useradd -w domain_name
</screen>
Then set the desired trust account password as shown here:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> smbldap-passwd domain_name\$
</screen>
Using a text editor, create the following file:
<screen format="linespecific">
dn: uid=domain_name$,ou=People,dc={your-domain},dc={your-top-level-domain}
changetype: modify
sambaAcctFlags: [I         ]
</screen>
Then apply the text file to the LDAP database as follows:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> ldapmodify -x -h localhost \
 -D "cn=Manager,dc={your-domain},dc={your-top-level-domain}" \
 -W -f /path-to/foobar
</screen>
Create a single-sided trust under the NT4 Domain User Manager, then execute:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> net rpc trustdom establish domain_name
</screen>
</para>

<para>
It works with Samba-3  and NT4 domains, and also with Samba-3 and Windows 200x ADS in mixed mode.
Both domain controllers, Samba and NT must have the same WINS server; otherwise,
the trust will never work. 
</para>

</sect2>

</sect1>

</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="msdfs">

<chapterinfo>
	<author>
		<firstname>Shirish</firstname><surname>Kalele</surname>
		<affiliation>
			<orgname>Samba Team &amp; Veritas Software</orgname>
			<address format="linespecific">
				<email>samba@samba.org</email>
			</address>
		</affiliation>
	</author>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
	
	<pubdate>12 Jul 2000</pubdate>
</chapterinfo>

<title>Hosting a Microsoft Distributed File System Tree</title>

<sect1>
<title>Features and Benefits</title>

	<para>
<indexterm significance="normal"><primary>distributed file system</primary><see>DFS</see></indexterm>
<indexterm significance="normal"><primary>physical locations</primary></indexterm>
<indexterm significance="normal"><primary>higher availability</primary></indexterm>
<indexterm significance="normal"><primary>load balancing</primary></indexterm>
<indexterm significance="normal"><primary>logical directories</primary></indexterm>
	The distributed file system (DFS) provides a means of separating the logical
	view of files and directories that users see from the actual physical locations
	of these resources on the network. It allows for higher availability, smoother
	storage expansion, load balancing, and so on.
	</para>

	<para>
<indexterm significance="normal"><primary>DFS</primary></indexterm>
<indexterm significance="normal"><primary>DFS tree</primary></indexterm>
<indexterm significance="normal"><primary>DFS-aware</primary></indexterm>
	For information about DFS, refer to the <ulink url="http://www.microsoft.com/NTServer/nts/downloads/winfeatures/NTSDistrFile/AdminGuide.asp">Microsoft
	documentation</ulink>.  This document explains how to host a DFS tree on a UNIX machine (for DFS-aware clients
	to browse) using Samba.
	</para>

	<para>
<indexterm significance="normal"><primary>DFS server</primary></indexterm>
<indexterm significance="normal"><primary>share-level</primary></indexterm>
<indexterm significance="normal"><primary>DFS junction</primary></indexterm>
<indexterm significance="normal"><primary>DFS-aware</primary></indexterm>
	A Samba server can be made a DFS server by setting the global Boolean <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="HOSTMSDFS" xlink:href="smb.conf.5.html#HOSTMSDFS">host msdfs</link>
	parameter in the <filename moreinfo="none">smb.conf</filename> file. You designate a share as a DFS root using the share-level Boolean
	<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="MSDFSROOT" xlink:href="smb.conf.5.html#MSDFSROOT">msdfs root</link> parameter. A DFS root directory on Samba hosts DFS links in the form of
	symbolic links that point to other servers. For example, a symbolic link
	<filename moreinfo="none">junction-&gt;msdfs:storage1\share1</filename> in the share directory acts as the DFS junction. When
	DFS-aware clients attempt to access the junction link, they are redirected to the storage location (in this
	case, <parameter moreinfo="none">\\storage1\share1</parameter>).
	</para>

	<para>
<indexterm significance="normal"><primary>DFS-aware</primary></indexterm>
<indexterm significance="normal"><primary>DFS tree</primary></indexterm>
<indexterm significance="normal"><primary>DFS links</primary></indexterm>
<indexterm significance="normal"><primary>DFS</primary></indexterm>
	DFS trees on Samba work with all DFS-aware clients ranging from Windows 95 to 200x.
	<link linkend="dfscfg">The following sample configuration</link> shows how to setup a DFS tree on a Samba server.
	In the <filename moreinfo="none">/export/dfsroot</filename> directory, you set up your DFS links to 
	other servers on the network.
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">cd /export/dfsroot</userinput>
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">chown root /export/dfsroot</userinput>
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">chmod 755 /export/dfsroot</userinput>
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">ln -s msdfs:storageA\\shareA linka</userinput>
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">ln -s msdfs:serverB\\share,serverC\\share linkb</userinput>
</screen>
</para>

<example id="dfscfg">
<title>smb.conf with DFS Configured</title>
<simplelist>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>netbios name</indexterm><parameter>netbios name = GANDALF</parameter></member>
<member><indexterm>host msdfs  </indexterm><parameter>host msdfs   = yes</parameter></member>

<member> </member><member><parameter>[dfs]</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /export/dfsroot</parameter></member>
<member><indexterm>msdfs root</indexterm><parameter>msdfs root = yes</parameter></member>
</simplelist>
</example>

	<para>
<indexterm significance="normal"><primary>DFS root</primary></indexterm>
<indexterm significance="normal"><primary>msdfs links</primary></indexterm>
<indexterm significance="normal"><primary>symbolic links</primary></indexterm>
	You should set up the permissions and ownership of the directory acting as the DFS root so that only
	designated users can create, delete, or modify the msdfs links. Also note that symlink names should be all
	lowercase. This limitation exists to have Samba avoid trying all the case combinations to get at the link
	name. Finally, set up the symbolic links to point to the network shares you want and start Samba.
	</para>

	<para>
<indexterm significance="normal"><primary>DFS-aware clients</primary></indexterm>
<indexterm significance="normal"><primary>DFS tree</primary></indexterm>
	Users on DFS-aware clients can now browse the DFS tree on the Samba server at
	<constant>\\samba\dfs</constant>. Accessing links linka or linkb (which appear as directories to the client)
	takes users directly to the appropriate shares on the network.
	</para>

</sect1>

<sect1>
<title>Common Errors</title>
	<itemizedlist>
		<listitem><para>Windows clients need to be rebooted 
		if a previously mounted non-DFS share is made a DFS 
		root, or vice versa. A better way is to introduce a 
		new share and make it the DFS root.</para>
		</listitem>
		
		<listitem><para>Currently, there's a restriction that msdfs 
		symlink names should all be lowercase.</para>
		</listitem>
		
		<listitem><para>For security purposes, the directory 
		acting as the root of the DFS tree should have ownership 
		and permissions set so only designated users can 
		modify the symbolic links in the directory.</para>
		</listitem>
	</itemizedlist>

	<sect2>
		<title>MSDFS UNIX Path Is Case-Critical</title>

		<para>
		A network administrator sent advice to the Samba mailing list
		after long sessions trying to determine why DFS was not working.
		His advice is worth noting.
		</para>

		<para><quote>
		I spent some time trying to figure out why my particular
		DFS root wasn't working. I noted in the documentation that
		the symlink should be in all lowercase. It should be
		amended that the entire path to the symlink should all be
		in lowercase as well.
		</quote></para>

		<para>
		<quote>For example, I had a share defined as such:</quote>
		<simplelist>
		<member> </member><member><parameter>[pub]</parameter></member>
		<member><indexterm>path</indexterm><parameter>path = /export/home/Shares/public_share</parameter></member>
		<member><indexterm>msdfs root</indexterm><parameter>msdfs root = yes</parameter></member>
		</simplelist>
		<quote>and I could not make my Windows 9x/Me (with the dfs client installed) follow this symlink:</quote>
		<screen format="linespecific">
		damage1 -&gt; msdfs:damage\test-share
		</screen>
		</para>

		<para>
		<quote>Running a debug level of 10 reveals:</quote>
		<programlisting format="linespecific">
		[2003/08/20 11:40:33, 5] msdfs/msdfs.c:is_msdfs_link(176)
		  is_msdfs_link: /export/home/shares/public_share/* does not exist.
		</programlisting>
		<quote>Curious. So I changed the directory name from <constant>.../Shares/...</constant> to
		<constant>.../shares/...</constant> (along with my service definition) and it worked!</quote>
		</para>

	</sect2>

</sect1>

</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="classicalprinting">

<chapterinfo>
	<author>
		<firstname>Kurt</firstname><surname>Pfeifle</surname>
		<affiliation>
			<orgname>Danka Deutschland GmbH</orgname>
			<address format="linespecific"><email>kpfeifle@danka.de</email></address>
		</affiliation>
	</author>
	<author>
<firstname>Gerald</firstname><surname>Carter</surname><othername>(Jerry)</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jerry@samba.org</email></address>
</affiliation></author>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
	<pubdate>May 31, 2003</pubdate>
</chapterinfo>

<title>Classical Printing Support</title>

<sect1>
<title>Features and Benefits</title>

<para>
<indexterm significance="normal"><primary>mission-critical</primary></indexterm>
Printing is often a mission-critical service for the users. Samba can provide this service reliably and
seamlessly for a client network consisting of Windows workstations.
</para>

<para>
<indexterm significance="normal"><primary>print service</primary></indexterm>
<indexterm significance="normal"><primary>domain member server</primary></indexterm>
<indexterm significance="normal"><primary>standalone server</primary></indexterm>
<indexterm significance="normal"><primary>file serving</primary></indexterm>
<indexterm significance="normal"><primary>dedicated print server</primary></indexterm>
<indexterm significance="normal"><primary>print server</primary></indexterm>
<indexterm significance="normal"><primary>printing support</primary></indexterm>
<indexterm significance="normal"><primary>Point'n'Print</primary></indexterm>
<indexterm significance="normal"><primary>Add Printer Wizard</primary></indexterm>
<indexterm significance="normal"><primary>upload drivers</primary></indexterm>
<indexterm significance="normal"><primary>manage drivers</primary></indexterm>
<indexterm significance="normal"><primary>install drivers</primary></indexterm>
<indexterm significance="normal"><primary>print accounting</primary></indexterm>
<indexterm significance="normal"><primary>Common UNIX Printing System</primary><see>CUPS</see></indexterm>
A Samba print service may be run on a standalone or domain member server, side by side with file serving
functions, or on a dedicated print server.  It can be made as tightly or as loosely secured as needs dictate.
Configurations may be simple or complex. Available authentication schemes are essentially the same as
described for file services in previous chapters. Overall, Samba's printing support is now able to replace an
NT or Windows 2000 print server full-square, with additional benefits in many cases. Clients may download and
install drivers and printers through their familiar <literal moreinfo="none">Point'n'Print</literal> mechanism. Printer
installations executed by <literal moreinfo="none">Logon Scripts</literal> are no problem. Administrators can upload and manage
drivers to be used by clients through the familiar <literal moreinfo="none">Add Printer Wizard</literal>. As an additional
benefit, driver and printer management may be run from the command line or through scripts, making it more
efficient in case of large numbers of printers. If a central accounting of print jobs (tracking every single
page and supplying the raw data for all sorts of statistical reports) is required, this function is best
supported by the newer Common UNIX Printing System (CUPS) as the print subsystem underneath the Samba hood.
</para>

<para>
<indexterm significance="normal"><primary>BSD</primary></indexterm>
<indexterm significance="normal"><primary>CUPS</primary></indexterm>
This chapter outlines the fundamentals of Samba printing as implemented by the more traditional UNIX
BSD- and System V-style printing systems. Much of the information in this chapter applies also to CUPS.  If
you use CUPS, you may be tempted to jump to the next chapter, but you will certainly miss a few things if you
do. For further information refer to <link linkend="CUPS-printing">CUPS Printing Support</link>.
</para>

<note>
<para>
<indexterm significance="normal"><primary>Windows XP Professional</primary></indexterm>
<indexterm significance="normal"><primary>Windows 200x/XP</primary></indexterm>
<indexterm significance="normal"><primary>Windows NT4</primary></indexterm>
Most of the following examples have been verified on Windows XP Professional clients. Where this document
describes the responses to commands given, bear in mind that Windows 200x/XP clients are quite similar but may
differ in minor details. Windows NT4 is somewhat different again.
</para>
</note>

</sect1>

<sect1>
<title>Technical Introduction</title>

<para>
<indexterm significance="normal"><primary>printing support</primary></indexterm>
<indexterm significance="normal"><primary>print subsystem</primary></indexterm>
<indexterm significance="normal"><primary>printing system</primary></indexterm>
Samba's printing support always relies on the installed print subsystem of the UNIX OS it runs on. Samba is a
<literal moreinfo="none">middleman.</literal> It takes print files from Windows (or other SMB) clients and passes them to the real
printing system for further processing; therefore, it needs to communicate with both sides: the Windows print
clients and the UNIX printing system. Hence, we must differentiate between the various client OS types, each
of which behave differently, as well as the various UNIX print subsystems, which themselves have different
features and are accessed differently.
</para>

<para>
<indexterm significance="normal"><primary>UNIX printing</primary></indexterm>
<indexterm significance="normal"><primary>CUPS</primary></indexterm>
This chapter deals with the traditional way of UNIX printing. The next chapter covers in great detail the more
modern CUPS.
</para>

<important><para>
<indexterm significance="normal"><primary>CUPS</primary></indexterm>
CUPS users, be warned: do not just jump on to the next chapter. You might miss important information only found here!
</para></important>

<para>
<indexterm significance="normal"><primary>print configuration</primary></indexterm>
<indexterm significance="normal"><primary>problematic print</primary></indexterm>
<indexterm significance="normal"><primary>print processing</primary></indexterm>
<indexterm significance="normal"><primary>print filtering</primary></indexterm>
It is apparent from postings on the Samba mailing list that print configuration is one of the most problematic
aspects of Samba administration today. Many new Samba administrators have the impression that Samba performs
some sort of print processing. Rest assured, Samba does not perform any type of print processing. It does not
do any form of print filtering.
</para>

<para>
<indexterm significance="normal"><primary>data stream</primary></indexterm>
<indexterm significance="normal"><primary>local spool area</primary></indexterm>
<indexterm significance="normal"><primary>spooled file</primary></indexterm>
<indexterm significance="normal"><primary>local system printing</primary></indexterm>
Samba obtains from its clients a data stream (print job) that it spools to a local spool area. When the entire
print job has been received, Samba invokes a local UNIX/Linux print command and passes the spooled file to it.
It is up to the local system printing subsystems to correctly process the print job and to submit it to the
printer.
</para>

<sect2>
<title>Client to Samba Print Job Processing</title>

<para>
Successful printing from a Windows client via a Samba print server to a UNIX
printer involves six (potentially seven) stages:
</para>

<orderedlist inheritnum="ignore" continuation="restarts">
	<listitem><para>Windows opens a connection to the printer share.</para></listitem>

	<listitem><para>Samba must authenticate the user.</para></listitem>

	<listitem><para>Windows sends a copy of the print file over the network
	into Samba's spooling area.</para></listitem>

	<listitem><para>Windows closes the connection.</para></listitem>

	<listitem><para>Samba invokes the print command to hand the file over
	to the UNIX print subsystem's spooling area.</para></listitem>

	<listitem><para>The UNIX print subsystem processes the print job.</para></listitem>

	<listitem><para>The print file may need to be explicitly deleted
	from the Samba spooling area. This item depends on your print spooler
	configuration settings.</para></listitem>
</orderedlist>
</sect2>

<sect2>
<title>Printing-Related Configuration Parameters</title>

<para>
<indexterm significance="normal"><primary>global-level</primary></indexterm>
<indexterm significance="normal"><primary>service-level</primary></indexterm>
<indexterm significance="normal"><primary>printing behavior</primary></indexterm>
There are a number of configuration parameters to control Samba's printing behavior. Please refer to the man
page for <filename moreinfo="none">smb.conf</filename> for an overview of these. As with other parameters, there are global-level (tagged with a
<emphasis>G</emphasis> in the listings) and service-level (<emphasis>S</emphasis>) parameters.
</para>

<variablelist>
	<varlistentry><term>Global Parameters</term>
		<listitem><para> These <emphasis>may not</emphasis> go into
		individual share definitions. If they go in by error,
		the <literal>testparm</literal> utility can discover this
		(if you run it) and tell you so.
		</para></listitem>
	</varlistentry>

	<varlistentry><term>Service-Level Parameters</term>
		<listitem><para> These may be specified in the
		<parameter>[global]</parameter> section of <filename moreinfo="none">smb.conf</filename>.
		In this case they define the default behavior of all individual
		or service-level shares (provided they do not have a different
		setting defined for the same parameter, thus overriding the
		global default).
		</para></listitem>
	</varlistentry>
</variablelist>
</sect2>

</sect1>

<sect1>
<title>Simple Print Configuration</title>

<para>
<indexterm significance="normal"><primary>BSD Printing</primary></indexterm>
<indexterm significance="normal"><primary>simple printing</primary></indexterm>
<indexterm significance="normal"><primary>enables clients to print</primary></indexterm>
<indexterm significance="normal"><primary>print environment</primary></indexterm>
<link linkend="simpleprc">Simple Configuration with BSD Printing</link> shows a simple printing configuration.
If you compare this with your own, you may find additional parameters that have been preconfigured by your OS
vendor. Following is a discussion and explanation of the parameters. This example does not use many
parameters.  However, in many environments these are enough to provide a valid <filename moreinfo="none">smb.conf</filename> file that enables
all clients to print.
</para>

<example id="simpleprc">
<title>Simple Configuration with BSD Printing</title>
<simplelist>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>printing</indexterm><parameter>printing = bsd</parameter></member>
<member><indexterm>load printers</indexterm><parameter>load printers = yes</parameter></member>

<member> </member><member><parameter>[printers]</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /var/spool/samba</parameter></member>
<member><indexterm>printable</indexterm><parameter>printable = yes</parameter></member>
<member><indexterm>public</indexterm><parameter>public = yes</parameter></member>
<member><indexterm>writable</indexterm><parameter>writable = no</parameter></member>
</simplelist>
</example>

<para>
<indexterm significance="normal"><primary>testparm</primary></indexterm>
<indexterm significance="normal"><primary>misconfigured settings</primary></indexterm>
<indexterm significance="normal"><primary>pager program</primary></indexterm>
This is only an example configuration. Samba assigns default values to all configuration parameters. The
defaults are conservative and sensible. When a parameter is specified in the <filename moreinfo="none">smb.conf</filename> file, this overwrites
the default value. The <literal>testparm</literal> utility when run as root is capable of reporting all
settings, both default as well as <filename moreinfo="none">smb.conf</filename> file settings. <literal>Testparm</literal> gives warnings for all
misconfigured settings. The complete output is easily 360 lines and more, so you may want to pipe it through a
pager program.
</para>

<para>  
<indexterm significance="normal"><primary>configuration syntax</primary></indexterm>
<indexterm significance="normal"><primary>syntax tolerates spelling errors</primary></indexterm>
<indexterm significance="normal"><primary>case-insensitive</primary></indexterm>
The syntax for the configuration file is easy to grasp. You should know that  is not very picky about its
syntax. As has been explained elsewhere in this book, Samba tolerates some spelling errors (such as
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="BROWSEABLE" xlink:href="smb.conf.5.html#BROWSEABLE">browseable</link> instead of <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="BROWSABLE" xlink:href="smb.conf.5.html#BROWSABLE">browsable</link>), and spelling is
case-insensitive. It is permissible to use <parameter moreinfo="none">Yes/No</parameter> or <parameter moreinfo="none">True/False</parameter>
for Boolean settings. Lists of names may be separated by commas, spaces, or tabs.
</para>

<sect2>
<title>Verifying Configuration with <literal>testparm</literal></title>

<para>
<indexterm significance="normal"><primary>printing-related settings</primary></indexterm>
<indexterm significance="normal"><primary>lp</primary></indexterm>
<indexterm significance="normal"><primary>print</primary></indexterm>
<indexterm significance="normal"><primary>spool</primary></indexterm>
<indexterm significance="normal"><primary>driver</primary></indexterm>
<indexterm significance="normal"><primary>ports</primary></indexterm>
<indexterm significance="normal"><primary>testparm</primary></indexterm>
<indexterm significance="normal"><primary>smbd</primary></indexterm>
<indexterm significance="normal"><primary>print configuration</primary></indexterm>
<indexterm significance="normal"><primary>printer shares </primary></indexterm>
<indexterm significance="normal"><primary>spooling path</primary></indexterm>
To see all (or at least most) printing-related settings in Samba, including the implicitly used ones, try the
command outlined below. This command greps for all occurrences of <constant>lp</constant>,
<constant>print</constant>, <constant>spool</constant>, <constant>driver</constant>,
<constant>ports</constant>, and <constant>[</constant> in <literal>testparm</literal>'s output. This provides
a convenient overview of the running <literal>smbd</literal> print configuration. This command does not show
individually created printer shares or the spooling paths they may use. Here is the output of my Samba setup,
with settings shown in <link linkend="simpleprc">the example above</link>:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">testparm -s -v | egrep "(lp|print|spool|driver|ports|\[)"</userinput>
 Load smb config files from /etc/samba/smb.conf
 Processing section "[homes]"
 Processing section "[printers]"
 
 [global]
        smb ports = 139 445
        lpq cache time = 10
        load printers = Yes
        printcap name = /etc/printcap
        disable spoolss = No
        enumports command =
        addprinter command = 
        deleteprinter command = 
        show add printer wizard = Yes
        os2 driver map =
        printer admin =
        min print space = 0
        max print jobs = 1000
        printable = No
        printing = bsd
        print command = lpr -r -P'%p' %s
        lpq command = lpq -P'%p'
        lprm command = lprm -P'%p' %j
        lppause command =
        lpresume command =
        printer name =
        use client driver = No

 [homes]

 [printers]
        path = /var/spool/samba
        printable = Yes
</screen>
</para>

<para>
You can easily verify which settings were implicitly added by Samba's default behavior. <emphasis>Remember: it
may be important in your future dealings with Samba.</emphasis>
</para>

<note><para>
The <literal>testparm</literal> in Samba-3 behaves differently from that in 2.2.x: used without the
<quote>-v</quote> switch, it only shows you the settings actually written into! To see the complete
configuration used, add the <quote>-v</quote> parameter to testparm.
</para></note>

</sect2>

<sect2>
<title>Rapid Configuration Validation</title>

<para> 
<indexterm significance="normal"><primary>troubleshoot</primary></indexterm>
<indexterm significance="normal"><primary>testparm</primary></indexterm>
<indexterm significance="normal"><primary>parameters</primary></indexterm>
<indexterm significance="normal"><primary>verify</primary></indexterm>
Should you need to troubleshoot at any stage, please always come back to this point first and verify if
<literal>testparm</literal> shows the parameters you expect. To give you a warning from personal experience,
try to just comment out the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOADPRINTERS" xlink:href="smb.conf.5.html#LOADPRINTERS">load printers</link> parameter. If your 2.2.x system behaves like
mine, you'll see this:
</para>

<para><screen format="linespecific">
<prompt moreinfo="none">root# </prompt>grep "load printers" /etc/samba/smb.conf
        #  load printers = Yes
        # This setting is commented out!!
 
<prompt moreinfo="none">root# </prompt>testparm -v /etc/samba/smb.conf | egrep "(load printers)"
        load printers = Yes
</screen></para>

<para>
<indexterm significance="normal"><primary>commenting out setting</primary></indexterm>
<indexterm significance="normal"><primary>publishing printers</primary></indexterm>
I assumed that commenting out of this setting should prevent Samba from
publishing my printers, but it still did. It took some time to figure out
the reason. But I am no longer fooled ... at least not by this.
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">grep -A1 "load printers" /etc/samba/smb.conf</userinput>
        load printers = No
        # The above setting is what I want!
        #  load printers = Yes
        # This setting is commented out!

<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">testparm -s -v smb.conf.simpleprinting | egrep "(load printers)"</userinput>
        load printers = No
</screen></para>

<para>
<indexterm significance="normal"><primary>explicitly set</primary></indexterm>
Only when the parameter is explicitly set to <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOADPRINTERS" xlink:href="smb.conf.5.html#LOADPRINTERS">load printers = No</link> would
Samba conform with my intentions. So, my strong advice is:
</para>

<itemizedlist>
	<listitem><para>Never rely on commented-out parameters.</para></listitem>

	<listitem><para>Always set parameters explicitly as you intend them to
	behave.</para></listitem>

	<listitem><para>Use <literal>testparm</literal> to uncover hidden
	settings that might not reflect your intentions.</para></listitem>
</itemizedlist>

<para>
The following is the most minimal configuration file:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">cat /etc/samba/smb.conf-minimal</userinput>
        [printers]
</screen></para>

<para>
<indexterm significance="normal"><primary>testparm</primary></indexterm>
<indexterm significance="normal"><primary>smbd</primary></indexterm>
This example should show that you can use <literal>testparm</literal> to test any Samba configuration file.
Actually, we encourage you <emphasis>not</emphasis> to change your working system (unless you know exactly
what you are doing). Don't rely on the assumption that changes will only take effect after you restart smbd!
This is not the case. Samba rereads it every 60 seconds and on each new client connection. You might have to
face changes for your production clients that you didn't intend to apply. You will now note a few more
interesting things; <literal>testparm</literal> is useful to identify what the Samba print configuration would
be if you used this minimalistic configuration. Here is what you can expect to find:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">testparm -v smb.conf-minimal | egrep "(print|lpq|spool|driver|ports|[)"</userinput>
 Processing section "[printers]"
 WARNING: [printers] service MUST be printable!
 No path in service printers - using /tmp

        lpq cache time = 10
        load printers = Yes
        printcap name = /etc/printcap
        disable spoolss = No
        enumports command =
        addprinter command =
        deleteprinter command =
        show add printer wizard = Yes
        os2 driver map =
        printer admin =
        min print space = 0
        max print jobs = 1000
        printable = No
        printing = bsd
        print command = lpr -r -P%p %s
        lpq command = lpq -P%p
        printer name =
        use client driver = No

 [printers]
        printable = Yes
</screen></para>

<para>
<literal>testparm</literal> issued two warnings:
</para>

<itemizedlist>
	<listitem><para>We did not specify the <parameter>[printers]</parameter> section as printable.</para></listitem>
	<listitem><para>We did not tell Samba which spool directory to use.</para></listitem>
</itemizedlist>

<para>
<indexterm significance="normal"><primary>compile-time options</primary></indexterm>
<indexterm significance="normal"><primary/></indexterm>
<indexterm significance="normal"><primary/></indexterm>
<indexterm significance="normal"><primary/></indexterm>
However, this was not fatal, and Samba will default to values that will work. Please, do not rely on this and
do not use this example. This was included to encourage you to be careful to design and specify your setup to
do precisely what you require. The outcome on your system may vary for some parameters given, since Samba may
have been built with  different compile-time options. <emphasis>Warning:</emphasis> do not put a comment sign
<emphasis>at the end</emphasis> of a valid line. It will cause the parameter to be ignored (just as if you had
put the comment sign at the front). At first I regarded this as a bug in my Samba versions. But the man page
clearly says: <literal moreinfo="none">Internal whitespace in a parameter value is retained verbatim.</literal> This means
that a line consisting of, for example,
<simplelist>
<member># This defines LPRng as the printing system</member>
<member><indexterm>printing</indexterm><parameter>printing =  lprng</parameter></member>
</simplelist>
</para>

<para>
will regard the whole of the string after the <literal moreinfo="none">=</literal> sign as the value you want to define. This
is an invalid value that will be ignored, and a default value will be used in its place.
</para>

</sect2>

</sect1>

<sect1>
<title>Extended Printing Configuration</title>

<para>
<indexterm significance="normal"><primary>Extended BSD Printing</primary></indexterm>
<indexterm significance="normal"><primary>BSD-style printing</primary></indexterm>
<indexterm significance="normal"><primary>CUPS</primary></indexterm>
<indexterm significance="normal"><primary>testparm</primary></indexterm>
<link linkend="extbsdpr">Extended BSD Printing Configuration</link> shows a more verbose configuration for
print-related settings in a BSD-style printing environment. What follows is a discussion and explanation of
the various parameters. We chose to use BSD-style printing here because it is still the most commonly used
system on legacy UNIX/Linux installations. New installations predominantly use CUPS, which is discussed in a
separate chapter. The example explicitly names many parameters that do not need to be specified because they
are set by default. You could use a much leaner <filename moreinfo="none">smb.conf</filename> file, or you can use <literal>testparm</literal> or
<literal>SWAT</literal> to optimize the <filename moreinfo="none">smb.conf</filename> file to remove all parameters that are set at default.
</para>

<example id="extbsdpr">
<title>Extended BSD Printing Configuration</title>
<simplelist>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>printing</indexterm><parameter>printing = bsd</parameter></member>
<member><indexterm>load printers</indexterm><parameter>load printers = yes</parameter></member>
<member><indexterm>show add printer wizard</indexterm><parameter>show add printer wizard = yes</parameter></member>
<member><indexterm>printcap name</indexterm><parameter>printcap name = /etc/printcap</parameter></member>
<member><indexterm>printer admin</indexterm><parameter>printer admin = @ntadmin, root</parameter></member>
<member><indexterm>max print jobs</indexterm><parameter>max print jobs = 100</parameter></member>
<member><indexterm>lpq cache time</indexterm><parameter>lpq cache time = 20</parameter></member>
<member><indexterm>use client driver</indexterm><parameter>use client driver = no</parameter></member>

<member> </member><member><parameter>[printers]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = All Printers</parameter></member>
<member><indexterm>printable</indexterm><parameter>printable = yes</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /var/spool/samba</parameter></member>
<member><indexterm>browseable</indexterm><parameter>browseable = no</parameter></member>
<member><indexterm>guest ok</indexterm><parameter>guest ok = yes</parameter></member>
<member><indexterm>public</indexterm><parameter>public = yes</parameter></member>
<member><indexterm>read only</indexterm><parameter>read only = yes</parameter></member>
<member><indexterm>writable</indexterm><parameter>writable = no       </parameter></member>

<member> </member><member><parameter>[my_printer_name]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = Printer with Restricted Access</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /var/spool/samba_my_printer</parameter></member>
<member><indexterm>printer admin</indexterm><parameter>printer admin = kurt</parameter></member>
<member><indexterm>browseable</indexterm><parameter>browseable = yes</parameter></member>
<member><indexterm>printable</indexterm><parameter>printable = yes</parameter></member>
<member><indexterm>writable</indexterm><parameter>writable = no</parameter></member>
<member><indexterm>hosts allow</indexterm><parameter>hosts allow = 0.0.0.0</parameter></member>
<member><indexterm>hosts deny</indexterm><parameter>hosts deny = turbo_xp, 10.160.50.23, 10.160.51.60</parameter></member>
<member><indexterm>guest ok</indexterm><parameter>guest ok = no</parameter></member>
</simplelist></example>

<para>
<indexterm significance="normal"><primary/></indexterm>
<indexterm significance="normal"><primary/></indexterm>
<indexterm significance="normal"><primary/></indexterm>
This is an example configuration. You may not find all the settings that are in the configuration file that
was provided by the OS vendor. Samba configuration parameters, if not explicitly set, default to a sensible
value.  To see all settings, as <constant>root</constant> use the <literal>testparm</literal> utility.
<literal>testparm</literal> gives warnings for misconfigured settings.
</para>

<sect2>
<title>Detailed Explanation Settings</title>

<para>
The following is a discussion of the settings from <link linkend="extbsdpr">Extended BSD Printing
Configuration</link> <link linkend="extbsdpr">Extended BSD Printing Configuration</link>.
</para>

<sect3>
<title>The [global] Section</title>

<para>
<indexterm significance="normal"><primary>global section</primary></indexterm>
<indexterm significance="normal"><primary>special sections</primary></indexterm>
<indexterm significance="normal"><primary>individual section</primary></indexterm>
<indexterm significance="normal"><primary>share</primary></indexterm>
The <parameter>[global]</parameter> section is one of four special sections (along with <parameter>[homes]</parameter>, <parameter>[printers]</parameter>, and <parameter>[print$]</parameter>). The
<parameter>[global]</parameter> contains all parameters that apply to the server as a whole. It is the place
for parameters that have only a global meaning. It may also contain service-level parameters that define
default settings for all other sections and shares. This way you can simplify the configuration and avoid
setting the same value repeatedly. (Within each individual section or share, you may, however, override these
globally set share settings and specify other values).
</para>

<variablelist>
	<varlistentry><term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTING" xlink:href="smb.conf.5.html#PRINTING">printing = bsd </link></term>
		<listitem><para>
<indexterm significance="normal"><primary>default print commands</primary></indexterm>
<indexterm significance="normal"><primary>RFC 1179</primary></indexterm>
<indexterm significance="normal"><primary>printing</primary></indexterm>
<indexterm significance="normal"><primary>CUPS</primary></indexterm>
<indexterm significance="normal"><primary>LPD</primary></indexterm>
<indexterm significance="normal"><primary>LPRNG</primary></indexterm>
<indexterm significance="normal"><primary>SYSV</primary></indexterm>
<indexterm significance="normal"><primary>HPUX</primary></indexterm>
<indexterm significance="normal"><primary>AIX</primary></indexterm>
<indexterm significance="normal"><primary>QNX</primary></indexterm>
<indexterm significance="normal"><primary>PLP</primary></indexterm>
<indexterm significance="normal"><primary>queue control</primary></indexterm>
		Causes Samba to use default print commands applicable for the BSD (also known as RFC 1179 style or LPR/LPD)
		printing system. In general, the <parameter moreinfo="none">printing</parameter> parameter informs Samba about the print
		subsystem it should expect. Samba supports CUPS, LPD, LPRNG, SYSV, HPUX, AIX, QNX, and PLP. Each of these
		systems defaults to a different <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTCOMMAND" xlink:href="smb.conf.5.html#PRINTCOMMAND">print command</link> (and other queue control commands).
		</para>

		<caution><para>
<indexterm significance="normal"><primary>service-level</primary></indexterm>
<indexterm significance="normal"><primary>SOFTQ printing system</primary></indexterm>
		The <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTING" xlink:href="smb.conf.5.html#PRINTING">printing</link> parameter is normally a service-level parameter. Since it is included
		here in the <parameter>[global]</parameter> section, it will take effect for all printer shares that are not
		defined differently. Samba-3 no longer supports the SOFTQ printing system.
		</para></caution>
	</listitem></varlistentry>

	<varlistentry><term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOADPRINTERS" xlink:href="smb.conf.5.html#LOADPRINTERS">load printers = yes </link></term>
		<listitem><para>
<indexterm significance="normal"><primary>printer shares</primary></indexterm>
<indexterm significance="normal"><primary>printcap</primary></indexterm>
<indexterm significance="normal"><primary>separate shares</primary></indexterm>
<indexterm significance="normal"><primary>UNIX printer</primary></indexterm>
		Tells Samba to create automatically all available printer shares. Available printer shares are discovered by
		scanning the printcap file. All created printer shares are also loaded for browsing. If you use this
		parameter, you do not need to specify separate shares for each printer. Each automatically created printer
		share will clone the configuration options found in the <parameter>[printers]</parameter> section. (The
		<parameter moreinfo="none">load printers = no</parameter> setting will allow you to specify each UNIX printer you want to
		share separately, leaving out some you do not want to be publicly visible and available).
		</para>
	</listitem></varlistentry>

	<varlistentry><term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SHOWADDPRINTERWIZARD" xlink:href="smb.conf.5.html#SHOWADDPRINTERWIZARD">show add printer wizard = yes </link></term>
		<listitem><para>
<indexterm significance="normal"><primary>Add Printer Wizard</primary></indexterm>
<indexterm significance="normal"><primary>Printers</primary></indexterm>
<indexterm significance="normal"><primary>Network Neighborhood</primary></indexterm>
<indexterm significance="normal"><primary>net view</primary></indexterm>
<indexterm significance="normal"><primary>uploaded driver</primary></indexterm>
		Setting is normally enabled by default (even if the parameter is not specified in <filename moreinfo="none">smb.conf</filename>).  It causes the
		<guiicon moreinfo="none">Add Printer Wizard</guiicon> icon to appear in the <guiicon moreinfo="none">Printers</guiicon> folder of the Samba
		host's share listing (as shown in <guiicon moreinfo="none">Network Neighborhood</guiicon> or by the <literal>net
		view</literal> command). To disable it, you need to explicitly set it to <constant>no</constant> (commenting
		it out will not suffice). The <parameter moreinfo="none">Add Printer Wizard</parameter> lets you upload a printer driver to
		the <parameter>[print$]</parameter> share and associate it with a printer (if the respective queue exists
		before the action), or exchange a printer's driver for any other previously uploaded driver.
		</para>
	</listitem></varlistentry>

	<varlistentry><term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="MAXPRINTJOBS" xlink:href="smb.conf.5.html#MAXPRINTJOBS">max print jobs = 100 </link></term>
		<listitem><para>
<indexterm significance="normal"><primary>print jobs</primary></indexterm>
		Sets the upper limit to 100 print jobs being active on the Samba server at any one time. Should a client
		submit a job that exceeds this number, a "no more space available on server" type of error message will be
		returned by Samba to the client. A setting of zero (the default) means there is <emphasis>no</emphasis> limit
		at all.
		</para>
	</listitem></varlistentry>

	<varlistentry><term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTCAPNAME" xlink:href="smb.conf.5.html#PRINTCAPNAME">printcap name = /etc/printcap </link></term>
		<listitem><para>
<indexterm significance="normal"><primary>CUPS</primary></indexterm>
<indexterm significance="normal"><primary>available printerd</primary></indexterm>
<indexterm significance="normal"><primary>printcap</primary></indexterm>
		Tells Samba where to look for a list of available printer names. Where CUPS is used, make sure that a printcap
		file is written. This is controlled by the <constant>Printcap</constant> directive in the
		<filename moreinfo="none">cupsd.conf</filename> file.
	</para></listitem></varlistentry>

	<varlistentry><term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTERADMIN" xlink:href="smb.conf.5.html#PRINTERADMIN">printer admin = @ntadmin </link></term>
		<listitem><para>
<indexterm significance="normal"><primary>add drivers</primary></indexterm>
<indexterm significance="normal"><primary>/etc/group</primary></indexterm>
<indexterm significance="normal"><primary>printer share</primary></indexterm>
<indexterm significance="normal"><primary>set printer properties</primary></indexterm>
		Members of the ntadmin group should be able to add drivers and set printer properties
		(<constant>ntadmin</constant> is only an example name; it needs to be a valid UNIX group name); root is
		implicitly always a <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTERADMIN" xlink:href="smb.conf.5.html#PRINTERADMIN">printer admin</link>. The <literal moreinfo="none">@</literal> sign precedes group names
		in the <filename moreinfo="none">/etc/group</filename>. A printer admin can do anything to printers via the remote
		administration interfaces offered by MS-RPC (see <link linkend="cups-msrpc">Printing Developments Since
		Samba-2.2</link>).  In larger installations, the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTERADMIN" xlink:href="smb.conf.5.html#PRINTERADMIN">printer admin</link> parameter is normally a
		per-share parameter. This permits different groups to administer each printer share.
	</para></listitem></varlistentry>

	<varlistentry><term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LPQCACHETIME" xlink:href="smb.conf.5.html#LPQCACHETIME">lpq cache time = 20 </link></term>
		<listitem><para>
<indexterm significance="normal"><primary>lpq command</primary></indexterm>
<indexterm significance="normal"><primary>lpq cache time</primary></indexterm>
		Controls the cache time for the results of the lpq command. It prevents the lpq command being called too often
		and reduces the load on a heavily used print server.
	</para></listitem></varlistentry>

	<varlistentry><term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="USECLIENTDRIVER" xlink:href="smb.conf.5.html#USECLIENTDRIVER">use client driver = no </link></term>
		<listitem><para>
<indexterm significance="normal"><primary>Windows NT/200x/XP</primary></indexterm>
		If set to <constant>yes</constant>, only takes effect for Windows NT/200x/XP clients (and not for Win
		95/98/ME). Its default value is <constant>No</constant> (or <constant>False</constant>).  It must
		<emphasis>not</emphasis> be enabled on print shares (with a <constant>yes</constant> or
		<constant>true</constant> setting) that have valid drivers installed on the Samba server. For more detailed
		explanations, see the <filename moreinfo="none">smb.conf</filename> man page.
	</para></listitem></varlistentry>
</variablelist>

</sect3>

<sect3 id="ptrsect">
<title>The [printers] Section</title>

<para>
<indexterm significance="normal"><primary>printers section</primary></indexterm>
<indexterm significance="normal"><primary>printcap</primary></indexterm>
The printers section is the second special section. If a section with this name appears in the <filename moreinfo="none">smb.conf</filename>,
users are able to connect to any printer specified in the Samba host's printcap file, because Samba on startup
then creates a printer share for every printer name it finds in the printcap file. You could regard this
section as a convenient shortcut to share all printers with minimal configuration. It is also a container for
settings that should apply as default to all printers. (For more details, see the <filename moreinfo="none">smb.conf</filename> man page.)
Settings inside this container must be share-level parameters.
</para>

<variablelist>
	<varlistentry><term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="COMMENT" xlink:href="smb.conf.5.html#COMMENT">comment = All printers </link></term>
		<listitem><para>
		The <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="COMMENT" xlink:href="smb.conf.5.html#COMMENT">comment</link> is shown next to the share if
		a client queries the server, either via <guiicon moreinfo="none">Network Neighborhood</guiicon> or with
		the <literal>net view</literal> command, to list available shares.
		</para></listitem>
	</varlistentry>

	<varlistentry><term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTABLE" xlink:href="smb.conf.5.html#PRINTABLE">printable = yes </link></term>
		<listitem><para>
		The <parameter>[printers]</parameter> service <emphasis>must</emphasis>
		be declared as printable. If you specify otherwise, smbd will refuse to load  at
		startup. This parameter allows connected clients to open, write to, and submit spool files
		into the directory specified with the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PATH" xlink:href="smb.conf.5.html#PATH">path</link>
		parameter for this service. It is used by Samba to differentiate printer shares from
		file shares. 
		</para></listitem>
	</varlistentry>

	<varlistentry><term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PATH" xlink:href="smb.conf.5.html#PATH">path = /var/spool/samba </link></term>
		<listitem><para>
		Must point to a directory used by Samba to spool incoming print files. <emphasis>It
		must not be the same as the spool directory specified in the configuration of your UNIX
		print subsystem!</emphasis> The path typically points to a directory that is world
		writable, with the <emphasis>sticky</emphasis> bit set to it.
		</para></listitem>
	</varlistentry>

	<varlistentry><term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="BROWSEABLE" xlink:href="smb.conf.5.html#BROWSEABLE">browseable = no </link></term>
		<listitem><para>
		Is always set to <constant>no</constant> if
		<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTABLE" xlink:href="smb.conf.5.html#PRINTABLE">printable = yes</link>. It makes
		the <parameter>[printer]</parameter> share itself invisible in the list of
		available shares in a <literal>net view</literal> command or in the Explorer browse
		list. (You will of course see the individual printers.)
		</para></listitem>
	</varlistentry>

	<varlistentry><term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="GUESTOK" xlink:href="smb.conf.5.html#GUESTOK">guest ok = yes </link></term>
		<listitem><para>
		If this parameter is set to <constant>yes</constant>, no password is required to
		connect to the printer's service. Access will be granted with the privileges of the
		<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="GUESTACCOUNT" xlink:href="smb.conf.5.html#GUESTACCOUNT">guest account</link>. On many systems the guest
		account will map to a user named "nobody." This user will usually be found
		in the UNIX passwd file with an empty password, but with no valid UNIX login. On some
		systems the guest account might not have the privilege to be able to print. Test this
		by logging in as your guest user using <literal>su - guest</literal> and run a system
		print command like:
		</para>
	
		<para>
		<userinput moreinfo="none">lpr -P printername /etc/motd</userinput>
		</para></listitem>
	</varlistentry>

	<varlistentry><term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PUBLIC" xlink:href="smb.conf.5.html#PUBLIC">public = yes </link></term>
		<listitem><para>
		Is a synonym for <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="GUESTOK" xlink:href="smb.conf.5.html#GUESTOK">guest ok = yes</link>.
		Since we have <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="GUESTOK" xlink:href="smb.conf.5.html#GUESTOK">guest ok = yes</link>, it
		really does not need to be here. (This leads to the interesting question, <quote>What if I
		by accident have two contradictory settings for the same share?</quote> The answer is that the
		last one encountered by Samba wins. <literal>testparm</literal> does not complain about different settings
		of the same parameter for the same share. You can test this by setting up multiple
		lines for the <parameter moreinfo="none">guest account</parameter> parameter with different usernames,
		and then run testparm to see which one is actually used by Samba.)
		</para></listitem>
	</varlistentry>

	<varlistentry><term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="READONLY" xlink:href="smb.conf.5.html#READONLY">read only = yes </link></term>
		<listitem><para>
		Normally (for other types of shares) prevents users from creating or modifying files
		in the service's directory. However, in a <emphasis>printable</emphasis> service, it is
		<emphasis>always</emphasis> allowed to write to the directory (if user privileges allow the
		connection), but only via print spooling operations. Normal write operations are not permitted.
		</para></listitem>
	</varlistentry>

	<varlistentry><term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="WRITABLE" xlink:href="smb.conf.5.html#WRITABLE">writable = no </link></term>
		<listitem><para>
		Is a synonym for <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="READONLY" xlink:href="smb.conf.5.html#READONLY">read only = yes</link>.
		</para></listitem>
	</varlistentry>
</variablelist>
</sect3>

<sect3>
<title>Any [my_printer_name] Section</title>

<para>
<indexterm significance="normal"><primary>loading printer drivers</primary></indexterm>
<indexterm significance="normal"><primary>name conflict</primary></indexterm>
If a <parameter moreinfo="none">[my_printer_name]</parameter> section appears in the <filename moreinfo="none">smb.conf</filename> file, which includes the
parameter <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTABLE" xlink:href="smb.conf.5.html#PRINTABLE">printable = yes</link> Samba will configure it as a printer share.
Windows 9x/Me clients may have problems with connecting or loading printer drivers if the share name has more
than eight characters. Do not name a printer share with a name that may conflict with an existing user or file
share name. On client connection requests, Samba always tries to find file shares with that name first. If it
finds one, it will connect to this and will not connect to a printer with the same name!
</para>

<?latex \newpage ?>
<variablelist>
	<varlistentry><term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="COMMENT" xlink:href="smb.conf.5.html#COMMENT">comment = Printer with Restricted Access </link></term>
		<listitem><para>
		The comment says it all.
		</para></listitem>
	</varlistentry>

	<varlistentry><term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PATH" xlink:href="smb.conf.5.html#PATH">path = /var/spool/samba_my_printer </link></term>
		<listitem><para>
		Sets the spooling area for this printer to a directory other than the default. It is not
		necessary to set it differently, but the option is available.
		</para></listitem>
	</varlistentry>

	<varlistentry><term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTERADMIN" xlink:href="smb.conf.5.html#PRINTERADMIN">printer admin = kurt </link></term>
		<listitem><para>
		The printer admin definition is different for this explicitly defined printer share from the general
		<parameter>[printers]</parameter> share. It is not a requirement; we did it to show that it is possible.
		</para></listitem>
	</varlistentry>

	<varlistentry><term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="BROWSEABLE" xlink:href="smb.conf.5.html#BROWSEABLE">browseable = yes </link></term>
		<listitem><para>
		This makes the printer browseable so the clients may conveniently find it when browsing the
		<guiicon moreinfo="none">Network Neighborhood</guiicon>.
		</para></listitem>
	</varlistentry>

	<varlistentry><term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTABLE" xlink:href="smb.conf.5.html#PRINTABLE">printable = yes </link></term>
		<listitem><para>
		See <link linkend="ptrsect">Section 20.4.1.2</link>.
		</para></listitem>
	</varlistentry>

	<varlistentry><term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="WRITABLE" xlink:href="smb.conf.5.html#WRITABLE">writable = no </link></term>
		<listitem><para>
		See <link linkend="ptrsect">Section 20.4.1.2</link>.
		</para></listitem>
	</varlistentry>

	<varlistentry><term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="HOSTSALLOW" xlink:href="smb.conf.5.html#HOSTSALLOW">hosts allow = 10.160.50.,10.160.51. </link></term>
		<listitem><para>
		Here we exercise a certain degree of access control by using the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="HOSTSALLOW" xlink:href="smb.conf.5.html#HOSTSALLOW">hosts allow</link>
		and <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="HOSTSDENY" xlink:href="smb.conf.5.html#HOSTSDENY">hosts deny</link> parameters. This is not by any means a safe bet. It is not a
		way to secure your printers. This line accepts all clients from a certain subnet in a first evaluation of
		access control.
		</para></listitem>
	</varlistentry>

	<varlistentry><term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="HOSTSDENY" xlink:href="smb.conf.5.html#HOSTSDENY">hosts deny = turbo_xp,10.160.50.23,10.160.51.60 </link></term>
		<listitem><para>
		All listed hosts are not allowed here (even if they belong to the allowed subnets). As
		you can see, you could name IP addresses as well as NetBIOS hostnames here.
		</para></listitem>
	</varlistentry>

	<varlistentry><term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="GUESTOK" xlink:href="smb.conf.5.html#GUESTOK">guest ok = no </link></term>
		<listitem><para>
		This printer is not open for the guest account.
		</para></listitem>
	</varlistentry>
</variablelist>
</sect3>

<sect3>
<title>Print Commands</title>

<para>
<indexterm significance="normal"><primary>print command</primary></indexterm>
<indexterm significance="normal"><primary>print subsystem</primary></indexterm>
<indexterm significance="normal"><primary>temporary location</primary></indexterm>
<indexterm significance="normal"><primary>shell scripts</primary></indexterm>
In each section defining a printer (or in the <parameter>[printers]</parameter> section),
a <parameter moreinfo="none">print command</parameter> parameter may be defined. It sets a command to process the files
that have been placed into the Samba print spool directory for that printer. (That spool directory was,
if you remember, set up with the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PATH" xlink:href="smb.conf.5.html#PATH">path</link> parameter). Typically,
this command will submit the spool file to the Samba host's print subsystem, using the suitable system
print command. But there is no requirement that this needs to be the case. For debugging or
some other reason, you may want to do something completely different than print the file. An example is a
command that just copies the print file to a temporary location for further investigation when you need
to debug printing. If you craft your own print commands (or even develop print command shell scripts),
make sure you pay attention to the need to remove the files from the Samba spool directory. Otherwise,
your hard disk may soon suffer from shortage of free space.
</para>
</sect3>

<sect3>
<title>Default UNIX System Printing Commands</title>

<para>
<indexterm significance="normal"><primary>default print command</primary></indexterm>
You learned earlier that Samba, in most cases, uses its built-in settings for many parameters if it cannot
find an explicitly stated one in its configuration file. The same is true for the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTCOMMAND" xlink:href="smb.conf.5.html#PRINTCOMMAND">print command</link>. The default print command varies depending on the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTING" xlink:href="smb.conf.5.html#PRINTING">printing</link> parameter
setting. In the commands listed in <link linkend="printOptions">Default Printing Settings</link> , you will
notice some parameters of the form <emphasis>%X</emphasis> where <emphasis>X</emphasis> is <emphasis>p, s,
J</emphasis>, and so on. These letters stand for printer name, spool file, and job ID, respectively.  They are
explained in more detail in <link linkend="printOptions">Default Printing Settings</link> presents an overview
of key printing options but excludes the special case of CUPS, is discussed in <link linkend="CUPS-printing">CUPS Printing Support</link>.
</para>

<table frame="all" id="printOptions">
	<title>Default Printing Settings</title>
	<tgroup cols="2" align="left" colsep="1" rowsep="1">
		<colspec align="left"/>
		<colspec align="left"/>
	<thead>
		<row>
		<entry>Setting</entry>
		<entry>Default Printing Commands</entry>
		</row>
	</thead>
	<tbody>
		<row>
		<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTING" xlink:href="smb.conf.5.html#PRINTING">printing = bsd|aix|lprng|plp</link></entry>
		<entry>print command is <literal>lpr -r -P%p %s</literal></entry>
		</row>
		<row>
		<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTING" xlink:href="smb.conf.5.html#PRINTING">printing = sysv|hpux</link></entry>
		<entry>print command is <literal>lp -c -P%p %s; rm %s</literal></entry>
		</row>
		<row>
		<entry> <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTING" xlink:href="smb.conf.5.html#PRINTING">printing = qnx</link></entry>
		<entry>print command is <literal>lp -r -P%p -s %s</literal></entry>
		</row>
		<row>
		<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTING" xlink:href="smb.conf.5.html#PRINTING">printing = bsd|aix|lprng|plp</link></entry>
		<entry>lpq command is <literal>lpq -P%p</literal></entry>
		</row>
		<row>
		<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTING" xlink:href="smb.conf.5.html#PRINTING">printing = sysv|hpux</link></entry>
		<entry>lpq command is <literal>lpstat -o%p</literal></entry>
		</row>
		<row>
		<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTING" xlink:href="smb.conf.5.html#PRINTING">printing = qnx</link></entry>
		<entry>lpq command is <literal>lpq -P%p</literal></entry>
		</row>
		<row>
		<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTING" xlink:href="smb.conf.5.html#PRINTING">printing = bsd|aix|lprng|plp</link></entry>
		<entry>lprm command is <literal>lprm -P%p %j</literal></entry>
		</row>
		<row>
		<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTING" xlink:href="smb.conf.5.html#PRINTING">printing = sysv|hpux</link></entry>
		<entry>lprm command is <literal>cancel %p-%j</literal></entry>
		</row>
		<row>
		<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTING" xlink:href="smb.conf.5.html#PRINTING">printing = qnx</link></entry>
		<entry>lprm command is <literal>cancel %p-%j</literal></entry>
		</row>
		<row>
		<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTING" xlink:href="smb.conf.5.html#PRINTING">printing = bsd|aix|lprng|plp</link></entry>
		<entry>lppause command is <literal>lp -i %p-%j -H hold</literal></entry>
		</row>
		<row>
		<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTING" xlink:href="smb.conf.5.html#PRINTING">printing = sysv|hpux</link></entry>
		<entry>lppause command   (...is empty)</entry>
		</row>
		<row>
		<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTING" xlink:href="smb.conf.5.html#PRINTING">printing = qnx</link></entry>
		<entry>lppause command   (...is empty)</entry>
		</row>
		<row>
		<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTING" xlink:href="smb.conf.5.html#PRINTING">printing = bsd|aix|lprng|plp</link></entry>
		<entry>lpresume command is <literal>lp -i %p-%j -H resume</literal></entry>
		</row>
		<row>
		<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTING" xlink:href="smb.conf.5.html#PRINTING">printing = sysv|hpux</link></entry>
		<entry>lpresume command   (...is empty)</entry>
		</row>
		<row>
		<entry><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTING" xlink:href="smb.conf.5.html#PRINTING">printing = qnx</link></entry>
		<entry>lpresume command   (...is empty)</entry>
		</row>
	</tbody>
	</tgroup>
</table>

<para>
<indexterm significance="normal"><primary>CUPS API</primary></indexterm>
<indexterm significance="normal"><primary>cupsd.conf</primary></indexterm>
<indexterm significance="normal"><primary>autogenerated printcap</primary></indexterm>
<indexterm significance="normal"><primary>libcups</primary></indexterm>
For <parameter moreinfo="none">printing = CUPS</parameter>, if Samba is compiled against libcups, it uses the CUPS API to
submit jobs. (It is a good idea also to set <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTCAP" xlink:href="smb.conf.5.html#PRINTCAP">printcap = cups</link> in case your
<filename moreinfo="none">cupsd.conf</filename> is set to write its autogenerated printcap file to an unusual place).
Otherwise, Samba maps to the System V printing commands with the -oraw option for printing; that is, it uses
<literal>lp -c -d%p -oraw; rm %s</literal>. With <parameter moreinfo="none">printing = cups</parameter>, and if Samba is
compiled against libcups, any manually set print command will be ignored!
</para>

</sect3>

<sect3>
<title>Custom Print Commands</title>

<para>
<indexterm significance="normal"><primary>print job</primary></indexterm>
<indexterm significance="normal"><primary>spooling</primary></indexterm>
After a print job has finished spooling to a service, the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTCOMMAND" xlink:href="smb.conf.5.html#PRINTCOMMAND">print command</link> will be used
by Samba via a system() call to process the spool file. Usually the command specified will submit the spool
file to the host's printing subsystem. But there is no requirement at all that this must be the case. The
print subsystem may not remove the spool file on its own, so whatever command you specify, you should ensure
that the spool file is deleted after it has been processed.
</para>

<para>
<indexterm significance="normal"><primary>traditional printing</primary></indexterm>
<indexterm significance="normal"><primary>customized print commands</primary></indexterm>
<indexterm significance="normal"><primary>built-in commands</primary></indexterm>
<indexterm significance="normal"><primary>macros</primary></indexterm>
There is no difficulty with using your own customized print commands with the traditional printing systems.
However, if you do not wish to roll your own, you should be well informed about the default built-in commands
that Samba uses for each printing subsystem (see <link linkend="printOptions">Default Printing
Settings</link>). In all the commands listed in the last paragraphs, you see parameters of the form
<emphasis>%X</emphasis>. These are <emphasis>macros</emphasis>, or shortcuts, used as placeholders for the
names of real objects. At the time of running a command with such a placeholder, Samba will insert the
appropriate value automatically. Print commands can handle all Samba macro substitutions. In regard to
printing, the following ones do have special relevance:
</para>

<itemizedlist>
	<listitem><para><parameter moreinfo="none">%s, %f</parameter> <?latex --- ?> the path to the spool file name.</para></listitem>
	<listitem><para><parameter moreinfo="none">%p</parameter> <?latex --- ?> the appropriate printer name.</para></listitem>
	<listitem><para><parameter moreinfo="none">%J</parameter> <?latex --- ?> the job name as transmitted by the client.</para></listitem>
	<listitem><para><parameter moreinfo="none">%c</parameter> <?latex --- ?> the number of printed pages of the spooled job (if known).</para></listitem>
	<listitem><para><parameter moreinfo="none">%z</parameter> <?latex --- ?> the size of the spooled print job (in bytes).</para></listitem>
</itemizedlist>

<para>
<indexterm significance="normal"><primary>default printer</primary></indexterm>
The print command must contain at least one occurrence of <parameter moreinfo="none">%s</parameter> or
<parameter moreinfo="none">%f</parameter>. The <parameter moreinfo="none">%p</parameter> is optional. If no printer name is supplied,
the <parameter moreinfo="none">%p</parameter> will be silently removed from the print command. In this case, the job is
sent to the default printer.
</para>

<para>
<indexterm significance="normal"><primary>global print command</primary></indexterm>
<indexterm significance="normal"><primary>spool files</primary></indexterm>
If specified in the <parameter>[global]</parameter> section, the print command given will be
used for any printable service that does not have its own print command specified. If there is neither a
specified print command for a printable service nor a global print command, spool files will be created
but not processed! Most importantly, print files will not be removed, so they will consume disk space.
</para>

<para>
<indexterm significance="normal"><primary>nobody account</primary></indexterm>
<indexterm significance="normal"><primary>guest account</primary></indexterm>
Printing may fail on some UNIX systems when using the <emphasis>nobody</emphasis> account. If this happens, create an
alternative guest account and give it the privilege to print. Set up this guest account in the
<parameter>[global]</parameter> section with the <parameter moreinfo="none">guest account</parameter> parameter.
</para>

<para>
<indexterm significance="normal"><primary>environment variables</primary></indexterm>
<indexterm significance="normal"><primary>print commands</primary></indexterm>
<indexterm significance="normal"><primary>print job</primary></indexterm>
You can form quite complex print commands. You need to realize that print commands are just
passed to a UNIX shell. The shell is able to expand the included environment variables as
usual. (The syntax to include a UNIX environment variable <parameter moreinfo="none">$variable</parameter>
in the Samba print command is <parameter moreinfo="none">%$variable</parameter>.) To give you a working
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTCOMMAND" xlink:href="smb.conf.5.html#PRINTCOMMAND">print command</link> example, the following will log a print job
to <filename moreinfo="none">/tmp/print.log</filename>, print the file, then remove it. The semicolon (<quote>;</quote>
is the usual separator for commands in shell scripts:
</para>

<para><simplelist>
	<member><indexterm>print command</indexterm><parameter>print command = echo Printing %s &gt;&gt; /tmp/print.log; lpr -P %p %s; rm %s</parameter></member>
</simplelist></para>

<para>
You may have to vary your own command considerably from this example depending on how you normally print
files on your system. The default for the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTCOMMAND" xlink:href="smb.conf.5.html#PRINTCOMMAND">print command</link>
parameter varies depending on the setting of the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTING" xlink:href="smb.conf.5.html#PRINTING">printing</link>
parameter. Another example is:
</para>

<para><simplelist>
<member><indexterm>print command</indexterm><parameter>print command = /usr/local/samba/bin/myprintscript %p %s</parameter></member>
</simplelist></para>
</sect3>
</sect2>
</sect1>

<sect1 id="cups-msrpc">
<title>Printing Developments Since Samba-2.2</title>

<para>
<indexterm significance="normal"><primary>LanMan</primary></indexterm>
<indexterm significance="normal"><primary>MS-RPC</primary></indexterm>
<indexterm significance="normal"><primary>SPOOLSS</primary></indexterm>
Prior to Samba-2.2.x, print server support for Windows clients was limited to <emphasis>LanMan</emphasis>
printing calls. This is the same protocol level as Windows 9x/Me PCs offer when they share printers.
Beginning with the 2.2.0 release, Samba started to support the native Windows NT printing mechanisms. These
are implemented via <emphasis>MS-RPC</emphasis> (Remote Procedure Calls).
MS-RPCs use the <emphasis>SPOOLSS</emphasis> named pipe for all printing.
</para>

<para>
The additional functionality provided by the new SPOOLSS support includes:
</para>

<itemizedlist>
	<listitem><para>
<indexterm significance="normal"><primary>Point'n'Print</primary></indexterm>
	Support for downloading printer driver files to Windows 95/98/NT/2000 clients upon
	demand (<emphasis>Point'n'Print</emphasis>).
	</para></listitem>

	<listitem><para>
<indexterm significance="normal"><primary>Add Printer Wizard</primary></indexterm>
	Uploading of printer drivers via the Windows NT <emphasis>Add Printer Wizard</emphasis> (APW)
	or the <ulink url="http://imprints.sourceforge.net/">Imprints</ulink> tool set.
	</para></listitem>

	<listitem><para>	
<indexterm significance="normal"><primary>MS-RPC</primary></indexterm>
<indexterm significance="normal"><primary>printing calls</primary></indexterm>
<indexterm significance="normal"><primary>StartDocPrinter</primary></indexterm>
<indexterm significance="normal"><primary>EnumJobs()</primary></indexterm>
<indexterm significance="normal"><primary>Win32 printing API</primary></indexterm>
	Support for the native MS-RPC printing calls such as StartDocPrinter, EnumJobs(), and so on. (See the
	<ulink url="http://msdn.microsoft.com/">MSDN documentation</ulink> for more information on the
	Win32 printing API).
	</para></listitem>

	<listitem><para>
<indexterm significance="normal"><primary>ACL</primary></indexterm>
<indexterm significance="normal"><primary>printer objects</primary></indexterm>
	Support for NT Access Control Lists (ACL) on printer objects.
	</para></listitem>

	<listitem><para>
<indexterm significance="normal"><primary>printer queue</primary></indexterm>
	Improved support for printer queue manipulation through the use of internal databases for spooled
	job information (implemented by various <filename moreinfo="none">*.tdb</filename> files).
	</para></listitem>
</itemizedlist>

<para>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
A benefit of updating is that Samba-3 is able to publish its printers to Active Directory (or LDAP).
</para>

<para>
<indexterm significance="normal"><primary>publish printers</primary></indexterm>
A fundamental difference exists between MS Windows NT print servers and Samba operation. Windows NT
permits the installation of local printers that are not shared. This is an artifact of the fact that
any Windows NT machine (server or client) may be used by a user as a workstation. Samba will publish all
printers that are made available, either by default or by specific declaration via printer-specific shares.
</para>

<para>
<indexterm significance="normal"><primary>SMB</primary></indexterm>
<indexterm significance="normal"><primary>MS-RPC</primary></indexterm>
<indexterm significance="normal"><primary>Everyone group</primary></indexterm>
<indexterm significance="normal"><primary>privileges</primary></indexterm>
<indexterm significance="normal"><primary>printer default permissions</primary></indexterm>
Windows NT/200x/XP Professional clients do not have to use the standard SMB printer share; they can
print directly to any printer on another Windows NT host using MS-RPC. This, of course, assumes that
the client has the necessary privileges on the remote host that serves the printer resource. The
default permissions assigned by Windows NT to a printer gives the print permissions to the well-known
<emphasis>Everyone</emphasis> group. (The older clients of type Windows 9x/Me can only print to shared
printers.)
</para>

<sect2>
<title>Point'n'Print Client Drivers on Samba Servers</title>

<para>
<indexterm significance="normal"><primary>printer drivers</primary></indexterm>
There is much confusion about what all this means. The question is often asked, <quote>Is it or is
it not necessary for printer drivers to be installed on a Samba host in order to support printing from
Windows clients?</quote> The answer to this is no, it is not necessary.
</para>

<para>
<indexterm significance="normal"><primary>install drivers</primary></indexterm>
<indexterm significance="normal"><primary>print queue</primary></indexterm>
Windows NT/2000 clients can, of course, also run their APW to install drivers <emphasis>locally</emphasis>
(which then connect to a Samba-served print queue). This is the same method used by Windows 9x/Me
clients. (However, a bug existed in Samba 2.2.0 that made Windows NT/2000 clients
require that the Samba server possess a valid driver for the printer. This was fixed in Samba 2.2.1).
</para>

<para>
<indexterm significance="normal"><primary>printer drivers</primary></indexterm>
<indexterm significance="normal"><primary>uploading</primary></indexterm>
But it is a new capability to install the printer drivers into the <parameter>[print$]</parameter>
share of the Samba server, and a big convenience, too. Then <emphasis>all</emphasis> clients
(including 95/98/ME) get the driver installed when they first connect to this printer share. The
<emphasis>uploading</emphasis> or <emphasis>depositing</emphasis> of the driver into this
<parameter>[print$]</parameter> share and the following binding of this driver to an existing
Samba printer share can be achieved by different means:
</para>

<itemizedlist>
	<listitem><para>
	Running the <emphasis>APW</emphasis> on an NT/200x/XP Professional client (this does not work from 95/98/ME clients).
	</para></listitem>

	<listitem><para>
	Using the <emphasis>Imprints</emphasis> toolset.
	</para></listitem>

	<listitem><para>
	Using the <emphasis>smbclient</emphasis> and <emphasis>rpcclient</emphasis> command-line tools.
	</para></listitem>

	<listitem><para>
	Using <emphasis>cupsaddsmb</emphasis> (only works for the CUPS printing system, not for LPR/LPD, LPRng, and so on).
	</para></listitem>
</itemizedlist>

<para>
<indexterm significance="normal"><primary>uploaded drivers</primary></indexterm>
<indexterm significance="normal"><primary>Point'n'Print</primary></indexterm>
Samba does not use these uploaded drivers in any way to process spooled files. These drivers are utilized
entirely by the clients who download and install them via the <quote>Point'n'Print</quote> mechanism
supported by Samba. The clients use these drivers to generate print files in the format the printer
(or the UNIX print system) requires. Print files received by Samba are handed over to the UNIX printing
system, which is responsible for all further processing, as needed.
</para>
</sect2>

<sect2>
<title>The Obsoleted [printer$] Section</title>

	<para>
<indexterm significance="normal"><primary>printer$ share</primary></indexterm>
<indexterm significance="normal"><primary>printer driver</primary></indexterm>
	Versions of Samba prior to 2.2 made it possible to use a share named <parameter moreinfo="none">[printer$]</parameter>. This
	name was taken from the same named service created by Windows 9x/Me clients when a printer was shared by them.
	Windows 9x/Me printer servers always have a <parameter>[printer$]</parameter> service that provides
	read-only access (with no password required) to support printer driver downloads. However, Samba's initial
	implementation allowed for a parameter named <parameter moreinfo="none">printer driver location</parameter> to be used on a
	per-share basis. This specified the location of the driver files associated with that printer. Another
	parameter named <parameter moreinfo="none">printer driver</parameter> provided a means of defining the printer driver name to
	be sent to the client.
	</para>

	<para>
<indexterm significance="normal"><primary>printer driver file</primary></indexterm>
<indexterm significance="normal"><primary>read-write access</primary></indexterm>
<indexterm significance="normal"><primary>ACLs</primary></indexterm>
	These parameters, including the <parameter moreinfo="none">printer driver file</parameter> parameter,
	are now removed and cannot be used in installations of Samba-3. The share name
	<parameter>[print$]</parameter> is now used for the location of downloadable printer
	drivers. It is taken from the <parameter>[print$]</parameter> service created
	by Windows NT PCs when a printer is shared by them. Windows NT print servers always have a
	<parameter>[print$]</parameter> service that provides read-write access (in the context
	of its ACLs) to support printer driver downloads and uploads. This does not mean Windows
	9x/Me clients are now thrown aside. They can use Samba's <parameter>[print$]</parameter>
	share support just fine.
	</para>
</sect2>

<sect2>
<title>Creating the [print$] Share</title>

<para>
<indexterm significance="normal"><primary>printer driver</primary></indexterm>
In order to support the uploading and downloading of printer driver files, you must first configure a
file share named <parameter>[print$]</parameter>. The public name of this share is hard coded
in the MS Windows clients. It cannot be renamed, since Windows clients are programmed to search for a
service of exactly this name if they want to retrieve printer driver files.
</para>

<para>
You should modify the server's file to add the global parameters and create the
<parameter>[print$]</parameter> file share (of course, some of the parameter values, such
as <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PATH" xlink:href="smb.conf.5.html#PATH">path</link>, are arbitrary and should be replaced with appropriate values for your
site). See <link linkend="prtdollar">[print\$] Example</link>.
</para>

<example id="prtdollar">
<title>[print$] Example</title>
<simplelist>
<member> </member><member><parameter>[global]</parameter></member>
<member># members of the ntadmin group should be able to add drivers and set</member>
<member># printer properties. root is implicitly always a 'printer admin'.</member>
<member><indexterm>printer admin</indexterm><parameter>printer admin = @ntadmin</parameter></member>
<member># ...</member>

<member> </member><member><parameter>[printers]</parameter></member>
<member># ...</member>

<member> </member><member><parameter>[print$]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = Printer Driver Download Area</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /etc/samba/drivers</parameter></member>
<member><indexterm>browseable</indexterm><parameter>browseable = yes</parameter></member>
<member><indexterm>guest ok</indexterm><parameter>guest ok = yes</parameter></member>
<member><indexterm>read only</indexterm><parameter>read only = yes</parameter></member>
<member><indexterm>write list</indexterm><parameter>write list = @ntadmin, root</parameter></member>
</simplelist>
</example>

<para>
Of course, you also need to ensure that the directory named by the
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PATH" xlink:href="smb.conf.5.html#PATH">path</link> parameter exists on the UNIX file system.
</para>

</sect2>

<sect2>
<title>[print$] Stanza Parameters</title>

<para>
<indexterm significance="normal"><primary>special section</primary></indexterm>
<indexterm significance="normal"><primary>special stanza</primary></indexterm>
<indexterm significance="normal"><primary>potential printer</primary></indexterm>
<indexterm significance="normal"><primary>driver download</primary></indexterm>
<indexterm significance="normal"><primary>local print driver</primary></indexterm>
The <parameter>[print$]</parameter> is a special section in <filename moreinfo="none">smb.conf</filename>. It contains settings relevant to
potential printer driver download and is used by Windows clients for local print driver installation.
The following parameters are frequently needed in this share section:
</para>

<variablelist>
	<varlistentry><term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="COMMENT" xlink:href="smb.conf.5.html#COMMENT">comment = Printer Driver Download Area </link></term>
		<listitem><para>
		The comment appears next to the share name if it is listed in a share list (usually Windows
		clients will not see it, but it will also appear up in a <literal>smbclient -L sambaserver
		</literal> output).
		</para></listitem>
	</varlistentry>

	<varlistentry><term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PATH" xlink:href="smb.conf.5.html#PATH">path = /etc/samba/printers </link></term>
		<listitem><para>
		The path to the location of the Windows driver file deposit from the UNIX point of view.
		</para></listitem>
	</varlistentry>

	<varlistentry><term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="BROWSEABLE" xlink:href="smb.conf.5.html#BROWSEABLE">browseable = no </link></term>
		<listitem><para>
		Makes the <parameter>[print$]</parameter> share invisible to clients from the
		<guimenu moreinfo="none">Network Neighborhood</guimenu>. By excuting from a <literal>cmd</literal> shell:
<screen format="linespecific">
<prompt moreinfo="none">C:\&gt; </prompt> <literal>net use g:\\sambaserver\print$</literal>
</screen>
		 you can still mount it from any client. This can also be done from the
		<guimenu moreinfo="none">Connect network drive menu&gt;</guimenu> from Windows Explorer.
		</para></listitem>
	</varlistentry>

	<varlistentry><term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="GUESTOK" xlink:href="smb.conf.5.html#GUESTOK">guest ok = yes </link></term>
		<listitem><para>
		Gives read-only access to this share for all guest users. Access may be granted to
		download and install printer drivers on clients. The requirement for <parameter moreinfo="none">guest ok
		= yes</parameter> depends on how your site is configured. If users will be guaranteed
		to have an account on the Samba host, then this is a non-issue.
		</para>

		<note><para> 
		If all your Windows NT users are guaranteed to be authenticated by the Samba server
		(for example, if Samba authenticates via an NT domain server and the user has already been
		validated by the domain controller in order to log on to the Windows NT session), then guest
		access is not necessary. Of course, in a workgroup environment where you just want
		to print without worrying about silly accounts and security, then configure the share for
		guest access. You should consider adding <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="MAPTOGUEST" xlink:href="smb.conf.5.html#MAPTOGUEST">map to guest = Bad	User</link>
		in the <parameter>[global]</parameter> section as well. Make sure you understand what this
		parameter does before using it.
		</para></note>
		</listitem>
	</varlistentry>

	<varlistentry><term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="READONLY" xlink:href="smb.conf.5.html#READONLY">read only = yes </link></term>
		<listitem><para>
		Because we do not want everybody to upload driver files (or even change driver settings),
		we tagged this share as not writable.
		</para></listitem>
	</varlistentry>

	<varlistentry><term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="WRITELIST" xlink:href="smb.conf.5.html#WRITELIST">write list = @ntadmin, root </link></term>
		<listitem><para>
		The <parameter>[print$]</parameter> was made read-only by the previous
		setting so we should create a <parameter moreinfo="none">write list</parameter> entry also. UNIX
		groups are denoted with a leading <quote>@</quote> character. Users listed here are allowed
		write-access (as an exception to the general public's read-only access), which they need to
		update files on the share. Normally, you will want to name only administrative-level user
		account in this setting. Check the file system permissions to make sure these accounts
		can copy files to the share. If this is a non-root account, then the account should also
		be mentioned in the global <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTERADMIN" xlink:href="smb.conf.5.html#PRINTERADMIN">printer admin</link>
		parameter. See the <filename moreinfo="none">smb.conf</filename> man page for more information on configuring file shares.
		</para></listitem>
	</varlistentry>
</variablelist>

</sect2>

<sect2>
<title>The [print$] Share Directory</title>

<para>
In order for a Windows NT print server to support the downloading of driver files by multiple client
architectures, you must create several subdirectories within the <parameter>[print$]</parameter>
service (i.e., the UNIX directory named by the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PATH" xlink:href="smb.conf.5.html#PATH">path</link>
parameter). These correspond to each of the supported client architectures. Samba follows this model as
well. Just like the name of the <parameter>[print$]</parameter> share itself, the subdirectories
must be exactly the names listed below (you may leave out the subdirectories of architectures you do
not need to support).
</para>

<para>
Therefore, create a directory tree below the
<parameter>[print$]</parameter> share for each architecture you wish
to support like this:
<programlisting format="linespecific">
[print$]--+
          |--W32X86           # serves drivers to Windows NT x86
          |--WIN40            # serves drivers to Windows 95/98
          |--W32ALPHA         # serves drivers to Windows NT Alpha_AXP
          |--W32MIPS          # serves drivers to Windows NT R4000
          |--W32PPC           # serves drivers to Windows NT PowerPC
</programlisting>
</para>

<important><title>Required Permissions</title>
	<para>
	In order to add a new driver to your Samba host, one of two conditions must hold true:
	</para>

	<itemizedlist>
		<listitem><para>
		The account used to connect to the Samba host must have a UID of 0 (i.e., a root account).
		</para></listitem>

		<listitem><para>
		The account used to connect to the Samba host must be named in the <emphasis>printer admin</emphasis> list.
		</para></listitem>
	</itemizedlist>

	<para>
	Of course, the connected account must still have write access to add files to the subdirectories beneath
	<parameter>[print$]</parameter>. Remember that all file shares are set to <quote>read-only</quote> by default.
	</para>
</important>

<para>
Once you have created the required <parameter>[print$]</parameter> service and
associated subdirectories, go to a Windows NT 4.0/200x/XP client workstation. Open <guiicon moreinfo="none">Network
Neighborhood</guiicon> or <guiicon moreinfo="none">My Network Places</guiicon> and browse for the Samba host. Once you
have located the server, navigate to its <guiicon moreinfo="none">Printers and Faxes</guiicon> folder. You should see
an initial listing of printers that matches the printer shares defined on your Samba host.
</para>
</sect2>
</sect1>

<sect1>
<title>Installing Drivers into [print$]</title>

<para>
Have you successfully created the <parameter>[print$]</parameter> share in <filename moreinfo="none">smb.conf</filename>, and have you forced
Samba to reread its <filename moreinfo="none">smb.conf</filename> file? Good. But you are not yet ready to use the new facility. The client
driver files need to be installed into this share. So far, it is still an empty share. Unfortunately, it is
not enough to just copy the driver files over. They need to be correctly installed so that appropriate records
for each driver will exist in the Samba internal databases so it can provide the correct drivers as they are
requested from MS Windows clients. And that is a bit tricky, to say the least. We now discuss two alternative
ways to install the drivers into <parameter>[print$]</parameter>:
</para>

<itemizedlist>
	<listitem><para>
	Using the Samba command-line utility <literal>rpcclient</literal> with its various subcommands (here,
	<literal>adddriver</literal> and <literal>setdriver</literal>) from any UNIX workstation.
	</para></listitem>

	<listitem><para>
	Running a GUI (<guiicon moreinfo="none">Printer Properties</guiicon> and <guiicon moreinfo="none">Add Printer Wizard</guiicon>)
	from any Windows NT/200x/XP client workstation.
	</para></listitem>
</itemizedlist>

<para>
The latter option is probably the easier one (even if the process may seem a little bit weird at first).
</para>

<sect2>
<title>Add Printer Wizard Driver Installation</title>

<para>
The printers initially listed in the Samba host's <guiicon moreinfo="none">Printers</guiicon> folder accessed from a
client's Explorer will have no real printer driver assigned to them. By default this driver name is set
to a null string. This must be changed now. The local <guiicon moreinfo="none">Add Printer Wizard</guiicon> (APW), run from
NT/2000/XP clients, will help us in this task.
</para>

<para>
Installation of a valid printer driver is not straightforward. You must attempt to view the printer properties
for the printer to which you want the driver assigned. Open Windows Explorer, open <guiicon moreinfo="none">Network
Neighborhood</guiicon>, browse to the Samba host, open Samba's <guiicon moreinfo="none">Printers</guiicon> folder, right-click
on the printer icon, and select <guimenu moreinfo="none">Properties...</guimenu>. You are now trying to view printer and
driver properties for a queue that has this default <constant>NULL</constant> driver assigned. This will
result in the following error message: <quote> Device settings cannot be displayed. The driver for the
specified printer is not installed, only spooler properties will be displayed. Do you want to install the
driver now?</quote>
</para>

<para>
Do <emphasis>not</emphasis> click on <guibutton moreinfo="none">Yes</guibutton>!  Instead, click on <guibutton moreinfo="none">No</guibutton>
in the error dialog.  Now you will be presented with the printer properties window. From here, the way to
assign a driver to a printer is open. You now have the choice of:
</para>

<itemizedlist>
	<listitem><para>
	Select a driver from the pop-up list of installed drivers. Initially this list will be empty.
	</para></listitem>

	<listitem><para>
	Click on <guibutton moreinfo="none">New Driver</guibutton> to install a new printer driver (which will
	start up the APW).
	</para></listitem>
</itemizedlist>

<para>
Once the APW is started, the procedure is exactly the same as the one you are familiar with in Windows (we
assume here that you are familiar with the printer driver installations procedure on Windows NT). Make sure
your connection is, in fact, set up as a user with <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTERADMIN" xlink:href="smb.conf.5.html#PRINTERADMIN">printer admin</link>
privileges (if in doubt, use <literal>smbstatus</literal> to check for this). If you wish to install
printer drivers for client operating systems other than <application moreinfo="none">Windows NT x86</application>,
you will need to use the <guilabel moreinfo="none">Sharing</guilabel> tab of the printer properties dialog.
</para>

<para>
Assuming you have connected with an administrative (or root) account (as named by the
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTERADMIN" xlink:href="smb.conf.5.html#PRINTERADMIN">printer admin</link> parameter), you will also be able to modify
other printer properties such as ACLs and default device settings using this dialog. For the default
device settings, please consider the advice given further in <link linkend="inst-rpc">Installing
Print Drivers Using <literal>rpcclient</literal></link>.
</para>
</sect2>

<sect2 id="inst-rpc">
<title>Installing Print Drivers Using <literal>rpcclient</literal></title>

<para>
The second way to install printer drivers into <parameter>[print$]</parameter> and set them
up in a valid way is to do it from the UNIX command line. This involves four distinct steps:
</para>

<orderedlist inheritnum="ignore" continuation="restarts">
	<listitem><para>
	Gather information about required driver files and collect the files.
	</para></listitem>

	<listitem><para>
	Deposit the driver files into the <parameter>[print$]</parameter> share's correct subdirectories
	(possibly by using <literal>smbclient</literal>).
	</para></listitem>

	<listitem><para>
	Run the <literal>rpcclient</literal> command-line utility once with the <literal>adddriver</literal>
	subcommand.
	</para></listitem>

	<listitem><para>
	Run <literal>rpcclient</literal> a second time with the <literal>setdriver</literal> subcommand.
	</para></listitem>
</orderedlist>

<para>
We provide detailed hints for each of these steps in the paragraphs that follow.
</para>

<sect3>
<title>Identifying Driver Files</title>

<para>
<indexterm significance="normal"><primary>driver files</primary></indexterm>
<indexterm significance="normal"><primary>driver CDROM</primary></indexterm>
<indexterm significance="normal"><primary>inf file</primary></indexterm>
To find out about the driver files, you have two options. You can check the contents of the driver
CDROM that came with your printer. Study the <filename moreinfo="none">*.inf</filename> files located on the CD-ROM. This
may not be possible, since the <filename moreinfo="none">*.inf</filename> file might be missing. Unfortunately, vendors have now started
to use their own installation programs. These installations packages are often in some Windows platform
archive format. Additionally, the files may be re-named during the installation process. This makes it
extremely difficult to identify the driver files required.
</para>

<para>
<indexterm significance="normal"><primary>W32X86</primary></indexterm>
Then you have the second option. Install the driver locally on a Windows client and
investigate which filenames and paths it uses after they are installed. (You need to repeat
this procedure for every client platform you want to support. We show it here for the
<application moreinfo="none">W32X86</application> platform only, a name used by Microsoft for all Windows NT/200x/XP
clients.)
</para>

<para>
<indexterm significance="normal"><primary>driver files</primary></indexterm>
A good method to recognize the driver files is to print the test page from the driver's
<guilabel moreinfo="none">Properties</guilabel> dialog (<guilabel moreinfo="none">General</guilabel> tab). Then look at the list of
driver files named on the printout. You'll need to recognize what Windows (and Samba) are calling the
<guilabel moreinfo="none">Driver File</guilabel>, <guilabel moreinfo="none">Data File</guilabel>, <guilabel moreinfo="none">Config File</guilabel>,
<guilabel moreinfo="none">Help File</guilabel>, and (optionally) <guilabel moreinfo="none">Dependent Driver Files</guilabel>
(this may vary slightly for Windows NT). You need to note all filenames for the next steps.
</para>

<para>
<indexterm significance="normal"><primary>rpcclient</primary></indexterm>
<indexterm significance="normal"><primary>enumdrivers</primary></indexterm>
<indexterm significance="normal"><primary>getdriver</primary></indexterm>
Another method to quickly test the driver filenames and related paths is provided by the
<literal>rpcclient</literal> utility. Run it with <literal>enumdrivers</literal> or with the
<literal>getdriver</literal> subcommand, each at the <filename moreinfo="none">3</filename> info level. In the following example,
<emphasis>TURBO_XP</emphasis> is the name of the Windows PC (in this case it was a Windows XP Professional
laptop). I installed the driver locally to TURBO_XP from a Samba server called <constant>KDE-BITSHOP</constant>.
We could run an interactive <literal>rpcclient</literal> session; then we would get an
<literal>rpcclient /&gt;</literal> prompt and would type the subcommands at this prompt. This is left as
a good exercise for you. For now, we use <literal>rpcclient</literal> with the <option>-c</option>
parameter to execute a single subcommand line and exit again. This is the method you use if you
want to create scripts to automate the procedure for a large number of printers and drivers. Note the
different quotation marks used to overcome the different spaces between words:
</para>

<para><screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">rpcclient -U'Danka%xxxx' -c \
	'getdriver "Heidelberg Digimaster 9110 (PS)" 3' TURBO_XP</userinput>
cmd = getdriver "Heidelberg Digimaster 9110 (PS)" 3

[Windows NT x86]
Printer Driver Info 3:
  Version: [2]
  Driver Name: [Heidelberg Digimaster 9110 (PS)]
  Architecture: [Windows NT x86]
  Driver Path: [C:\WINNT\System32\spool\DRIVERS\W32X86\2\HDNIS01_de.DLL]
  Datafile: [C:\WINNT\System32\spool\DRIVERS\W32X86\2\Hddm91c1_de.ppd]
  Configfile: [C:\WINNT\System32\spool\DRIVERS\W32X86\2\HDNIS01U_de.DLL]
  Helpfile: [C:\WINNT\System32\spool\DRIVERS\W32X86\2\HDNIS01U_de.HLP]
  
  Dependentfiles: [C:\WINNT\System32\spool\DRIVERS\W32X86\2\Hddm91c1_de.DLL]
  Dependentfiles: [C:\WINNT\System32\spool\DRIVERS\W32X86\2\Hddm91c1_de.INI]
  Dependentfiles: [C:\WINNT\System32\spool\DRIVERS\W32X86\2\Hddm91c1_de.dat]
  Dependentfiles: [C:\WINNT\System32\spool\DRIVERS\W32X86\2\Hddm91c1_de.cat]
  Dependentfiles: [C:\WINNT\System32\spool\DRIVERS\W32X86\2\Hddm91c1_de.def]
  Dependentfiles: [C:\WINNT\System32\spool\DRIVERS\W32X86\2\Hddm91c1_de.hre]
  Dependentfiles: [C:\WINNT\System32\spool\DRIVERS\W32X86\2\Hddm91c1_de.vnd]
  Dependentfiles: [C:\WINNT\System32\spool\DRIVERS\W32X86\2\Hddm91c1_de.hlp]
  Dependentfiles: [C:\WINNT\System32\spool\DRIVERS\W32X86\2\HDNIS01Aux.dll]
  Dependentfiles: [C:\WINNT\System32\spool\DRIVERS\W32X86\2\HDNIS01_de.NTF]
  
  Monitorname: []
  Defaultdatatype: []
</screen></para>

<para>
<indexterm significance="normal"><primary>Driver File</primary></indexterm>
<indexterm significance="normal"><primary>Driver Path</primary></indexterm>
<indexterm significance="normal"><primary>WIN40</primary></indexterm>
<indexterm significance="normal"><primary>W32X86</primary></indexterm>
You may notice that this driver has quite a large number of <guilabel moreinfo="none">Dependent files</guilabel>
(there are worse cases, however). Also, strangely, the
<guilabel moreinfo="none">Driver File</guilabel> is tagged here
<guilabel moreinfo="none">Driver Path</guilabel>. We do not yet have support for the so-called
<application moreinfo="none">WIN40</application> architecture installed. This name is used by Microsoft for the Windows
9x/Me platforms. If we want to support these, we need to install the Windows 9x/Me driver files in
addition to those for <application moreinfo="none">W32X86</application> (i.e., the Windows NT 2000/XP clients) onto a
Windows PC. This PC can also host the Windows 9x/Me drivers, even if it runs on Windows NT, 2000, or XP.
</para>

<para>
<indexterm significance="normal"><primary>UNC notation</primary></indexterm>
<indexterm significance="normal"><primary>Windows Explorer</primary></indexterm>
<indexterm significance="normal"><primary/></indexterm>
Since the <parameter>[print$]</parameter> share is usually accessible through the <guiicon moreinfo="none">Network
Neighborhood</guiicon>, you can also use the UNC notation from Windows Explorer to poke at it. The Windows
9x/Me driver files will end up in subdirectory <filename moreinfo="none">0</filename> of the <filename moreinfo="none">WIN40</filename>
directory. The full path to access them is <filename moreinfo="none">\\WINDOWSHOST\print$\WIN40\0\</filename>.
</para>

<note><para>
More recent drivers on Windows 2000 and Windows XP are installed into the <quote>3</quote> subdirectory
instead of the <quote>2</quote>. The version 2 of drivers, as used in Windows NT, were running in kernel
mode. Windows 2000 changed this. While it still can use the kernel mode drivers (if this is enabled by
the Admin), its native mode for printer drivers is user mode execution. This requires drivers designed
for this purpose. These types of drivers install into the <quote>3</quote> subdirectory.
</para></note>
</sect3>

<sect3>
<title>Obtaining Driver Files from Windows Client [print$] Shares</title>

<para>
Now we need to collect all the driver files we identified in our previous step. Where do we get them
from? Well, why not retrieve them from the very PC and the same <parameter>[print$]</parameter>
share that we investigated in our last step to identify the files? We can use <literal>smbclient</literal>
to do this. We will use the paths and names that were leaked to us by <literal>getdriver</literal>. The
listing is edited to include line breaks for readability:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">smbclient //TURBO_XP/print\$ -U'Danka%xxxx' \ 
   -c 'cd W32X86/2;mget HD*_de.* hd*ppd Hd*_de.* Hddm*dll HDN*Aux.DLL'</userinput>

added interface ip=10.160.51.60 bcast=10.160.51.255 nmask=255.255.252.0
Got a positive name query response from 10.160.50.8 ( 10.160.50.8 )
Domain=[DEVELOPMENT] OS=[Windows 5.1] Server=[Windows 2000 LAN Manager]
<prompt moreinfo="none">Get file Hddm91c1_de.ABD? </prompt><userinput moreinfo="none">n</userinput>
<prompt moreinfo="none">Get file Hddm91c1_de.def? </prompt><userinput moreinfo="none">y</userinput>
getting file \W32X86\2\Hddm91c1_de.def of size 428 as Hddm91c1_de.def
<prompt moreinfo="none">Get file Hddm91c1_de.DLL? </prompt><userinput moreinfo="none">y</userinput>
getting file \W32X86\2\Hddm91c1_de.DLL of size 876544 as Hddm91c1_de.DLL
[...]
</screen></para>

<para>
After this command is complete, the files are in our current local directory. You probably have noticed
that this time we passed several commands to the <option>-c</option> parameter, separated by semicolons.
This ensures that all commands are executed in sequence on the remote Windows server before
<literal>smbclient</literal> exits again.
</para>

<para>
<indexterm significance="normal"><primary>WIN40</primary></indexterm>
Remember to repeat the procedure for the <application moreinfo="none">WIN40</application> architecture should you need to
support Windows 9x/Me/XP clients. Remember too, the files for these architectures are in the
<filename moreinfo="none">WIN40/0/</filename> subdirectory. Once this is complete, we can run <literal>smbclient. .
.put</literal> to store the collected files on the Samba server's <parameter>[print$]</parameter> share.
</para>
</sect3>

<sect3>
<title>Installing Driver Files into [print$]</title>

<para>
We are now going to locate the driver files into the <parameter>[print$]</parameter> share. Remember, the
UNIX path to this share has been defined previously in your <filename moreinfo="none">smb.conf</filename> file. You also have created
subdirectories for the different Windows client types you want to support. If, for example, your
<parameter>[print$]</parameter> share maps to the UNIX path <filename moreinfo="none">/etc/samba/drivers/</filename>, your
driver files should now go here:
</para>

<itemizedlist>
	<listitem><para>
	For all Windows NT, 2000, and XP clients, <filename moreinfo="none">/etc/samba/drivers/W32X86/</filename> but
	not (yet) into the <filename moreinfo="none">2</filename> subdirectory.
	</para></listitem>

	<listitem><para>
	For all Windows 95, 98, and Me clients, <filename moreinfo="none">/etc/samba/drivers/WIN40/</filename> but not
	(yet) into the <filename moreinfo="none">0</filename> subdirectory.
	</para></listitem>
</itemizedlist>

<para>
<indexterm significance="normal"><primary>smbclient</primary></indexterm>
<indexterm significance="normal"><primary>getdriver</primary></indexterm>
We again use smbclient to transfer the driver files across the network. We specify the same files
and paths as were leaked to us by running <literal>getdriver</literal> against the original
<emphasis>Windows</emphasis> install. However, now we are going to store the files into a
<emphasis>Samba/UNIX</emphasis> print server's <parameter>[print$]</parameter> share.
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">smbclient //SAMBA-CUPS/print\$ -U'root%xxxx' -c \
  'cd W32X86; put HDNIS01_de.DLL; \
  put Hddm91c1_de.ppd; put HDNIS01U_de.DLL;        \
  put HDNIS01U_de.HLP; put Hddm91c1_de.DLL;        \
  put Hddm91c1_de.INI; put Hddm91c1KMMin.DLL;      \
  put Hddm91c1_de.dat; put Hddm91c1_de.dat;        \
  put Hddm91c1_de.def; put Hddm91c1_de.hre;        \
  put Hddm91c1_de.vnd; put Hddm91c1_de.hlp;        \
  put Hddm91c1_de_reg.HLP; put HDNIS01Aux.dll;     \
  put HDNIS01_de.NTF'</userinput>

added interface ip=10.160.51.60 bcast=10.160.51.255 nmask=255.255.252.0
Got a positive name query response from 10.160.51.162 ( 10.160.51.162 )
Domain=[CUPS-PRINT] OS=[UNIX] Server=[Samba 2.2.7a]
putting file HDNIS01_de.DLL as \W32X86\HDNIS01_de.DLL
putting file Hddm91c1_de.ppd as \W32X86\Hddm91c1_de.ppd
putting file HDNIS01U_de.DLL as \W32X86\HDNIS01U_de.DLL
putting file HDNIS01U_de.HLP as \W32X86\HDNIS01U_de.HLP
putting file Hddm91c1_de.DLL as \W32X86\Hddm91c1_de.DLL
putting file Hddm91c1_de.INI as \W32X86\Hddm91c1_de.INI
putting file Hddm91c1KMMin.DLL as \W32X86\Hddm91c1KMMin.DLL
putting file Hddm91c1_de.dat as \W32X86\Hddm91c1_de.dat
putting file Hddm91c1_de.dat as \W32X86\Hddm91c1_de.dat
putting file Hddm91c1_de.def as \W32X86\Hddm91c1_de.def
putting file Hddm91c1_de.hre as \W32X86\Hddm91c1_de.hre
putting file Hddm91c1_de.vnd as \W32X86\Hddm91c1_de.vnd
putting file Hddm91c1_de.hlp as \W32X86\Hddm91c1_de.hlp
putting file Hddm91c1_de_reg.HLP as \W32X86\Hddm91c1_de_reg.HLP
putting file HDNIS01Aux.dll as \W32X86\HDNIS01Aux.dll
putting file HDNIS01_de.NTF as \W32X86\HDNIS01_de.NTF
</screen>
<indexterm significance="normal"><primary>PPD</primary></indexterm>
<indexterm significance="normal"><primary>PostScript driver</primary></indexterm>
<indexterm significance="normal"><primary>adddriver</primary></indexterm>
Whew <?latex --- ?> that was a lot of typing! Most drivers are a lot smaller <?latex --- ?> many have only three generic
PostScript driver files plus one PPD. While we did retrieve the files from the <filename moreinfo="none">2</filename>
subdirectory of the <filename moreinfo="none">W32X86</filename> directory from the Windows box, we do not put them
(for now) in this same subdirectory of the Samba box. This relocation will automatically be done by the
<literal>adddriver</literal> command, which we will run shortly (and do not forget to also put the files
for the Windows 9x/Me architecture into the <filename moreinfo="none">WIN40/</filename> subdirectory should you need them).
</para>

</sect3>

<sect3>
<title><literal>smbclient</literal> to Confirm Driver Installation</title>

<para>
<indexterm significance="normal"><primary>smbclient</primary></indexterm>
<indexterm significance="normal"><primary>SSH</primary></indexterm>
For now we verify that our files are there. This can be done with <literal>smbclient</literal>, too
(but, of course, you can log in via SSH also and do this through a standard UNIX shell access):
</para>

<para><screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">smbclient //SAMBA-CUPS/print\$ -U 'root%xxxx' \
	-c 'cd W32X86; pwd; dir; cd 2; pwd; dir'</userinput>
 added interface ip=10.160.51.60 bcast=10.160.51.255 nmask=255.255.252.0
Got a positive name query response from 10.160.51.162 ( 10.160.51.162 )
Domain=[CUPS-PRINT] OS=[UNIX] Server=[Samba 2.2.8a]

Current directory is \\SAMBA-CUPS\print$\W32X86\
.                                  D        0  Sun May  4 03:56:35 2003
..                                 D        0  Thu Apr 10 23:47:40 2003
2                                   D        0  Sun May  4 03:56:18 2003
HDNIS01Aux.dll                      A    15356  Sun May  4 03:58:59 2003
Hddm91c1KMMin.DLL                   A    46966  Sun May  4 03:58:59 2003
HDNIS01_de.DLL                      A   434400  Sun May  4 03:58:59 2003
HDNIS01_de.NTF                      A   790404  Sun May  4 03:56:35 2003
Hddm91c1_de.DLL                     A   876544  Sun May  4 03:58:59 2003
Hddm91c1_de.INI                     A      101  Sun May  4 03:58:59 2003
Hddm91c1_de.dat                     A     5044  Sun May  4 03:58:59 2003
Hddm91c1_de.def                     A      428  Sun May  4 03:58:59 2003
Hddm91c1_de.hlp                     A    37699  Sun May  4 03:58:59 2003
Hddm91c1_de.hre                     A   323584  Sun May  4 03:58:59 2003
Hddm91c1_de.ppd                     A    26373  Sun May  4 03:58:59 2003
Hddm91c1_de.vnd                     A    45056  Sun May  4 03:58:59 2003
HDNIS01U_de.DLL                     A   165888  Sun May  4 03:58:59 2003
HDNIS01U_de.HLP                     A    19770  Sun May  4 03:58:59 2003
Hddm91c1_de_reg.HLP                 A   228417  Sun May  4 03:58:59 2003
              40976 blocks of size 262144. 709 blocks available

Current directory is \\SAMBA-CUPS\print$\W32X86\2\
.                                  D        0  Sun May  4 03:56:18 2003
..                                 D        0  Sun May  4 03:56:35 2003
ADOBEPS5.DLL                        A   434400  Sat May  3 23:18:45 2003
laserjet4.ppd                       A     9639  Thu Apr 24 01:05:32 2003
ADOBEPSU.DLL                        A   109568  Sat May  3 23:18:45 2003
ADOBEPSU.HLP                        A    18082  Sat May  3 23:18:45 2003
PDFcreator2.PPD                     A    15746  Sun Apr 20 22:24:07 2003
              40976 blocks of size 262144. 709 blocks available
</screen></para>

<para>
<indexterm significance="normal"><primary>Point'n'Print</primary></indexterm>
<indexterm significance="normal"><primary>printer driver files</primary></indexterm>
<indexterm significance="normal"><primary>print queue</primary></indexterm>
Notice that there are already driver files present in the <filename moreinfo="none">2</filename> subdirectory (probably from a
previous installation). Once the files for the new driver are there too, you are still a few steps away from
being able to use them on the clients. The only thing you could do now is retrieve them from a client just
like you retrieve ordinary files from a file share, by opening print$ in Windows Explorer. But that wouldn't
install them per Point'n'Print. The reason is that Samba does not yet know that these files are something
special, namely <emphasis>printer driver files</emphasis>, and it does not know to which print queue(s) these
driver files belong.
</para>
</sect3>

<sect3>
<title>Running <literal>rpcclient</literal> with <literal>adddriver</literal></title>

<para>
<indexterm significance="normal"><primary>adddriver</primary></indexterm>
<indexterm significance="normal"><primary>register driver files</primary></indexterm>
<indexterm significance="normal"><primary>TDB database</primary></indexterm>
Next, you must tell Samba about the special category of the files you just uploaded into the
<parameter>[print$]</parameter> share. This is done by the <literal>adddriver</literal>
command. It will prompt Samba to register the driver files into its internal TDB database files. The
following command and its output has been edited for readability:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">rpcclient -Uroot%xxxx -c 'adddriver "Windows NT x86" \
  "dm9110:HDNIS01_de.DLL: \
  Hddm91c1_de.ppd:HDNIS01U_de.DLL:HDNIS01U_de.HLP:   \
  NULL:RAW:Hddm91c1_de.DLL,Hddm91c1_de.INI,          \
  Hddm91c1_de.dat,Hddm91c1_de.def,Hddm91c1_de.hre,   \
  Hddm91c1_de.vnd,Hddm91c1_de.hlp,Hddm91c1KMMin.DLL, \
  HDNIS01Aux.dll,HDNIS01_de.NTF,                     \
  Hddm91c1_de_reg.HLP' SAMBA-CUPS</userinput>

cmd = adddriver "Windows NT x86" \
  "dm9110:HDNIS01_de.DLL:Hddm91c1_de.ppd:HDNIS01U_de.DLL:   \
  HDNIS01U_de.HLP:NULL:RAW:Hddm91c1_de.DLL,Hddm91c1_de.INI, \
  Hddm91c1_de.dat,Hddm91c1_de.def,Hddm91c1_de.hre,          \
  Hddm91c1_de.vnd,Hddm91c1_de.hlp,Hddm91c1KMMin.DLL,        \
  HDNIS01Aux.dll,HDNIS01_de.NTF,Hddm91c1_de_reg.HLP"

Printer Driver dm9110 successfully installed.
</screen></para>

<para>
<indexterm significance="normal"><primary>NT_STATUS_UNSUCCESSFUL</primary></indexterm>
<indexterm significance="normal"><primary>error message</primary></indexterm>
<indexterm significance="normal"><primary>adddriver</primary></indexterm>
After this step, the driver should be recognized by Samba on the print server. You need to be very
careful when typing the command. Don't exchange the order of the fields. Some changes would lead to
an <computeroutput moreinfo="none">NT_STATUS_UNSUCCESSFUL</computeroutput> error message. These become obvious. Other
changes might install the driver files successfully but render the driver unworkable. So take care!
Hints about the syntax of the adddriver command are in the man page. 
provides a more detailed description, should you need it.
</para>
</sect3>

<sect3>
<title>Checking <literal>adddriver</literal> Completion</title>

<para>
One indication for Samba's recognition of the files as driver files is the <computeroutput moreinfo="none">successfully
installed</computeroutput> message. Another one is the fact that our files have been moved by the
<literal>adddriver</literal> command into the <filename moreinfo="none">2</filename> subdirectory. You can check this
again with <literal>smbclient</literal>:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">smbclient //SAMBA-CUPS/print\$ -Uroot%xx \
	-c 'cd W32X86;dir;pwd;cd 2;dir;pwd'</userinput>
 added interface ip=10.160.51.162 bcast=10.160.51.255 nmask=255.255.252.0
 Domain=[CUPS-PRINT] OS=[UNIX] Server=[Samba 2.2.7a]

  Current directory is \\SAMBA-CUPS\print$\W32X86\
  .                                  D        0  Sun May  4 04:32:48 2003
  ..                                 D        0  Thu Apr 10 23:47:40 2003
  2                                   D        0  Sun May  4 04:32:48 2003
                40976 blocks of size 262144. 731 blocks available 

  Current directory is \\SAMBA-CUPS\print$\W32X86\2\
  .                                  D        0  Sun May  4 04:32:48 2003
  ..                                 D        0  Sun May  4 04:32:48 2003
  DigiMaster.PPD                      A   148336  Thu Apr 24 01:07:00 2003
  ADOBEPS5.DLL                        A   434400  Sat May  3 23:18:45 2003
  laserjet4.ppd                       A     9639  Thu Apr 24 01:05:32 2003
  ADOBEPSU.DLL                        A   109568  Sat May  3 23:18:45 2003
  ADOBEPSU.HLP                        A    18082  Sat May  3 23:18:45 2003
  PDFcreator2.PPD                     A    15746  Sun Apr 20 22:24:07 2003
  HDNIS01Aux.dll                      A    15356  Sun May  4 04:32:18 2003
  Hddm91c1KMMin.DLL                   A    46966  Sun May  4 04:32:18 2003
  HDNIS01_de.DLL                      A   434400  Sun May  4 04:32:18 2003
  HDNIS01_de.NTF                      A   790404  Sun May  4 04:32:18 2003
  Hddm91c1_de.DLL                     A   876544  Sun May  4 04:32:18 2003
  Hddm91c1_de.INI                     A      101  Sun May  4 04:32:18 2003
  Hddm91c1_de.dat                     A     5044  Sun May  4 04:32:18 2003
  Hddm91c1_de.def                     A      428  Sun May  4 04:32:18 2003
  Hddm91c1_de.hlp                     A    37699  Sun May  4 04:32:18 2003
  Hddm91c1_de.hre                     A   323584  Sun May  4 04:32:18 2003
  Hddm91c1_de.ppd                     A    26373  Sun May  4 04:32:18 2003
  Hddm91c1_de.vnd                     A    45056  Sun May  4 04:32:18 2003
  HDNIS01U_de.DLL                     A   165888  Sun May  4 04:32:18 2003
  HDNIS01U_de.HLP                     A    19770  Sun May  4 04:32:18 2003
  Hddm91c1_de_reg.HLP                 A   228417  Sun May  4 04:32:18 2003
                40976 blocks of size 262144. 731 blocks available
</screen></para>

<para>
Another verification is that the timestamp of the printing TDB files is now updated
(and possibly their file size has increased).
</para>
</sect3>

<sect3>
<title>Check Samba for Driver Recognition</title>

<para>
<indexterm significance="normal"><primary>registered</primary></indexterm>
Now the driver should be registered with Samba. We can easily verify this and will do so in a
moment. However, this driver is not yet associated with a particular printer. We may check the driver
status of the files by at least three methods:
</para>

<itemizedlist>
	<listitem><para>
<indexterm significance="normal"><primary>Network Neighborhood</primary></indexterm>
<indexterm significance="normal"><primary>Printers and Faxes</primary></indexterm>
<indexterm significance="normal"><primary>printer icon</primary></indexterm>
<indexterm significance="normal"><primary>Windows95/98/ME</primary></indexterm>
<indexterm significance="normal"><primary>Windows NT/2000/XP</primary></indexterm>
	From any Windows client browse Network Neighborhood, find the Samba host, and open the Samba
	<guiicon moreinfo="none">Printers and Faxes</guiicon> folder. Select any printer icon, right-click and select
	the printer <guimenuitem moreinfo="none">Properties</guimenuitem>. Click the <guilabel moreinfo="none">Advanced</guilabel>
	tab. Here is a field indicating the driver for that printer. A drop-down menu allows you to
	change that driver (be careful not to do this unwittingly). You can use this list to view
	all drivers known to Samba. Your new one should be among them. (Each type of client will
	see only its own architecture's list. If you do not have every driver installed for each platform,
	the list will differ if you look at it from Windows95/98/ME or Windows NT/2000/XP.)
	</para></listitem>

	<listitem><para>
<indexterm significance="normal"><primary>Network Neighborhood</primary></indexterm>
	From a Windows 200x/XP client (not Windows NT) browse <guiicon moreinfo="none">Network Neighborhood</guiicon>,
	search for the Samba server, open the server's <guiicon moreinfo="none">Printers</guiicon> folder,
	and right-click on the white background (with no printer highlighted). Select <guimenuitem moreinfo="none">Server
	Properties</guimenuitem>. On the <guilabel moreinfo="none">Drivers</guilabel> tab you will see the new driver
	listed. This view enables you to also inspect the list of files belonging to that driver
	(this does not work on Windows NT, but only on Windows 2000 and Windows XP; Windows NT does not
	provide the <guimenuitem moreinfo="none">Drivers</guimenuitem> tab). An alternative and much quicker method for
	Windows 2000/XP to start this dialog is by typing into a DOS box (you must of course adapt the
	name to your Samba server instead of <replaceable>SAMBA-CUPS</replaceable>):
	<screen format="linespecific">
	<userinput moreinfo="none">rundll32 printui.dll,PrintUIEntry /s /t2 /n\\<replaceable>SAMBA-CUPS</replaceable></userinput>
	</screen>
	</para>
	</listitem>

	<listitem><para>
	From a UNIX prompt, run this command (or a variant thereof), where
	<replaceable>SAMBA-CUPS</replaceable> is the name of the Samba host and xxxx represents the
	actual Samba password assigned to root:
	<screen format="linespecific">
	<userinput moreinfo="none">rpcclient -U'root%xxxx' -c 'enumdrivers' <replaceable>SAMBA-CUPS</replaceable></userinput>
	</screen>
	</para>

	<para>
	You will see a listing of all drivers Samba knows about. Your new one should be among
	them. But it is only listed under the <parameter moreinfo="none">[Windows NT x86]</parameter> heading, not under
	<parameter>[Windows 4.0]</parameter>, since you didn't install that part. Or did you?
	In our example it is named <constant>dm9110</constant>. Note that the third column shows the other
	installed drivers twice, one time for each supported architecture. Our new driver only shows up
	for <application moreinfo="none">Windows NT 4.0 or 2000</application>. To have it present for <application moreinfo="none">Windows
	95, 98, and Me</application>, you'll have to repeat the whole procedure with the WIN40 architecture
	and subdirectory.
	</para></listitem>
</itemizedlist>
</sect3>

<sect3>
<title>Specific Driver Name Flexibility</title>

<para>
<indexterm significance="normal"><primary>adddriver</primary></indexterm>
You can name the driver as you like. If you repeat the <literal>adddriver</literal> step with the same
files as before but with a different driver name, it will work the same:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">rpcclient -Uroot%xxxx         \
  -c 'adddriver "Windows NT x86"                     \
  "mydrivername:HDNIS01_de.DLL:              \
  Hddm91c1_de.ppd:HDNIS01U_de.DLL:HDNIS01U_de.HLP:   \
  NULL:RAW:Hddm91c1_de.DLL,Hddm91c1_de.INI,          \
  Hddm91c1_de.dat,Hddm91c1_de.def,Hddm91c1_de.hre,   \
  Hddm91c1_de.vnd,Hddm91c1_de.hlp,Hddm91c1KMMin.DLL, \
  HDNIS01Aux.dll,HDNIS01_de.NTF,Hddm91c1_de_reg.HLP' SAMBA-CUPS
  </userinput>

cmd = adddriver "Windows NT x86" \
 "mydrivername:HDNIS01_de.DLL:Hddm91c1_de.ppd:HDNIS01U_de.DLL:\
  HDNIS01U_de.HLP:NULL:RAW:Hddm91c1_de.DLL,Hddm91c1_de.INI,           \
  Hddm91c1_de.dat,Hddm91c1_de.def,Hddm91c1_de.hre,                    \
  Hddm91c1_de.vnd,Hddm91c1_de.hlp,Hddm91c1KMMin.DLL,                  \
  HDNIS01Aux.dll,HDNIS01_de.NTF,Hddm91c1_de_reg.HLP"

Printer Driver mydrivername successfully installed.
</screen></para>

<para>
<indexterm significance="normal"><primary>print queue</primary></indexterm>
<indexterm significance="normal"><primary>rpcclient</primary></indexterm>
<indexterm significance="normal"><primary>adddriver</primary></indexterm>
You will be able to bind that driver to any print queue (however, you are responsible that
you associate drivers to queues that make sense with respect to target printers). You cannot run the
<literal>rpcclient</literal> <literal>adddriver</literal> command repeatedly. Each run consumes the
files you had put into the <parameter>[print$]</parameter> share by moving them into the
respective subdirectories, so you must execute an <literal>smbclient ... put</literal> command before
each <literal>rpcclient ... adddriver</literal> command.
</para>
</sect3>

<sect3>
<title>Running <literal>rpcclient</literal> with <literal>setdriver</literal></title>

<para> 
<indexterm significance="normal"><primary>mapping printer driver</primary></indexterm>
<indexterm significance="normal"><primary>TDB</primary></indexterm>
Samba needs to know which printer owns which driver. Create a mapping of the driver to a printer, and
store this information in Samba's memory, the TDB files. The <literal>rpcclient setdriver</literal> command
achieves exactly this:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">rpcclient -U'root%xxxx' -c 'setdriver dm9110 mydrivername' <replaceable>SAMBA-CUPS</replaceable></userinput>
 cmd = setdriver dm9110 mydrivername

Successfully set dm9110 to driver mydrivername.
</screen></para>

<para>
Ah, no, I did not want to do that. Repeat, this time with the name I intended: 
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">rpcclient -U'root%xxxx' -c 'setdriver dm9110 dm9110' <replaceable>SAMBA-CUPS</replaceable></userinput>
 cmd = setdriver dm9110 dm9110
Successfully set dm9110 to driver dm9110.
</screen></para>

<para>
The syntax of the command is:
<screen format="linespecific">
<userinput moreinfo="none">rpcclient -U'root%<replaceable>sambapassword</replaceable>' -c 'setdriver <replaceable>printername</replaceable> \
 <replaceable>drivername</replaceable>' <replaceable>SAMBA-Hostname</replaceable></userinput>. 
</screen>
Now we have done most of the work, but not all of it.
</para>

<note><para>
The <literal>setdriver</literal> command will only succeed if the printer is already known to Samba. A
bug in 2.2.x prevented Samba from recognizing freshly installed printers. You had to restart Samba,
or at least send an HUP signal to all running smbd processes to work around this: <userinput moreinfo="none">kill -HUP
`pidof smbd`</userinput>.
</para></note>
</sect3>
</sect2>
</sect1>

<sect1>
<title>Client Driver Installation Procedure</title>

<para>
As Don Quixote said, <quote>The proof of the pudding is in the eating.</quote> The proof
for our setup lies in the printing. So let's install the printer driver onto the client PCs. This is
not as straightforward as it may seem. Read on.
</para>

<sect2>
<title>First Client Driver Installation</title>

<para>
Especially important is the installation onto the first client PC (for each architectural platform
separately). Once this is done correctly, all further clients are easy to set up and shouldn't need further
attention. What follows is a description for the recommended first procedure. You now work from a client
workstation. You should check that your connection is not unwittingly mapped to <emphasis>bad
user</emphasis> nobody. In a DOS box type:
</para>

<para><userinput moreinfo="none">net use \\<replaceable>SAMBA-SERVER</replaceable>\print$ /user:root</userinput></para>

<para>
Replace root, if needed, by another valid <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTERADMIN" xlink:href="smb.conf.5.html#PRINTERADMIN">printer admin</link> user as given in
the definition. Should you already be connected as a different user, you will get an error message. There
is no easy way to get rid of that connection, because Windows does not seem to know a concept of logging
off from a share connection (do not confuse this with logging off from the local workstation; that is
a different matter).  On Windows NT/200x, you can force a logoff from all smb/cifs connections by restarting the 
<emphasis>workstation</emphasis> service. You can try to close all Windows file explorers and Internet Explorer for 
Windows. As a last resort, you may have to reboot. Make sure there is no automatic reconnection set up. It may be
easier to go to a different workstation and try from there. After you have made sure you are connected
as a printer admin user (you can check this with the <literal>smbstatus</literal> command on Samba),
do this from the Windows workstation:
</para>

<procedure>
	<step performance="required"><para>
	Open <guiicon moreinfo="none">Network Neighborhood</guiicon>.
	</para></step>

	<step performance="required"><para>
	Browse to Samba server.
	</para></step>

	<step performance="required"><para>
	Open its <guiicon moreinfo="none">Printers and Faxes</guiicon> folder.
	</para></step>

	<step performance="required"><para>
	Highlight and right-click on the printer.
	</para></step>

	<step performance="required"><para>
	Select <guimenuitem moreinfo="none">Connect</guimenuitem> (for Windows NT4/200x
	it is possibly <guimenuitem moreinfo="none">Install</guimenuitem>).
	</para></step>
</procedure>

<para>
A new printer (named <replaceable>printername</replaceable> on Samba server) should now have
appeared in your <emphasis>local</emphasis> Printer folder (check <guimenu moreinfo="none">Start</guimenu> -&gt;
<guimenuitem moreinfo="none">Settings</guimenuitem> -&gt; <guimenuitem moreinfo="none">Control Panel</guimenuitem> -&gt; <guiicon moreinfo="none">Printers
and Faxes</guiicon>).
</para>

<para>
<indexterm significance="normal"><primary>print test page</primary></indexterm>
Most likely you are tempted to try to print a test page. After all, you now can open the printer
properties, and on the <guimenu moreinfo="none">General</guimenu> tab there is a button offering to do just that. But
chances are that you get an error message saying "<literal moreinfo="none">Unable to print Test Page</literal>." The
reason might be that there is not yet a valid device mode set for the driver or that the <quote>printer
driver data</quote> set is still incomplete.
</para>

<para>
You must make sure that a valid <parameter moreinfo="none">device mode</parameter> is set for the
driver. We now explain what that means.
</para>
</sect2>

<sect2 id="prt-modeset">
<title>Setting Device Modes on New Printers</title>

<para>
For a printer to be truly usable by a Windows NT/200x/XP client, it must possess:
</para>

<itemizedlist>
	<listitem><para>
<indexterm significance="normal"><primary>device mode</primary></indexterm>
	A valid <emphasis>device mode</emphasis> generated by the driver for the printer (defining things
	like paper size, orientation and duplex settings).
	</para></listitem>

	<listitem><para>
<indexterm significance="normal"><primary>printer driver data</primary></indexterm>
	A complete set of <emphasis>printer driver data</emphasis> generated by the driver.
	</para></listitem>
</itemizedlist>

<para>
<indexterm significance="normal"><primary>ntprinters.tdb</primary></indexterm>
<indexterm significance="normal"><primary>ntdrivers.tdb</primary></indexterm>
<indexterm significance="normal"><primary>printing.tdb</primary></indexterm>
<indexterm significance="normal"><primary>ntforms.tdb</primary></indexterm>
<indexterm significance="normal"><primary>TDB database files</primary></indexterm>
If either of these is incomplete, the clients can produce less than optimal output at best. In the
worst cases, unreadable garbage or nothing at all comes from the printer, or it produces a harvest of
error messages when attempting to print. Samba stores the named values and all printing-related information in
its internal TDB database files <filename moreinfo="none">(ntprinters.tdb</filename>, <filename moreinfo="none">ntdrivers.tdb</filename>,
<filename moreinfo="none">printing.tdb</filename>, and <filename moreinfo="none">ntforms.tdb</filename>).
</para>

<para>
The device mode and the set of printer driver data are basically collections
of settings for all print queue properties, initialized in a sensible way. Device modes and
printer driver data should initially be set on the print server (the Samba host) to healthy
values so the clients can start to use them immediately. How do we set these initial healthy values?
This can be achieved by accessing the drivers remotely from an NT (or 200x/XP) client, as discussed
in the following paragraphs.
</para>

<para>
Be aware that a valid device mode can only be initiated by a <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTERADMIN" xlink:href="smb.conf.5.html#PRINTERADMIN">printer admin</link> or root
(the reason should be obvious). Device modes can be correctly set only by executing the printer driver program
itself. Since Samba cannot execute this Win32 platform driver code, it sets this field initially to NULL
(which is not a valid setting for clients to use). Fortunately, most drivers automatically generate the
printer driver data that is needed when they are uploaded to the <parameter>[print$]</parameter> share with
the help of the APW or rpcclient.
</para>

<para>
The generation and setting of a first valid device mode, however, requires some tickling from a client
to set it on the Samba server. The easiest means of doing so is to simply change the page orientation on
the server's printer. This executes enough of the printer driver program on the client for the desired
effect to happen and feeds back the new device mode to our Samba server. You can use the native Windows
NT/200x/XP printer properties page from a Window client for this:
</para>

<procedure>
<title>Procedure to Initialize the Printer Driver Settings</title>
	<step performance="required"><para>
	Browse the <guiicon moreinfo="none">Network Neighborhood</guiicon>.
	</para></step>

	<step performance="required"><para>
	Find the Samba server.
	</para></step>

	<step performance="required"><para>
	Open the Samba server's <guiicon moreinfo="none">Printers and Faxes</guiicon> folder.
	</para></step>

	<step performance="required"><para>
	Highlight the shared printer in question.
	</para></step>

	<step performance="required"><para>
	Right-click on the printer (you may already be here if you followed the last section's description).
	</para></step>

	<step performance="required"><para>
	At the bottom of the context menu select <guimenu moreinfo="none">Properties</guimenu> (if the menu still offers the 
	<guimenuitem moreinfo="none">Connect</guimenuitem> entry further above, you
	need to click on that one first to achieve the driver
	installation, as shown in the last section).
	</para></step>

	<step performance="required"><para>
	Go to the <guilabel moreinfo="none">Advanced</guilabel> tab; click on <guibutton moreinfo="none">Printing Defaults</guibutton>.
	</para></step>

	<step performance="required"><para>
	Change the <guimenuitem moreinfo="none">Portrait</guimenuitem> page setting to <guimenuitem moreinfo="none">Landscape</guimenuitem> (and back).
	</para></step>

	<step performance="required"><para>
	Make sure to apply changes between swapping the page orientation to cause the change to actually take effect.
	</para></step>

	<step performance="required"><para>
	While you are at it, you may also want to set the desired printing defaults here, which then apply to all future
	client driver installations.
	</para></step>
</procedure>

<para>
This procedure executes the printer driver program on the client platform and feeds back the correct
device mode to Samba, which now stores it in its TDB files. Once the driver is installed on the client,
you can follow the analogous steps by accessing the <emphasis>local</emphasis> <guiicon moreinfo="none">Printers</guiicon>
folder, too, if you are a Samba printer admin user. From now on, printing should work as expected.
</para>

<para>
<indexterm significance="normal"><primary>default devmode</primary></indexterm>
Samba includes a service-level parameter name <parameter moreinfo="none">default devmode</parameter> for generating a default
device mode for a printer. Some drivers function well with Samba's default set of properties. Others
may crash the client's spooler service. So use this parameter with caution. It is always better to have
the client generate a valid device mode for the printer and store it on the server for you.
</para>
</sect2>

<sect2>
<title>Additional Client Driver Installation</title>

<para>
<indexterm significance="normal"><primary>additional driver</primary></indexterm>
Every additional driver may be installed in the same way as just described.  Browse <literal>Network
Neighborhood</literal>, open the <guiicon moreinfo="none">Printers</guiicon> folder on Samba server, right-click on
<guiicon moreinfo="none">Printer</guiicon>, and choose <guimenuitem moreinfo="none">Connect...</guimenuitem>. Once this completes (should be
not more than a few seconds, but could also take a minute, depending on network conditions), you should find
the new printer in your client workstation local <guiicon moreinfo="none">Printers and Faxes</guiicon> folder.
</para>

<para>
You can also open your local <guiicon moreinfo="none">Printers and Faxes</guiicon> folder by
using this command on Windows 200x/XP Professional workstations:
<screen format="linespecific">
<userinput moreinfo="none">rundll32 shell32.dll,SHHelpShortcuts_RunDLL PrintersFolder</userinput>
</screen>
or this command on Windows NT 4.0 workstations:
<indexterm significance="normal"><primary>rundll32</primary></indexterm>
<screen format="linespecific">
<userinput moreinfo="none">rundll32 shell32.dll,Control_RunDLL MAIN.CPL @2</userinput>
</screen>
</para>

<para>
You can enter the commands either inside a <guilabel moreinfo="none">DOS box</guilabel> window or in the <guimenuitem moreinfo="none">Run
command...</guimenuitem> field from the <guimenu moreinfo="none">Start</guimenu> menu.
</para>
</sect2>

<sect2>
<title>Always Make First Client Connection as root or <quote>printer admin</quote></title>

<para>
After you installed the driver on the Samba server (in its <parameter>[print$]</parameter> share), you
should always make sure that your first client installation completes correctly. Make it a habit for yourself
to build the very first connection from a client as <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTERADMIN" xlink:href="smb.conf.5.html#PRINTERADMIN">printer admin</link>. This is to make
sure that:
</para>

<itemizedlist>
	<listitem><para>
	A first valid <emphasis>device mode</emphasis> is really initialized (see above <link linkend="prt-modeset">Setting Device Modes on New Printers</link>) for more explanation details).
	</para></listitem>

	<listitem><para>	
	The default print settings of your printer for all further client installations are as you want them.
	</para></listitem>
</itemizedlist>

<para>
Do this by changing the orientation to landscape, click on <guiicon moreinfo="none">Apply</guiicon>, and then change it
back again. Next, modify the other settings (for example, you do not want the default media size set to
<guiicon moreinfo="none">Letter</guiicon> when you are all using <guiicon moreinfo="none">A4</guiicon>, right? You may want to set the
printer for <guiicon moreinfo="none">duplex</guiicon> as the default, and so on).
</para>

<para>
<indexterm significance="normal"><primary>runas</primary></indexterm>
To connect as root to a Samba printer, try this command from a Windows 200x/XP DOS box command prompt:
<screen format="linespecific">
<prompt moreinfo="none">C:\&gt; </prompt><userinput moreinfo="none">runas /netonly /user:root "rundll32 printui.dll,PrintUIEntry /p /t3 /n 
	\\<replaceable>SAMBA-SERVER</replaceable>\<replaceable>printername</replaceable>"</userinput>
</screen>
</para>

<para>
You will be prompted for <constant>root</constant>'s Samba password; type it, wait a few seconds, click on
<guibutton moreinfo="none">Printing Defaults</guibutton>, and proceed to set the job options that should be used as defaults
by all clients. Alternatively, instead of root you can name one other member of the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTERADMIN" xlink:href="smb.conf.5.html#PRINTERADMIN">printer admin</link> from the setting.
</para>

<para>
Now all the other users downloading and installing the driver the same way (using
<literal moreinfo="none">Point'n'Print</literal>) will have the same defaults set for them. If you miss this step, you'll get a
lot of help desk calls from your users, but maybe you like to talk to people.
</para>
</sect2>
</sect1>

<sect1>
<title>Other Gotchas</title>

<para>
Your driver is installed. It is now ready for Point'n'Print installation by the clients. You may have tried to
download and use it on your first client machine, but wait. Let's make sure you are acquainted first with a
few tips and tricks you may find useful. For example, suppose you did not set the defaults on the printer, as
advised in the preceding paragraphs. Your users complain about various issues (such as, <quote>We need to set
the paper size for each job from Letter to A4 and it will not store it</quote>).
</para>

<sect2>
<title>Setting Default Print Options for Client Drivers</title>

<para>
The last sentence might be viewed with mixed feelings by some users and Admins. They have struggled for hours
and could not arrive at a point where their settings seemed to be saved. It is not their fault. The confusing
thing is that in the multitabbed dialog that pops up when you right-click on the printer name and select
<guimenuitem moreinfo="none">Properties</guimenuitem>, you can arrive at two dialogs that appear identical, each claiming that
they help you to set printer options in three different ways. Here is the definitive answer to the Samba
default driver setting FAQ:
</para>

<formalpara><title><quote>I can not set and save default print options
for all users on Windows 200x/XP. Why not?</quote></title>

<para>
How are you doing it? I bet the wrong way. (It is not easy to find out, though.) There are three different
ways to bring you to a dialog that seems to set everything. All three dialogs look the same, but only one of
them does what you intend. You need to be Administrator or Print Administrator to do this for all users. Here
is how I reproduce it in an XP Professional:
</para>

<orderedlist numeration="upperalpha" inheritnum="ignore" continuation="restarts">
	<listitem><para>The first <quote>wrong</quote> way:
	<orderedlist numeration="arabic" inheritnum="ignore" continuation="restarts">
		<listitem><para>Open the <guiicon moreinfo="none">Printers</guiicon> folder.</para></listitem>

		<listitem><para>Right-click on the printer (<emphasis>remoteprinter on cupshost</emphasis>) and
		select in context menu <guimenu moreinfo="none">Printing Preferences...</guimenu>.</para></listitem>

		<listitem><para>Look at this dialog closely and remember what it looks like.</para></listitem>
	</orderedlist></para></listitem>

	<listitem><para>The second <quote>wrong</quote> way:
		<orderedlist numeration="arabic" inheritnum="ignore" continuation="restarts">
			<listitem><para>Open the <guimenu moreinfo="none">Printers</guimenu> folder.</para></listitem>

			<listitem><para>Right-click on the printer (<emphasis>remoteprinter on
			cupshost</emphasis>) and select in the context menu
			<guimenuitem moreinfo="none">Properties</guimenuitem></para></listitem>.

			<listitem><para>Click on the <guilabel moreinfo="none">General</guilabel>
			tab.</para></listitem>

			<listitem><para>Click on the <guibutton moreinfo="none">Printing
			Preferences...</guibutton> button.</para></listitem>

			<listitem><para>A new dialog opens. Keep this dialog open and go back
			to the parent dialog.</para></listitem>
		</orderedlist>
	</para></listitem>

	<listitem><para>
	The third and correct way (should you do this from the beginning, just carry out steps 1
	and 2 from the second method above):
	</para>

		<orderedlist numeration="arabic" inheritnum="ignore" continuation="restarts">
			<listitem><para>Click on the <guilabel moreinfo="none">Advanced</guilabel>
			tab. (If everything is <quote>grayed out,</quote> then you are not logged
			in as a user with enough privileges.)</para></listitem>

			<listitem><para>Click on the <guibutton moreinfo="none">Printing
			Defaults</guibutton> button.</para></listitem>

			<listitem><para>On any of the two new tabs,
			click on the
			<guilabel moreinfo="none">Advanced</guilabel> button.</para></listitem>

			<listitem><para>A new dialog opens. Compare
			this one to the other. Are they
			identical when you compare one from
			<quote>B.5</quote> and one from A.3?</para></listitem>
		</orderedlist>
	</listitem>
</orderedlist>

<para>
Do you see any difference in the two settings dialogs? I do not either. However, only the last one, which you
arrived at with steps C.1 through C.6 will permanently save any settings which will then become the defaults
for new users. If you want all clients to have the same defaults, you need to conduct these steps as
administrator (<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTERADMIN" xlink:href="smb.conf.5.html#PRINTERADMIN">printer admin</link>) before a client downloads the driver (the clients can
later set their own per-user defaults by following procedures A or B above). Windows 200x/XP allow per-user
default settings and the ones the administrator gives them before they set up their own. The parents of the
identical-looking dialogs have a slight difference in their window names; one is called
<computeroutput moreinfo="none">Default Print Values for Printer Foo on Server Bar</computeroutput> (which is the one you
need) and the other is called <quote><computeroutput moreinfo="none">Print Settings for Printer Foo on Server
Bar</computeroutput></quote>. The last one is the one you arrive at when you right-click on the printer and
select <guimenuitem moreinfo="none">Print Settings...</guimenuitem>. This is the one that you were taught to use back in the
days of Windows NT, so it is only natural to try the same way with Windows 200x/XP. You would not dream that
there is now a different path to arrive at an identical-looking, but functionally different, dialog to set
defaults for all users.
</para></formalpara>

<tip><para>Try (on Windows 200x/XP) to run this command (as a user with the right privileges):
<indexterm significance="normal"><primary>rundll32</primary></indexterm>
</para>

<para><userinput moreinfo="none">
rundll32 printui.dll,PrintUIEntry /p /t3 /n\\<replaceable>SAMBA-SERVER</replaceable>\<replaceable>printersharename</replaceable>
</userinput></para>

<para>
To see the tab with the <guilabel moreinfo="none">Printing Defaults</guilabel> button (the one you need), also run this command:
</para>

<para><userinput moreinfo="none">
rundll32 printui.dll,PrintUIEntry /p /t0 /n\\<replaceable>SAMBA-SERVER</replaceable>\<replaceable>printersharename</replaceable>
</userinput></para>

<para>
To see the tab with the <guilabel moreinfo="none">Printing Preferences</guilabel>
button (the one that does not set systemwide defaults), you can
start the commands from inside a DOS box or from <guimenu moreinfo="none">Start</guimenu> -&gt; <guimenuitem moreinfo="none">Run</guimenuitem>.
</para>
</tip>

</sect2>

<sect2>
<title>Supporting Large Numbers of Printers</title>

<para>
One issue that has arisen during the recent development phase of Samba is the need to support driver
downloads for hundreds of printers. Using Windows NT APW for this task is somewhat awkward (to say the least). If
you do not want to acquire RSS pains from the printer installation clicking orgy alone, you need
to think about a non-interactive script.
</para>

<para>
If more than one printer is using the same driver, the <literal>rpcclient setdriver</literal>
command can be used to set the driver associated with an installed queue. If the driver is uploaded to
<parameter>[print$]</parameter> once and registered with the printing TDBs, it can be used by
multiple print queues. In this case, you just need to repeat the <literal>setprinter</literal> subcommand of
<literal>rpcclient</literal> for every queue (without the need to conduct the <literal>adddriver</literal>
repeatedly). The following is an example of how this can be accomplished:
</para>

<para><screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">rpcclient <replaceable>SAMBA-CUPS</replaceable> -U root%<replaceable>secret</replaceable> -c 'enumdrivers'</userinput>
 cmd = enumdrivers
 
 [Windows NT x86]
 Printer Driver Info 1:
   Driver Name: [infotec  IS 2075 PCL 6]
 
 Printer Driver Info 1:
   Driver Name: [DANKA InfoStream]
 
 Printer Driver Info 1:
   Driver Name: [Heidelberg Digimaster 9110 (PS)]
 
 Printer Driver Info 1:
   Driver Name: [dm9110]

 Printer Driver Info 1:
   Driver Name: [mydrivername]

 [....]
</screen>

<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">rpcclient <replaceable>SAMBA-CUPS</replaceable> -U root%<replaceable>secret</replaceable> -c 'enumprinters'</userinput>
 cmd = enumprinters
   flags:[0x800000]
   name:[\\SAMBA-CUPS\dm9110]
   description:[\\SAMBA-CUPS\dm9110,,110ppm HiVolume DANKA Stuttgart]
   comment:[110 ppm HiVolume DANKA Stuttgart]
 [....]
</screen>

<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">rpcclient <replaceable>SAMBA-CUPS</replaceable> -U root%<replaceable>secret</replaceable> -c \
  'setdriver <replaceable>dm9110</replaceable> "<replaceable>Heidelberg Digimaster 9110 (PS)</replaceable>"'</userinput>
 cmd = setdriver dm9110 Heidelberg Digimaster 9110 (PPD)
 Successfully set dm9110 to driver Heidelberg Digimaster 9110 (PS).
</screen>

<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">rpcclient <replaceable>SAMBA-CUPS</replaceable> -U root%<replaceable>secret</replaceable> -c 'enumprinters'</userinput>
 cmd = enumprinters
   flags:[0x800000]
   name:[\\SAMBA-CUPS\dm9110]
   description:[\\SAMBA-CUPS\dm9110,Heidelberg Digimaster 9110 (PS),\
     110ppm HiVolume DANKA Stuttgart]
   comment:[110ppm HiVolume DANKA Stuttgart]
 [....]
</screen>

<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">rpcclient <replaceable>SAMBA-CUPS</replaceable> -U root%<replaceable>secret</replaceable> -c 'setdriver <replaceable>dm9110</replaceable> <replaceable>mydrivername</replaceable>'</userinput>
 cmd = setdriver dm9110 mydrivername
 Successfully set dm9110 to mydrivername.
</screen>

<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">rpcclient <replaceable>SAMBA-CUPS</replaceable> -U root%<replaceable>secret</replaceable> -c 'enumprinters'</userinput>
 cmd = enumprinters
   flags:[0x800000]
   name:[\\SAMBA-CUPS\dm9110]
   description:[\\SAMBA-CUPS\dm9110,mydrivername,\
     110ppm HiVolume DANKA Stuttgart]
   comment:[110ppm HiVolume DANKA Stuttgart]
 [....]
</screen></para>

<para>
It may not be easy to recognize that the first call to <literal>enumprinters</literal> showed the
<quote>dm9110</quote> printer with an empty string where the driver should have been listed (between
the two commas in the description field). After the <literal>setdriver</literal> command
succeeds, all is well.
</para>
</sect2>

<sect2>
<title>Adding New Printers with the Windows NT APW</title>

<para>
By default, Samba exhibits all printer shares defined in <filename moreinfo="none">smb.conf</filename> in the <guiicon moreinfo="none">Printers</guiicon>
folder. Also located in this folder is the Windows NT Add Printer Wizard icon. The APW will be shown only if:
</para>

<itemizedlist>
	<listitem><para>
	The connected user is able to successfully execute an <literal>OpenPrinterEx(\\server)</literal> with
	administrative privileges (i.e., root or <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTERADMIN" xlink:href="smb.conf.5.html#PRINTERADMIN">printer admin</link>).
	</para>

	<tip><para> Try this from a Windows 200x/XP DOS box command prompt:
	</para>

	<para><userinput moreinfo="none">
	runas /netonly /user:root rundll32 printui.dll,PrintUIEntry /p /t0 /n \\<replaceable>SAMBA-SERVER</replaceable>\<replaceable>printersharename</replaceable>
	</userinput></para>

	<para>
	Click on <guibutton moreinfo="none">Printing Preferences</guibutton>.
	</para></tip></listitem>

	<listitem><para>... contains the setting
	<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SHOWADDPRINTERWIZARD" xlink:href="smb.conf.5.html#SHOWADDPRINTERWIZARD">show add printer wizard = yes</link> (the
	default).</para></listitem>
</itemizedlist>

<para>
The APW can do various things:
</para>

<itemizedlist>
	<listitem><para>
	Upload a new driver to the Samba <parameter>[print$]</parameter> share.
	</para></listitem>

	<listitem><para>
	Associate an uploaded driver with an existing (but still driverless) print queue.
	</para></listitem>

	<listitem><para>
	Exchange the currently used driver for an existing print queue with one that has been uploaded before.
	</para></listitem>

	<listitem><para>
	Add an entirely new printer to the Samba host (only in conjunction with a working
	<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="ADDPRINTERCOMMAND" xlink:href="smb.conf.5.html#ADDPRINTERCOMMAND">add printer command</link>. A corresponding
	<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="DELETEPRINTERCOMMAND" xlink:href="smb.conf.5.html#DELETEPRINTERCOMMAND">delete printer command</link> for removing entries from the
	<guiicon moreinfo="none">Printers</guiicon> folder may also be provided).
	</para></listitem>
</itemizedlist>

<para>
The last one (add a new printer) requires more effort than the previous ones. To use the APW to successfully
add a printer to a Samba server, the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="ADDPRINTERCOMMAND" xlink:href="smb.conf.5.html#ADDPRINTERCOMMAND">add printer command</link> must have a defined value.
The program hook must successfully add the printer to the UNIX print system (i.e., to
<filename moreinfo="none">/etc/printcap</filename>, <filename moreinfo="none">/etc/cups/printers.conf</filename> or other appropriate files)
and to <filename moreinfo="none">smb.conf</filename> if necessary.
</para>

<para>
When using the APW from a client, if the named printer share does not exist, smbd will execute the
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="ADDPRINTERCOMMAND" xlink:href="smb.conf.5.html#ADDPRINTERCOMMAND">add printer command</link> and reparse to attempt to locate the new printer share. If the
share is still not defined, an error of "<errorname>Access Denied"</errorname> is returned to the client. The
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="ADDPRINTERCOMMAND" xlink:href="smb.conf.5.html#ADDPRINTERCOMMAND">add printer command</link> is executed under the context of the connected user, not
necessarily a root account. A <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="MAPTOGUEST" xlink:href="smb.conf.5.html#MAPTOGUEST">map to guest = bad user</link> may have connected
you unwittingly under the wrong privilege. You should check it by using the <literal>smbstatus</literal>
command.
</para>

</sect2>

<sect2>
<title>Error Message: <quote>Cannot connect under a different Name</quote></title>

<para>
Once you are connected with the wrong credentials, there is no means to reverse the situation other than
to close all Explorer windows, and perhaps reboot.
</para>

<itemizedlist>
	<listitem><para>
<indexterm significance="normal"><primary>net use</primary></indexterm>
	The <literal>net use \\SAMBA-SERVER\sharename /user:root</literal> gives you an error message:
	<quote>Multiple connections to a server or a shared resource by the same user utilizing
	several user names are not allowed. Disconnect all previous connections to the server,
	esp. the shared resource, and try again.</quote>
	</para></listitem>

	<listitem><para>
	Every attempt to <quote>connect a network drive</quote> to <filename moreinfo="none">\\SAMBASERVER\\print$</filename>
	to <constant>z:</constant> is countered by the pertinacious message: <quote>This
	network folder is currently connected under different credentials (username and password).
	Disconnect first any existing connection to this network share in order to connect again under
	a different username and password</quote>.
	</para></listitem>
</itemizedlist>

<para>
So you close all connections. You try again. You get the same message. You check from the Samba side, using
<literal>smbstatus</literal>. Yes, there are more connections. You kill them all. The client still gives you
the same error message. You watch the smbd.log file on a high debug level and try reconnect. Same error
message, but not a single line in the log. You start to wonder if there was a connection attempt at all. You
run ethereal and tcpdump while you try to connect. Result: not a single byte goes on the wire. Windows still
gives the error message. You close all Explorer windows and start it again. You try to connect <?latex --- ?> and
this times it works!  Windows seems to cache connection information somewhere and does not keep it up to date
(if you are unlucky, you might need to reboot to get rid of the error message).
</para>

<para>
The easiest way to forcefully terminate all connections from your client to a server is by executing:
<screen format="linespecific">
<prompt moreinfo="none">C:\&gt; </prompt> net use * /delete
</screen>
This will also disconnect all mapped drives and will allow you create fresh connection as required.
</para>
</sect2>

<sect2>
<title>Take Care When Assembling Driver Files</title>

<para>
You need to be extremely careful when you take notes about the files belonging to a particular
driver. Don't confuse the files for driver version <quote>0</quote> (for Windows 9x/Me, going into
<filename moreinfo="none">[print$]/WIN/0/</filename>), driver version <filename moreinfo="none">2</filename> (kernel mode driver for Windows NT,
going into <filename moreinfo="none">[print$]/W32X86/2/</filename>; may be used on Windows 200x/XP also), and
driver version <quote>3</quote> (non-kernel mode driver going into <filename moreinfo="none">[print$]/W32X86/3/</filename>;
cannot be used on Windows NT). Quite often these different driver versions contain
files that have the same name but actually are very different. If you look at them from
the Windows Explorer (they reside in <filename moreinfo="none">%WINDOWS%\system32\spool\drivers\W32X86\</filename>),
you will probably see names in capital letters, while an <literal>enumdrivers</literal> command from Samba
would show mixed or lowercase letters, so it is easy to confuse them. If you install them manually using
<literal>rpcclient</literal> and subcommands, you may even succeed without an error message. Only later,
when you try install on a client, you will encounter error messages like <computeroutput moreinfo="none">This server
has no appropriate driver for the printer</computeroutput>.
</para>

<para>
Here is an example. You are invited to look closely at the various files, compare their names and
their spelling, and discover the differences in the composition of the version 2 and 3 sets. Note: the
version 0 set contained 40 <parameter moreinfo="none">Dependentfiles</parameter>, so I left it out for space reasons:
</para>

<para><screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">rpcclient -U 'Administrator%<replaceable>secret</replaceable>' -c 'enumdrivers 3' 10.160.50.8 </userinput>

 Printer Driver Info 3:
         Version: [3]
         Driver Name: [Canon iR8500 PS3]
         Architecture: [Windows NT x86]
         Driver Path: [\\10.160.50.8\print$\W32X86\3\cns3g.dll]
         Datafile: [\\10.160.50.8\print$\W32X86\3\iR8500sg.xpd]
         Configfile: [\\10.160.50.8\print$\W32X86\3\cns3gui.dll]
         Helpfile: [\\10.160.50.8\print$\W32X86\3\cns3g.hlp]
 
         Dependentfiles: [\\10.160.50.8\print$\W32X86\3\aucplmNT.dll]
         Dependentfiles: [\\10.160.50.8\print$\W32X86\3\ucs32p.dll]
         Dependentfiles: [\\10.160.50.8\print$\W32X86\3\tnl32.dll]
         Dependentfiles: [\\10.160.50.8\print$\W32X86\3\aussdrv.dll]
         Dependentfiles: [\\10.160.50.8\print$\W32X86\3\cnspdc.dll]
         Dependentfiles: [\\10.160.50.8\print$\W32X86\3\aussapi.dat]
         Dependentfiles: [\\10.160.50.8\print$\W32X86\3\cns3407.dll]
         Dependentfiles: [\\10.160.50.8\print$\W32X86\3\CnS3G.cnt]
         Dependentfiles: [\\10.160.50.8\print$\W32X86\3\NBAPI.DLL]
         Dependentfiles: [\\10.160.50.8\print$\W32X86\3\NBIPC.DLL]
         Dependentfiles: [\\10.160.50.8\print$\W32X86\3\cpcview.exe]
         Dependentfiles: [\\10.160.50.8\print$\W32X86\3\cpcdspl.exe]
         Dependentfiles: [\\10.160.50.8\print$\W32X86\3\cpcedit.dll]
         Dependentfiles: [\\10.160.50.8\print$\W32X86\3\cpcqm.exe]
         Dependentfiles: [\\10.160.50.8\print$\W32X86\3\cpcspl.dll]
         Dependentfiles: [\\10.160.50.8\print$\W32X86\3\cfine32.dll]
         Dependentfiles: [\\10.160.50.8\print$\W32X86\3\cpcr407.dll]
         Dependentfiles: [\\10.160.50.8\print$\W32X86\3\Cpcqm407.hlp]
         Dependentfiles: [\\10.160.50.8\print$\W32X86\3\cpcqm407.cnt]
         Dependentfiles: [\\10.160.50.8\print$\W32X86\3\cns3ggr.dll]
 
         Monitorname: []
         Defaultdatatype: []

 Printer Driver Info 3:
         Version: [2]
         Driver Name: [Canon iR5000-6000 PS3]
         Architecture: [Windows NT x86]
         Driver Path: [\\10.160.50.8\print$\W32X86\2\cns3g.dll]
         Datafile: [\\10.160.50.8\print$\W32X86\2\IR5000sg.xpd]
         Configfile: [\\10.160.50.8\print$\W32X86\2\cns3gui.dll]
         Helpfile: [\\10.160.50.8\print$\W32X86\2\cns3g.hlp]
 
         Dependentfiles: [\\10.160.50.8\print$\W32X86\2\AUCPLMNT.DLL]
         Dependentfiles: [\\10.160.50.8\print$\W32X86\2\aussdrv.dll]
         Dependentfiles: [\\10.160.50.8\print$\W32X86\2\cnspdc.dll]
         Dependentfiles: [\\10.160.50.8\print$\W32X86\2\aussapi.dat]
         Dependentfiles: [\\10.160.50.8\print$\W32X86\2\cns3407.dll]
         Dependentfiles: [\\10.160.50.8\print$\W32X86\2\CnS3G.cnt]
         Dependentfiles: [\\10.160.50.8\print$\W32X86\2\NBAPI.DLL]
         Dependentfiles: [\\10.160.50.8\print$\W32X86\2\NBIPC.DLL]
         Dependentfiles: [\\10.160.50.8\print$\W32X86\2\cns3gum.dll]
 
         Monitorname: [CPCA Language Monitor2]
         Defaultdatatype: []

</screen></para>

<para>
If we write the <quote>version 2</quote> files and the <quote>version 3</quote> files
into different text files and compare the result, we see this
picture:
</para>

<para><screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">sdiff 2-files 3-files</userinput>


 cns3g.dll                     cns3g.dll
 iR8500sg.xpd                  iR8500sg.xpd
 cns3gui.dll                   cns3gui.dll
 cns3g.hlp                     cns3g.hlp
 AUCPLMNT.DLL                | aucplmNT.dll
                             &gt; ucs32p.dll
                             &gt; tnl32.dll
 aussdrv.dll                   aussdrv.dll
 cnspdc.dll                    cnspdc.dll
 aussapi.dat                   aussapi.dat
 cns3407.dll                   cns3407.dll
 CnS3G.cnt                     CnS3G.cnt
 NBAPI.DLL                     NBAPI.DLL
 NBIPC.DLL                     NBIPC.DLL
 cns3gum.dll                 | cpcview.exe
                             &gt; cpcdspl.exe 
                             &gt; cpcqm.exe
                             &gt; cpcspl.dll
                             &gt; cfine32.dll
                             &gt; cpcr407.dll
                             &gt; Cpcqm407.hlp
                             &gt; cpcqm407.cnt
                             &gt; cns3ggr.dll

</screen>

Do not be fooled! Driver files for each version with identical
names may be different in their content, as you can see from this size
comparison:
</para>

<para><screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">for i in cns3g.hlp cns3gui.dll cns3g.dll; do                  \
           smbclient //10.160.50.8/print\$ -U 'Administrator%xxxx' \
           -c "cd W32X86/3; dir $i; cd .. ; cd 2; dir $i";      \
		   done</userinput>

  CNS3G.HLP               A   122981  Thu May 30 02:31:00 2002
  CNS3G.HLP               A    99948  Thu May 30 02:31:00 2002

  CNS3GUI.DLL             A  1805824  Thu May 30 02:31:00 2002
  CNS3GUI.DLL             A  1785344  Thu May 30 02:31:00 2002

  CNS3G.DLL               A  1145088  Thu May 30 02:31:00 2002
  CNS3G.DLL               A    15872  Thu May 30 02:31:00 2002
</screen></para>

<para>
In my example were even more differences than shown here. Conclusion: you must be careful to select the
correct driver files for each driver version. Don't rely on the names alone, and don't interchange files
belonging to different driver versions.
</para>
</sect2>

<sect2>
<title>Samba and Printer Ports</title>

<para>
<indexterm significance="normal"><primary>LPT1:</primary></indexterm>
<indexterm significance="normal"><primary>COM1:</primary></indexterm>
<indexterm significance="normal"><primary>FILE:</primary></indexterm>
<indexterm significance="normal"><primary>available port</primary></indexterm>
Windows NT/2000 print servers associate a port with each printer. These normally take the form of
<filename moreinfo="none">LPT1:</filename>, <filename moreinfo="none">COM1:</filename>, <filename moreinfo="none">FILE:</filename>, and so on. Samba must also
support the concept of ports associated with a printer. By default, only one printer port, named <quote>Samba
Printer Port</quote>, exists on a system. Samba does not really need such a <quote>port</quote> in order to
print; rather it is a requirement of Windows clients. They insist on being told about an available port when
they request this information; otherwise, they throw an error message at you. So Samba fakes the port
information to keep the Windows clients happy.
</para>

<para>
<indexterm significance="normal"><primary>Printer Pooling</primary></indexterm>
Samba does not support the concept of <constant>Printer Pooling</constant> internally either. Printer
pooling assigns a logical printer to multiple ports as a form of load balancing or failover.
</para>

<para>
If you require multiple ports to be defined for some reason or another (my users and my boss should not know
that they are working with Samba), configure the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="ENUMPORTSCOMMAND" xlink:href="smb.conf.5.html#ENUMPORTSCOMMAND">enumports command</link>,
which can be used to define an external program that generates a listing of ports on a system.
</para>
</sect2>

<sect2>
<title>Avoiding Common Client Driver Misconfiguration</title>

<para>
So now the printing works, but there are still problems. Most jobs print well, some do not print at
all. Some jobs have problems with fonts, which do not look good. Some jobs print fast and some
are dead-slow. We cannot cover it all, but we want to encourage you to read the brief paragraph about
<quote>Avoiding the Wrong PostScript Driver Settings</quote> in <link linkend="CUPS-printing">CUPS Printing
Chapter</link>, <link linkend="cups-avoidps1">Avoiding Critical PostScript Driver Settings on the
Client</link>.
</para>
</sect2>
</sect1>

<sect1>
<title>The Imprints Toolset</title>

<para>
<indexterm significance="normal"><primary>Imprints</primary></indexterm>
The Imprints tool set provides a UNIX equivalent of the Windows NT APW.  For complete information, please
refer to the <ulink url="http://imprints.sourceforge.net/">Imprints</ulink> Web site as well as the
documentation included with the Imprints source distribution. This section provides only a brief introduction
to the features of Imprints.
</para>

<para>
Unfortunately, the Imprints toolset is no longer maintained. As of December 2000, the project is in
need of a new maintainer. The most important skill to have is Perl coding and an interest in MS-RPC-based
printing used in Samba. If you wish to volunteer, please coordinate your efforts on the Samba technical
mailing list. The toolset is still in usable form, but only for a series of older printer models where
there are prepared packages to use. Packages for more up-to-date print devices are needed if Imprints
should have a future. Information regarding the Imprints toolset can be obtained from the <ulink url="http://imprints.sourceforge.net/">Imprints</ulink> home page.
</para>

<sect2>
<title>What Is Imprints?</title>

<para>
Imprints is a collection of tools for supporting these goals:
</para>

<itemizedlist>
	<listitem><para>
	Providing a central repository of information regarding Windows NT and 95/98 printer driver packages.
	</para></listitem>

	<listitem><para>
	Providing the tools necessary for creating the Imprints printer driver packages.
	</para></listitem>

	<listitem><para>
	Providing an installation client that will obtain printer drivers from a central Internet (or intranet) Imprints Server
	repository and install them on remote Samba and Windows NT4 print servers.
	</para></listitem>
</itemizedlist>
</sect2>

<sect2>
<title>Creating Printer Driver Packages</title>

<para>
The process of creating printer driver packages is beyond the scope of this document (refer to Imprints.txt,
included with the Samba distribution for more information). In short, an Imprints driver package
is a gzipped tarball containing the driver files, related INF files, and a control file needed by the
installation client.
</para>
</sect2>

<sect2>
<title>The Imprints Server</title>

<para>
The Imprints server is really a database server that may be queried via standard HTTP mechanisms. Each
printer entry in the database has an associated URL for the actual downloading of the package. Each
package is digitally signed via GnuPG, which can be used to verify that
the package downloaded is actually
the one referred in the Imprints database. It is strongly recommended that this security check
not be disabled.
</para>
</sect2>

<sect2>
<title>The Installation Client</title>

<para>
More information regarding the Imprints installation client is available from the documentation file
<filename moreinfo="none">Imprints-Client-HOWTO.ps</filename> that is included with the Imprints source package. The Imprints
installation client comes in two forms:
</para>

<itemizedlist>
	<listitem><para>A set of command-line Perl scripts.</para></listitem>
	<listitem><para>A GTK+-based graphical interface to the command-line Perl scripts.</para></listitem>
</itemizedlist>

<para>
The installation client (in both forms) provides a means of querying the Imprints database server for
a matching list of known printer model names as well as a means to download and install the drivers on
remote Samba and Windows NT print servers.
</para>

<para>
The basic installation process is in four steps, and Perl code is wrapped around smbclient and rpcclient.
</para>

<itemizedlist>
	<listitem><para>
	For each supported architecture for a given driver:
	<orderedlist inheritnum="ignore" continuation="restarts">
		<listitem><para>rpcclient: Get the appropriate upload directory on the remote server.</para></listitem>
		<listitem><para>smbclient: Upload the driver files.</para></listitem>
		<listitem><para>rpcclient: Issues an AddPrinterDriver() MS-RPC.</para></listitem>
	</orderedlist>
	</para></listitem>

	<listitem><para>rpcclient: Issues an AddPrinterEx() MS-RPC to actually create the printer.</para></listitem>
</itemizedlist>

<para>
One of the problems encountered when implementing the Imprints tool set was the namespace issues between
various supported client architectures. For example, Windows NT includes a driver named <quote>Apple LaserWriter
II NTX v51.8</quote>, and Windows 95 calls its version of this driver <quote>Apple LaserWriter II NTX</quote>.
</para>

<para>
The problem is how to know what client drivers have been uploaded for a printer. An astute reader will
remember that the Windows NT Printer Properties dialog only includes space for one printer driver name. A
quick look in the Windows NT 4.0 system registry at:
</para>

<para><filename moreinfo="none">
 HKLM\System\CurrentControlSet\Control\Print\Environment
</filename></para>

<para>
will reveal that Windows NT always uses the NT driver name. This is okay because Windows NT always requires
that at least the Windows NT version of the printer driver is present. Samba does not have the
requirement internally; therefore, <quote>How can you use the NT driver name if it has not already been installed?</quote>
</para>

<para>
The way of sidestepping this limitation is to require that all Imprints printer driver packages include both the Intel Windows NT and
95/98 printer drivers and that the NT driver is installed first.
</para>
</sect2>
</sect1>

<sect1>
<title>Adding Network Printers without User Interaction</title>

<para>
The following MS Knowledge Base article may be of some help if you need to handle Windows 2000 clients:
<emphasis>How to Add Printers with No User Interaction in Windows 2000,</emphasis> (<ulink url="http://support.microsoft.com/default.aspx?scid=kb;en-us;189105">Microsoft KB 189105</ulink>).  It also
applies to Windows XP Professional clients.  The ideas sketched out in this section are inspired by this
article, which describes a command-line method that can be applied to install network and local printers and
their drivers. This is most useful if integrated in Logon Scripts. You can see what options are available by
typing in the command prompt (<literal>DOS box</literal>):
</para>

<para><userinput moreinfo="none">rundll32 printui.dll,PrintUIEntry /?</userinput></para>

<para>
A window pops up that shows you all of the command-line switches available. An extensive list of examples
is also provided. This is only for Windows 200x/XP; it does not work on Windows NT. Windows NT probably has
some other tools in the respective Resource Kit. Here is a suggestion about what a client logon script
might contain, with a short explanation of what the lines actually do (it works if 200x/XP Windows
clients access printers via Samba, and works for Windows-based print servers too):
</para>

<para><screen format="linespecific">
<userinput moreinfo="none">rundll32 printui.dll,PrintUIEntry /dn /n "\\cupsserver\infotec2105-IPDS" /q</userinput>
<userinput moreinfo="none">rundll32 printui.dll,PrintUIEntry /in /n "\\cupsserver\infotec2105-PS"</userinput>
<userinput moreinfo="none">rundll32 printui.dll,PrintUIEntry /y /n "\\cupsserver\infotec2105-PS"</userinput>
</screen></para>

<para>
Here is a list of the used command-line parameters: 
</para>

<variablelist>
	<varlistentry><term>/dn</term>
		<listitem><para>deletes a network printer.</para></listitem>
	</varlistentry>
	<varlistentry><term>/q</term>
		<listitem><para>quiet modus.</para></listitem>
	</varlistentry>
	<varlistentry><term>/n</term>
		<listitem><para>names a printer.</para></listitem>
	</varlistentry>
	<varlistentry><term>/in</term>
		<listitem><para>adds a network printer connection.</para></listitem>
	</varlistentry>
	<varlistentry><term>/y</term>
		<listitem><para>sets printer as default printer.</para></listitem>
	</varlistentry>
</variablelist>

<itemizedlist>
	<listitem><para>
	Line 1 deletes a possibly existing previous network printer <emphasis>infotec2105-IPDS</emphasis>
	(which had used native Windows drivers with LPRng that were removed from the server that was
	converted to CUPS). The <literal>/q</literal> at the end prevents confirm
	or error dialog boxes from popping up. They should not be presented to the user logging on.
	</para></listitem>

	<listitem><para>	
	Line 2 adds the new printer
	<emphasis>infotec2105-PS</emphasis> (which actually is the same
	physical device but is now run by the new CUPS printing system and associated with the
	CUPS/Adobe PS drivers). The printer and its driver must have been added to Samba prior to
	the user logging in (e.g., by a procedure as discussed earlier in this chapter or by running
	<literal>cupsaddsmb</literal>). The driver is now autodownloaded to the client PC where the
	user is about to log in.
	</para></listitem>

	<listitem><para>
	Line 3 sets the default printer to this new network printer (there might be several other
	printers installed with this same method, and some may be local as well, so we decide for a
	default printer). The default printer selection may, of course, be different for different users.
	</para></listitem>
</itemizedlist>

<para>
The second line only works if the printer <emphasis>infotec2105-PS</emphasis> has an already working
print queue on the <constant>cupsserver</constant> and if the
printer drivers have been successfully uploaded
(via the <literal>APW</literal>, <literal>smbclient/rpcclient</literal>, or <literal>cupsaddsmb</literal>)
into the <parameter>[print$]</parameter> driver repository of Samba. Some Samba versions
prior to version 3.0 required a restart of smbd after the printer install and the driver upload;
otherwise the script (or any other client driver download) would fail.
</para>

<para>
Since there is no easy way to test for the existence of an installed network printer from the logon script,
do not bother checking. Just allow the de-installation/re-installation to occur every time a user logs in;
it's really quick anyway (1 to 2 seconds).
</para>

<para>
The additional benefits for this are:
</para>

<itemizedlist>
	<listitem><para>
	It puts in place any printer default setup changes automatically at every user logon.
	</para></listitem>

	<listitem><para>
	It allows for <quote>roaming</quote> users' login to the domain from different workstations.
	</para></listitem>
</itemizedlist>

<para>
Since network printers are installed per user, this much simplifies the process of keeping the installation
up to date. The few extra seconds at logon time will not really be noticeable. Printers can be centrally
added, changed, and deleted at will on the server with no user intervention required from the clients
(you just need to keep the logon scripts up to date).
</para>
</sect1>

<sect1>
<title>The <literal>addprinter</literal> Command</title>

<para>
The <literal>addprinter</literal> command can be configured to be a shell script or program executed by
Samba. It is triggered by running the APW from a client against the Samba print server. The APW asks
the user to fill in several fields (such as printer name, driver to be used, comment, port monitor,
and so on). These parameters are passed on to Samba by the APW. If the addprinter command is designed in a
way that it can create a new printer (through writing correct printcap entries on legacy systems or
by executing the <literal>lpadmin</literal> command on more modern systems) and create the associated share,
then the APW will in effect really create a new printer on Samba and the UNIX print subsystem!
</para>
</sect1>

<sect1>
<title>Migration of Classical Printing to Samba</title>

<para>
The basic NT-style printer driver management has not changed considerably in 3.0 over the 2.2.x releases
(apart from many small improvements). Here migration should be quite easy, especially if you followed
previous advice to stop using deprecated parameters in your setup. For migrations from an existing 2.0.x
setup, or if you continued Windows 9x/Me-style printing in your Samba 2.2 installations, it is more of
an effort. Please read the appropriate release notes and the HOWTO Collection for Samba-2.2.x. You can
follow several paths. Here are possible scenarios for migration:
</para>

<itemizedlist>
	<listitem><para>
	You need to study and apply the new Windows NT printer and driver support. Previously used
	parameters <parameter moreinfo="none">printer driver file</parameter>, <parameter moreinfo="none">printer driver</parameter>,
	and <parameter moreinfo="none">printer driver location</parameter> are no longer supported.
	</para></listitem>

	<listitem><para>
	If you want to take advantage of Windows NT printer driver support, you also need to migrate the
	Windows 9x/Me drivers to the new setup.
	</para></listitem>

	<listitem><para>
	An existing <filename moreinfo="none">printers.def</filename> file (the one specified in the now removed parameter
	<parameter moreinfo="none">printer driver file</parameter>) will no longer work with Samba-3. In 3.0, smbd attempts
	to locate Windows 9x/Me driver files for the printer in <parameter>[print$]</parameter>
	and additional settings in the TDB and only there; if it fails, it will <emphasis>not</emphasis>
	(as 2.2.x used to do) drop down to using a <filename moreinfo="none">printers.def</filename> (and all associated
	parameters). The make_printerdef tool is removed and there is no backward compatibility for this.
	</para></listitem>

	<listitem><para>You need to install a Windows 9x/Me driver into the
	<parameter>[print$]</parameter> share for a printer on your Samba
	host. The driver files will be stored in the <quote>WIN40/0</quote> subdirectory of
	<parameter>[print$]</parameter>, and some other settings and information go
	into the printing-related TDBs.</para></listitem>

	<listitem><para>
	If you want to migrate an existing <filename moreinfo="none">printers.def</filename> file into the new setup, the only current
	solution is to use the Windows NT APW to install the NT drivers and the 9x/Me drivers. This can be scripted
	using smbclient and rpcclient. See the Imprints installation client on the <ulink noescape="1" url="http://imprints.sourceforge.net/">Imprints</ulink> web site for example. See also the discussion of
	rpcclient usage in <link linkend="CUPS-printing">CUPS Printing</link>.
	</para></listitem>
</itemizedlist>
</sect1>

<sect1>
<title>Publishing Printer Information in Active Directory or LDAP</title>

<para>
This topic has also been addressed in <link linkend="NetCommand">Remote and Local Management <?latex --- ?> The
Net Command</link>. If you wish to volunteer your services to help document this further, please contact 
<ulink url="mail://jht@samba.org">John H. Terpstra</ulink>.
</para>
</sect1>

<sect1>
<title>Common Errors</title>

<sect2>
<title>I Give My Root Password but I Do Not Get Access</title>

<para>
Do not confuse the root password, which is valid for the UNIX system (and in most cases stored in the
form of a one-way hash in a file named <filename moreinfo="none">/etc/shadow</filename>), with the password used to
authenticate against Samba. Samba does not know the UNIX password. Root access to Samba resources
requires that a Samba account for root must first be created. This is done with the <literal>smbpasswd</literal>
command as follows:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> smbpasswd -a root
New SMB password: secret
Retype new SMB password: secret
</screen>
</para>

</sect2>

<sect2>
<title>My Print Jobs Get Spooled into the Spooling Directory, but Then Get Lost</title>

<para>
Do not use the existing UNIX print system spool directory for the Samba spool directory. It may seem
convenient and a savings of space, but it only leads to problems. The two must be separate. The UNIX/Linux
system print spool directory (e.g., <filename moreinfo="none">/var/spool/cups</filename>) is typically owned by a
non-privileged user such as <literal moreinfo="none">cups</literal> or <literal moreinfo="none">lp</literal>. Additionally. the permissions on
the spool directory are typically restrictive to the owner and/or group. On the other hand, the Samba
spool directory must be world writable, and should have the 't' bit set to ensure that only a temporary
spool file owner can change or delete the file.
</para>

<para>
Depending on the type of print spooling system in use on the UNIX/Linux host, files that the spool
management application finds and that are not currently part of job queue that it is managing can be deleted.
This may explain the observation that jobs are spooled (by Samba) into this directory and just disappear.
</para>

</sect2>
</sect1>

</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="CUPS-printing">

<chapterinfo>

	<author>
		<firstname>Kurt</firstname><surname>Pfeifle</surname>
		<affiliation>
			<orgname>Danka Deutschland GmbH </orgname>
                        <address format="linespecific"><email>kpfeifle@danka.de</email></address>
		</affiliation>
	</author>
	<author>
		<firstname>Ciprian</firstname><surname>Vizitiu</surname>
		<affiliation>
			<address format="linespecific"><email>CVizitiu@gbif.org</email></address>
		</affiliation>
		<contrib>drawings</contrib>
	</author>

	<author>
<firstname>Jelmer</firstname><othername>R.</othername><surname>Vernooij</surname><othername>R.</othername>
<affiliation>
	<orgname>The Samba Team</orgname>
	<address format="linespecific"><email>jelmer@samba.org</email></address>
</affiliation><contrib>drawings</contrib></author>

	<pubdate> (27 Jan 2004) </pubdate>
</chapterinfo>

<title>CUPS Printing Support</title>

<sect1>

	<title>Introduction</title>

	<sect2>
		<title>Features and Benefits</title>

		<para>
<indexterm significance="normal"><primary>default printing</primary></indexterm>
		The Common UNIX Print System (<ulink url="http://www.cups.org/">CUPS</ulink>)
		has become quite popular. All major Linux distributions now ship it as their default printing
		system. To many, it is still a mystical tool. Mostly, it just works.  People tend to regard
		it as a <quote>black box</quote> that they do not want to look into as long as it works. But once
		there is a little problem, they have trouble finding out where to start debugging it. Refer to
		<link linkend="classicalprinting">Classical Printing</link>, which contains much information
		that is also relevant to CUPS.
		</para>

		<para>
<indexterm significance="normal"><primary>CUPS</primary></indexterm>
		CUPS sports quite a few unique and powerful features. While its basic functions may be grasped quite
		easily, they are also new. Because it is different from other, more traditional printing systems, it is best
		not to try to apply any prior knowledge about printing to this new system. Rather, try to understand CUPS from
		the beginning. This documentation will lead you to a complete understanding of CUPS. Let's start with the most
		basic things first.
		</para>

	</sect2>

	<sect2>
	<title>Overview</title>

	<para>
<indexterm significance="normal"><primary>print spooling system</primary></indexterm>
<indexterm significance="normal"><primary>CUPS</primary></indexterm>
<indexterm significance="normal"><primary>printer management system</primary></indexterm>
<indexterm significance="normal"><primary>IETF</primary></indexterm>
<indexterm significance="normal"><primary>Internet Printing Protocol</primary><see>IPP</see></indexterm>
<indexterm significance="normal"><primary>Internet Engineering Task Force</primary><see>IETF</see></indexterm>
<indexterm significance="normal"><primary>GUI</primary></indexterm>
<indexterm significance="normal"><primary>KDEPrint</primary></indexterm>
	CUPS is more than just a print spooling system. It is a complete printer management system that
	complies with the new Internet Printing Protocol (IPP). IPP is an industry and Internet Engineering Task Force
	(IETF) standard for network printing. Many of its functions can be managed remotely (or locally) via a Web
	browser (giving you platform-independent access to the CUPS print server). Additionally, it has the
	traditional command line and several more modern GUI interfaces (GUI interfaces developed by third parties,
	like KDE's overwhelming <ulink url="http://printing.kde.org/">KDEPrint</ulink>).
	</para>

	<para>
<indexterm significance="normal"><primary>raw printers</primary></indexterm>
<indexterm significance="normal"><primary>smart printers</primary></indexterm>
	CUPS allows creation of <emphasis>raw</emphasis> printers (i.e., no print file format translation) as
	well as <emphasis>smart</emphasis> printers (i.e., CUPS does file format conversion as required for the
	printer). In many ways, this gives CUPS capabilities similar to the MS Windows print monitoring system. Of
	course, if you are a CUPS advocate, you would argue that CUPS is better! In any case, let us now explore how
	to configure CUPS for interfacing with MS Windows print clients via Samba.
	</para>

	</sect2>

</sect1>

<sect1>
	<title>Basic CUPS Support Configuration</title>

	<para>
<indexterm significance="normal"><primary>CUPS</primary></indexterm>
<indexterm significance="normal"><primary>cupsd.conf</primary></indexterm>
<indexterm significance="normal"><primary>/etc/printcap</primary></indexterm>
<indexterm significance="normal"><primary>Printcap</primary></indexterm>
<indexterm significance="normal"><primary>PrintcapFormat</primary></indexterm>
Printing with CUPS in the most basic <filename moreinfo="none">smb.conf</filename> setup in Samba-3.0 (as was true for 2.2.x) requires just two
parameters: <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTING" xlink:href="smb.conf.5.html#PRINTING">printing = cups</link> and <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTCAP" xlink:href="smb.conf.5.html#PRINTCAP">printcap = cups</link>. CUPS does not need a printcap file.  However, the
<filename moreinfo="none">cupsd.conf</filename> configuration file knows of two related directives that control how such a
file will be automatically created and maintained by CUPS for the convenience of third-party applications
(example: <parameter moreinfo="none">Printcap /etc/printcap</parameter> and <parameter moreinfo="none">PrintcapFormat BSD</parameter>).
Legacy programs often require the existence of a printcap file containing printer names or they will refuse to
print. Make sure CUPS is set to generate and maintain a printcap file. For details, see <literal>man
cupsd.conf</literal> and other CUPS-related documentation, like the wealth of documents regarding the CUPS
server itself available from the <ulink noescape="1" url="http://localhost:631/documentation.html">CUPS</ulink> web site.
	</para>

	<sect2>
	<title>Linking smbd with libcups.so</title>

	<para>
<indexterm significance="normal"><primary>libcups.so</primary></indexterm>
	Samba has a special relationship to CUPS. Samba can be compiled with CUPS library support.
	Most recent installations have this support enabled. By default, CUPS linking is compiled
	into smbd and other Samba binaries. Of course, you can use CUPS even
	if Samba is not linked against <filename moreinfo="none">libcups.so</filename> <?latex --- ?> but
	there are some differences in required or supported configuration.
	</para>

	<para>
<indexterm significance="normal"><primary>libcups</primary></indexterm>
<indexterm significance="normal"><primary>ldd</primary></indexterm>
	When Samba is compiled and linked with <filename moreinfo="none">libcups</filename>, <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTCAP" xlink:href="smb.conf.5.html#PRINTCAP">printcap = cups</link>
	uses the CUPS API to list printers, submit jobs, query queues, and so on. Otherwise it maps to the System V
	commands with an additional <literal>-oraw</literal> option for printing. On a Linux
	system, you can use the <literal>ldd</literal> utility to find out if smbd has been linked with the
	libcups library (<literal>ldd</literal> may not be present on other OS platforms, or its function may be embodied
	by a different command):
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">ldd `which smbd`</userinput>
libssl.so.0.9.6 =&gt; /usr/lib/libssl.so.0.9.6 (0x4002d000)
libcrypto.so.0.9.6 =&gt; /usr/lib/libcrypto.so.0.9.6 (0x4005a000)
libcups.so.2 =&gt; /usr/lib/libcups.so.2 (0x40123000)
[....]
</screen>
	</para>

	<para>
<indexterm significance="normal"><primary>libcups.so.2</primary></indexterm>
	The line <computeroutput moreinfo="none">libcups.so.2 =&gt; /usr/lib/libcups.so.2 (0x40123000)</computeroutput> shows
	there is CUPS support compiled into this version of Samba. If this is the case, and printing = cups
	is set, then <emphasis>any otherwise manually set print command in <filename moreinfo="none">smb.conf</filename> is ignored</emphasis>.
	This is an important point to remember!
	</para>

	<tip><para> Should it be necessary, for any reason, to set your own print commands, you can do this by setting
	<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTING" xlink:href="smb.conf.5.html#PRINTING">printing = sysv</link>. However, you will lose all the benefits
	of tight CUPS-Samba integration. When you do this, you must manually configure the printing system commands
	(most important: 
	<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTCOMMAND" xlink:href="smb.conf.5.html#PRINTCOMMAND">print command</link>; other commands are
	<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LPPAUSECOMMAND" xlink:href="smb.conf.5.html#LPPAUSECOMMAND">lppause command</link>,
	<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LPRESUMECOMMAND" xlink:href="smb.conf.5.html#LPRESUMECOMMAND">lpresume command</link>,
	<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LPQCOMMAND" xlink:href="smb.conf.5.html#LPQCOMMAND">lpq command</link>,
	<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LPRMCOMMAND" xlink:href="smb.conf.5.html#LPRMCOMMAND">lprm command</link>,
	<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="QUEUEPAUSECOMMAND" xlink:href="smb.conf.5.html#QUEUEPAUSECOMMAND">queuepause command</link> and
	<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="QUEUERESUMECOMMAND" xlink:href="smb.conf.5.html#QUEUERESUMECOMMAND">queue resume command</link>).
	</para></tip>

	</sect2>

	<sect2>
	<title>Simple <filename moreinfo="none">smb.conf</filename> Settings for CUPS</title>

	<para>
	To summarize, <link linkend="cups-exam-simple">the Simplest Printing-Related 
	<filename moreinfo="none">smb.conf</filename> file</link> shows the simplest printing-related setup for <filename moreinfo="none">smb.conf</filename> to 
	enable basic CUPS support:
	</para>

	<example id="cups-exam-simple">
	<title>Simplest Printing-Related smb.conf</title>
	<simplelist>
	<member> </member><member><parameter>[global]</parameter></member>
	<member><indexterm>load printers</indexterm><parameter>load printers = yes</parameter></member>
	<member><indexterm>printing</indexterm><parameter>printing = cups</parameter></member>
	<member><indexterm>printcap name</indexterm><parameter>printcap name = cups</parameter></member>

	<member> </member><member><parameter>[printers]</parameter></member>
	<member><indexterm>comment</indexterm><parameter>comment = All Printers</parameter></member>
	<member><indexterm>path</indexterm><parameter>path = /var/spool/samba</parameter></member>
	<member><indexterm>browseable</indexterm><parameter>browseable = no</parameter></member>
	<member><indexterm>guest ok</indexterm><parameter>guest ok = yes</parameter></member>
	<member><indexterm>writable</indexterm><parameter>writable = no</parameter></member>
	<member><indexterm>printable</indexterm><parameter>printable = yes</parameter></member>
	<member><indexterm>printer admin</indexterm><parameter>printer admin = root, @ntadmins, @smbprintadm</parameter></member>
	</simplelist>
	</example>

	<para>
<indexterm significance="normal"><primary>PDF</primary></indexterm>
<indexterm significance="normal"><primary>PostScript</primary></indexterm>
<indexterm significance="normal"><primary>printer driver</primary></indexterm>
	This is all you need for basic printing setup for CUPS. It will print all graphic, text, PDF, and PostScript
	files submitted from Windows clients. However, most of your Windows users would not know how to send these
	kinds of files to print without opening a GUI application. Windows clients tend to have local printer drivers
	installed, and the GUI application's print buttons start a printer driver. Your users also rarely send files
	from the command line. Unlike UNIX clients, they rarely submit graphic, text, or PDF formatted files directly
	to the spooler. They nearly exclusively print from GUI applications with a <quote>printer driver</quote>
	hooked between the application's native format and the print data stream. If the backend printer is not a
	PostScript device, the print data stream is <quote>binary,</quote> sensible only for the target printer. Read
	on to learn what problem this may cause and how to avoid it.
	</para>

	</sect2>

	<sect2>
	<title>More Complex CUPS <filename moreinfo="none">smb.conf</filename> Settings</title>

	<para>
	<link linkend="overridesettings">The Overriding Global CUPS Settings for One Printer example</link> 
	is a slightly more complex printing-related setup for <filename moreinfo="none">smb.conf</filename>. It enables general CUPS printing
	support for all printers, but defines one printer share, which is set up differently. 
	</para>

	<example id="overridesettings">
	<title>Overriding Global CUPS Settings for One Printer</title>
	<simplelist>
	<member> </member><member><parameter>[global]</parameter></member>
	<member><indexterm>printing</indexterm><parameter>printing = cups</parameter></member>
	<member><indexterm>printcap name</indexterm><parameter>printcap name = cups</parameter></member>
	<member><indexterm>load printers</indexterm><parameter>load printers = yes</parameter></member>

	<member> </member><member><parameter>[printers]</parameter></member>
	<member><indexterm>comment</indexterm><parameter>comment = All Printers</parameter></member>
	<member><indexterm>path</indexterm><parameter>path = /var/spool/samba</parameter></member>
	<member><indexterm>guest ok</indexterm><parameter>guest ok = yes</parameter></member>
	<member><indexterm>writable</indexterm><parameter>writable = no</parameter></member>
	<member><indexterm>printable</indexterm><parameter>printable = yes</parameter></member>
	<member><indexterm>printer admin</indexterm><parameter>printer admin = root, @ntadmins, @smbprintadm</parameter></member>

	<member> </member><member><parameter>[special_printer]</parameter></member>
	<member><indexterm>comment</indexterm><parameter>comment = A special printer with his own settings</parameter></member>
	<member><indexterm>path</indexterm><parameter>path = /var/spool/samba-special</parameter></member>
	<member><indexterm>printing</indexterm><parameter>printing = sysv</parameter></member>
	<member><indexterm>printcap</indexterm><parameter>printcap = lpstat</parameter></member>
	<member><indexterm>print command</indexterm><parameter>print command = echo "NEW: `date`: printfile %f" &gt;&gt; /tmp/smbprn.log ; echo "     `date`: p-%p s-%s f-%f" &gt;&gt; /tmp/smbprn.log ; echo "     `date`: j-%j J-%J z-%z c-%c" &gt;&gt; /tmp/smbprn.log ; rm %f </parameter></member>
	<member><indexterm>guest ok</indexterm><parameter>guest ok = no</parameter></member>
	<member><indexterm>writable</indexterm><parameter>writable = no</parameter></member>
	<member><indexterm>printable</indexterm><parameter>printable = yes</parameter></member>
	<member><indexterm>printer admin</indexterm><parameter>printer admin = kurt</parameter></member>
	<member><indexterm>hosts deny</indexterm><parameter>hosts deny = 0.0.0.0</parameter></member>
	<member><indexterm>hosts allow</indexterm><parameter>hosts allow = turbo_xp, 10.160.50.23, 10.160.51.60</parameter></member>
	</simplelist>
	</example>

	<para>
	This special share is only for testing purposes. It does not write the print job to a file. It just logs the job parameters
	known to Samba into the <filename moreinfo="none">/tmp/smbprn.log</filename> file and deletes the job-file. Moreover, the
	<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTERADMIN" xlink:href="smb.conf.5.html#PRINTERADMIN">printer admin</link> of this share is <quote>kurt</quote> (not the <quote>@ntadmins</quote> group),
	guest access is not allowed, the share isn't published to the Network Neighborhood (so you need to know it is there), and it
	allows access from only three hosts. To prevent CUPS from kicking in and taking over the print jobs for that share, we need to set
	<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTING" xlink:href="smb.conf.5.html#PRINTING">printing = sysv</link> and <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTCAP" xlink:href="smb.conf.5.html#PRINTCAP">printcap = lpstat</link>.
	</para>

	</sect2>

</sect1>

<sect1>
	<title>Advanced Configuration</title>

	<para>
	Before we delve into all the configuration options, let us clarify a few points. <emphasis>Network printing
	needs to be organized and set up correctly</emphasis>. This frequently doesn't happen. Legacy systems or small
	business LAN environments often lack design and good housekeeping.
	</para>


	<sect2>
	<title>Central Spooling vs. <quote>Peer-to-Peer</quote> Printing</title>


	<para>
<indexterm significance="normal"><primary>spooling</primary></indexterm>
	<indexterm significance="normal"><primary>spooling</primary><secondary>central</secondary></indexterm>
	<indexterm significance="normal"><primary>spooling</primary><secondary>peer-to-peer</secondary></indexterm>
	Many small office or home networks, as well as badly organized larger environments, allow each client a direct
	access to available network printers. This is generally a bad idea. It often blocks one client's access to the
	printer when another client's job is printing. It might freeze the first client's application while it is
	waiting to get rid of the job. Also, there are frequent complaints about various jobs being printed with their
	pages mixed with each other. A better concept is the use of a print server: it routes all jobs through one
	central system, which responds immediately, takes jobs from multiple concurrent clients, and transfers them to
	the printer(s) in the correct order.
	</para>

	</sect2>

	<sect2>
	<title>Raw Print Serving: Vendor Drivers on Windows Clients</title>

	<para>
	<indexterm significance="normal"><primary>spooling-only</primary></indexterm>
	<indexterm significance="normal"><primary>raw printing</primary></indexterm>
	Most traditionally configured UNIX print servers acting on behalf of
	Samba's Windows clients represented a really simple setup. Their only
	task was to manage the <quote>raw</quote> spooling of all jobs handed to them by
	Samba. This approach meant that the Windows clients were expected to
	prepare the print job file that is ready to be sent to the printing
	device. In this case, a native (vendor-supplied) Windows printer driver needs to
	be installed on each and every client for the target device.
	</para>

	<para>
<indexterm significance="normal"><primary>render</primary></indexterm>
<indexterm significance="normal"><primary>vendor-provided drivers</primary></indexterm>
	It is possible to configure CUPS, Samba, and your Windows clients in the
	same traditional and simple way. When CUPS printers are configured
	for raw print-through mode operation, it is the responsibility of the
	Samba client to fully render the print job (file). The file must be
	sent in a format that is suitable for direct delivery to the
	printer. Clients need to run the vendor-provided drivers to do
	this. In this case, CUPS will not do any print file format conversion
	work.
	</para>

	<para>
	The easiest printing configuration possible is raw print-through.
	This is achieved by installation of the printer as if it were physically
	attached to the Windows client. You then redirect output to a raw network
	print queue. This procedure may be followed to achieve this:
	</para>

	<procedure>
	<title>Configuration Steps for Raw CUPS Printing Support</title>

		<step performance="required"><para>
<indexterm significance="normal"><primary>/etc/cups/mime.types</primary></indexterm>
		Edit <filename moreinfo="none">/etc/cups/mime.types</filename> to uncomment the line
		near the end of the file that has:
<screen format="linespecific">
#application/octet-...
</screen>
		</para></step>

		<step performance="required"><para>
<indexterm significance="normal"><primary>/etc/cups/mime.convs</primary></indexterm>
		Do the same for the file <filename moreinfo="none">/etc/cups/mime.convs</filename>.
		</para></step>

		<step performance="required"><para>
		Add a raw printer using the Web interface. Point your browser at
		<constant>http://localhost:631</constant>. Enter Administration, and add
		the printer following the prompts. Do not install any drivers for it.
		Choose Raw. Choose queue name <constant>Raw Queue</constant>.
		</para></step>

		<step performance="required"><para>
		In the <filename moreinfo="none">smb.conf</filename> file <constant>[printers]</constant> section add
		<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="USECLIENTDRIVER" xlink:href="smb.conf.5.html#USECLIENTDRIVER">use client driver = Yes</link>,
		and in the <constant>[global]</constant> section add
		<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTING" xlink:href="smb.conf.5.html#PRINTING">printing = CUPS</link>, plus
		<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTCAP" xlink:href="smb.conf.5.html#PRINTCAP">printcap = CUPS</link>.
		</para></step>

		<step performance="required"><para>
		Install the printer as if it is a local printer, that is, Printing to <constant>LPT1:</constant>.
		</para></step>

		<step performance="required"><para>
		Edit the configuration under the <guimenu moreinfo="none">Detail</guimenu> tab and create a
		<constant>local port</constant> that points to the raw printer queue that
		you have configured above. Example: <constant>\\server\raw_q</constant>.
		Here, the name <constant>raw_q</constant> is the name you gave the print
		queue in the CUPS environment.
		</para></step>
	</procedure>

	</sect2>

	<sect2>
	<title>Installation of Windows Client Drivers</title>

	<para>
	The printer drivers on the Windows clients may be installed
	in two functionally different ways:
	</para>

	<itemizedlist>
	<listitem><para>Manually install the drivers locally on each client,
	one by one; this yields the old LanMan style
	printing and uses a <filename moreinfo="none">\\sambaserver\printershare</filename>
	type of connection.</para></listitem>


	<listitem><para>
	<indexterm significance="normal"><primary>point 'n' print</primary></indexterm>
			Deposit and prepare the drivers (for later download) on
			the print server (Samba); this enables the clients to use
	<quote>Point'n'Print</quote> to get drivers semi-automatically installed the
	first time they access the printer; with this method NT/200x/XP
	clients use the <emphasis>SPOOLSS/MS-RPC</emphasis>
	type printing calls.</para></listitem>
	</itemizedlist>

	<para>
	The second method is recommended for use over the first as it reduces the
	administrative efforts and prevents that different versions of the drivers
	are used accidentally.
	</para>
	</sect2>

	<sect2 id="cups-raw">
	<title>Explicitly Enable <quote>raw</quote> Printing for <emphasis>application/octet-stream</emphasis></title>


	<para>
	<indexterm significance="normal"><primary>application/octet-stream</primary></indexterm>
	<indexterm significance="normal"><primary>raw printing</primary></indexterm>
	<indexterm significance="normal"><primary>MIME</primary><secondary>raw</secondary></indexterm>
	If you use the first option (drivers are installed on the client
	side), there is one setting to take care of: CUPS needs to be told
	that it should allow <quote>raw</quote> printing of deliberate (binary) file
	formats. The CUPS files that need to be correctly set for raw mode
	printers to work are:
	</para>

	<itemizedlist>
		<listitem><para><filename moreinfo="none">/etc/cups/mime.types</filename></para></listitem>
		<listitem><para><filename moreinfo="none">/etc/cups/mime.convs</filename></para></listitem>
	</itemizedlist>

	<para>
	Both contain entries (at the end of the respective files) that must be uncommented to allow RAW mode
	operation.  In <filename moreinfo="none">/etc/cups/mime.types</filename>, make sure this line is present:
<programlisting format="linespecific">
application/octet-stream
</programlisting>
	<indexterm significance="normal"><primary>/etc/cups/mime.convs</primary></indexterm>
	<indexterm significance="normal"><primary>/etc/cups/mime.types</primary></indexterm>
	In <filename moreinfo="none">/etc/cups/mime.convs</filename>, have this line:
	<indexterm significance="normal"><primary>application/vnd.cups-raw</primary></indexterm>
<programlisting format="linespecific">
application/octet-stream   application/vnd.cups-raw   0   - 
</programlisting>
	If these two files are not set up correctly for raw Windows client
	printing, you may encounter the dreaded <computeroutput moreinfo="none">Unable to
	convert file 0</computeroutput> in your CUPS <filename moreinfo="none">error_log</filename> file. 
	</para>

	<note><para>
	Editing the <filename moreinfo="none">mime.convs</filename> and the <filename moreinfo="none">mime.types</filename> file does
	not <emphasis>enforce</emphasis> <quote>raw</quote> printing, it only <emphasis>allows</emphasis> it.
	</para></note>

	<formalpara><title>Background</title>

	<para>
	<indexterm significance="normal"><primary>application/octet-stream</primary></indexterm>
<indexterm significance="normal"><primary>MIME type</primary></indexterm>
	That CUPS is a more security-aware printing system than traditional ones does not by default allow a user to
	send deliberate (possibly binary) data to printing devices. This could be easily abused to launch a
	<quote>Denial of Service</quote> attack on your printer(s), causing at least the loss of a lot of paper and
	ink. <quote>Unknown</quote> data are tagged by CUPS as <parameter moreinfo="none">MIME type: application/octet-stream</parameter>
	and not allowed to go to the printer. By default, you can only send other (known) MIME types <quote>raw.</quote>
	Sending data <quote>raw</quote> means that CUPS does not try to convert them and passes them to the printer
	untouched.
	</para>
	</formalpara>

	<para>
	This is all you need to know to get the CUPS/Samba combo printing
	<quote>raw</quote> files prepared by Windows clients, which have vendor drivers
	locally installed. If you are not interested in background information about
	more advanced CUPS/Samba printing, simply skip the remaining sections
	of this chapter.
	</para>

	</sect2>

	<sect2>
	<title>Driver Upload Methods</title>

	<para>
	This section describes three familiar methods, plus one new one, by which
	printer drivers may be uploaded.
	</para>

	<para>
	<indexterm significance="normal"><primary>point'n'print</primary></indexterm>
	If you want to use the MS-RPC-type printing, you must upload the
	drivers onto the Samba server first (<parameter>[print$]</parameter>
	share). For a discussion on how to deposit printer drivers on the
	Samba host (so the Windows clients can download and use them via
	<quote>Point'n'Print</quote>), please refer to the <link linkend="classicalprinting">Classical Printing
	chapter</link> of this book. There you will find a description or reference to
	three methods of preparing the client drivers on the Samba server:
	</para>

	<itemizedlist>
		<listitem><para>
		<indexterm significance="normal"><primary>add printer wizard</primary></indexterm>
		The GUI, <quote>Add Printer Wizard</quote> <emphasis>upload-from-a-Windows-client</emphasis> method.
		</para></listitem>

		<listitem><para>
		The command line, <quote>smbclient/rpcclient</quote> upload-from-a-UNIX-workstation method.
		</para></listitem>

		<listitem><para>
		<indexterm significance="normal"><primary>imprints</primary></indexterm>
		The Imprints tool set method.
		</para></listitem>
	</itemizedlist>

	<para> 
<indexterm significance="normal"><primary>cupsaddsmb</primary></indexterm>
	These three methods apply to CUPS all the same. The <literal>cupsaddsmb</literal> utility is a new and more
	convenient way to load the Windows drivers into Samba and is provided if you use CUPS.
	</para>

	<para>
	<literal>cupsaddsmb</literal> is discussed in much detail later in this chapter. But we first
	explore the CUPS filtering system and compare the Windows and UNIX printing architectures.
	</para>

	</sect2>

</sect1>

<sect1>
	<title>Advanced Intelligent Printing with PostScript Driver Download</title>

	<para>
	<indexterm significance="normal"><primary>PostScript</primary><seealso>Ghostscript</seealso></indexterm>
	We now know how to set up a <quote>dump</quote> print server, that is, a server that spools
	print jobs <quote>raw</quote>, leaving the print data untouched.
	</para>

	<para>
	You might need to set up CUPS in a smarter way. The reasons could be manifold:
	</para>

<indexterm significance="normal"><primary>print statistics</primary></indexterm>
<indexterm significance="normal"><primary>average print run</primary></indexterm>
<indexterm significance="normal"><primary>print quota</primary></indexterm>
	<itemizedlist>
	<listitem><para>Maybe your boss wants to get monthly statistics: Which
	printer did how many pages? What was the average data size of a job?
	What was the average print run per day? What are the typical hourly
	peaks in printing? Which department prints how much?</para></listitem>

	<listitem><para>Maybe you are asked to set up a print quota system:
	Users should not be able to print more jobs once they have surpassed
	a given limit per period.</para></listitem>

	<listitem><para>Maybe your previous network printing setup is a mess
	and must be re-organized from a clean beginning.</para></listitem>

	<listitem><para>Maybe you are experiencing too many <quote>blue screens</quote>
	originating from poorly debugged printer drivers running in NT <quote>kernel mode</quote>?</para></listitem>
	</itemizedlist>

	<para>
	These goals cannot be achieved by a raw print server. To build a
	server meeting these requirements, you'll first need to learn
	how CUPS works and how you can enable its features.
	</para>

	<para>
	What follows is the comparison of some fundamental concepts for
	Windows and UNIX printing, then a description of the
	CUPS filtering system, how it works, and how you can tweak it.
	</para>

	<sect2 id="gdipost">
	<title>GDI on Windows, PostScript on UNIX</title>

	<para>
	<indexterm significance="normal"><primary>GDI</primary></indexterm>
	<indexterm significance="normal"><primary>PostScript</primary></indexterm>
	Network printing is one of the most complicated and error-prone
	day-to-day tasks any user or administrator may encounter. This is
	true for all OS platforms, and there are reasons it is so.
	</para>


	<para>
	<indexterm significance="normal"><primary>PCL</primary></indexterm>
	<indexterm significance="normal"><primary>PDL</primary></indexterm>
<indexterm significance="normal"><primary>PostScript</primary></indexterm>
<indexterm significance="normal"><primary>Adobe</primary></indexterm>
<indexterm significance="normal"><primary>page description languages</primary><see>PDL</see></indexterm>
	You can't expect to throw just any file format at a printer and have it get printed. A file format conversion
	must take place. The problem is that there is no common standard for print file formats across all
	manufacturers and printer types. While PostScript (trademark held by Adobe) and, to an extent, PCL (trademark
	held by Hewlett-Packard) have developed into semi-official <quote>standards</quote> by being the most widely
	used page description languages (PDLs), there are still many manufacturers who <quote>roll their own</quote>
	(their reasons may be unacceptable license fees for using printer-embedded PostScript interpreters, and so on).
	</para>

	</sect2>

	<sect2>
	<title>Windows Drivers, GDI, and EMF</title>

	<para>
	<indexterm significance="normal"><primary>GDI</primary></indexterm>
	<indexterm significance="normal"><primary>EMF</primary></indexterm>
	<indexterm significance="normal"><primary>WYSIWYG</primary></indexterm>
<indexterm significance="normal"><primary>Enhanced MetaFile</primary><see>EMF</see></indexterm>
	In Windows OS, the format conversion job is done by the printer drivers. On MS Windows OS platforms all
	application programmers have at their disposal a built-in API, the graphical device interface (GDI), as part
	and parcel of the OS itself to base themselves on. This GDI core is used as one common unified ground for all
	Windows programs to draw pictures, fonts, and documents <emphasis>on screen</emphasis> as well as <emphasis>on
	paper</emphasis> (print). Therefore, printer driver developers can standardize on a well-defined GDI output
	for their own driver input. Achieving WYSIWYG (What You See Is What You Get) is relatively easy, because the
	on-screen graphic primitives, as well as the on-paper drawn objects, come from one common source. This source,
	the GDI, often produces a file format called Enhanced MetaFile (EMF). The EMF is processed by the printer
	driver and converted to the printer-specific file format.
	</para>

	<note><para>
	<indexterm significance="normal"><primary>PDF</primary></indexterm>
<indexterm significance="normal"><primary>Xprint</primary></indexterm>
<indexterm significance="normal"><primary>core graphic engine</primary></indexterm>
	To the GDI foundation in MS Windows, Apple has chosen to put paper and screen output on a common foundation
	for its (BSD-UNIX-based, did you know?) Mac OS X and Darwin operating <indexterm significance="normal"><primary>X Window
	System</primary></indexterm> <indexterm significance="normal"><primary>PostScript</primary></indexterm>
	<indexterm significance="normal"><primary>PCL</primary></indexterm> <indexterm significance="normal"><primary>Xprint</primary></indexterm> systems.
	Apple's <emphasis>core graphic engine</emphasis> uses a <emphasis>PDF</emphasis> derivative for all display work.
	</para></note>

	<para>
	The example in <link linkend="1small">Windows Printing to a Local Printer</link> illustrates local Windows
	printing.
	</para>

	<figure id="1small" float="0">
		<title>Windows Printing to a Local Printer.</title>
		<mediaobject><imageobject role="html"><imagedata fileref="images/1small.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/1small.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/1small"/></imageobject></mediaobject>
	</figure>

	</sect2>

	<sect2>
	<title>UNIX Printfile Conversion and GUI Basics</title>

	<para>
	<indexterm significance="normal"><primary>X Window System</primary></indexterm>
	<indexterm significance="normal"><primary>PostScript</primary></indexterm>
	<indexterm significance="normal"><primary>PCL</primary></indexterm>
	<indexterm significance="normal"><primary>Xprint</primary></indexterm>
	In UNIX and Linux, there is no comparable layer built into the OS kernel(s) or the X (screen display) server.
	Every application is responsible for itself to create its print output. Fortunately, most use PostScript and
	that at least gives some common ground. Unfortunately, there are many different levels of quality for this
	PostScript. And worse, there is a huge difference (and no common root) in the way the same document is
	displayed on screen and how it is presented on paper. WYSIWYG is more difficult to achieve. This goes back to
	the time, decades ago, when the predecessors of X.org, designing the UNIX foundations and protocols for
	graphical user interfaces, refused to take responsibility for <quote>paper output</quote>, as some had
	demanded at the time, and restricted itself to <quote>on-screen only.</quote> (For some years now, the
	<quote>Xprint</quote> project has been under development, attempting to build printing support into the X
	framework, including a PostScript and a PCL driver, but it is not yet ready for prime time.) You can see this
	unfavorable inheritance up to the present day by looking into the various <quote>font</quote> directories on
	your system; there are separate ones for fonts used for X display and fonts to be used on paper.
	</para>

	<formalpara>
	<title>Background</title>

	<para>
	<indexterm significance="normal"><primary>PostScript</primary></indexterm>
<indexterm significance="normal"><primary>color</primary></indexterm>
<indexterm significance="normal"><primary>linewidth</primary></indexterm>
<indexterm significance="normal"><primary>scale</primary></indexterm>
<indexterm significance="normal"><primary>distort</primary></indexterm>
<indexterm significance="normal"><primary>rotate</primary></indexterm>
<indexterm significance="normal"><primary>shift</primary></indexterm>
<indexterm significance="normal"><primary>raster images</primary></indexterm>
<indexterm significance="normal"><primary>display PostScript</primary></indexterm>
<indexterm significance="normal"><primary>graphical objects</primary></indexterm>
	The PostScript programming language is an <quote>invention</quote> by Adobe, but its specifications have been
	published extensively. Its strength lies in its powerful abilities to describe graphical objects (fonts,
	shapes, patterns, lines, curves, and dots), their attributes (color, linewidth), and the way to manipulate
	(scale, distort, rotate, shift) them. Because of its open specification, anybody with the skill can start
	writing his or her own implementation of a PostScript interpreter and use it to display PostScript files on
	screen or on paper. Most graphical output devices are based on the concept of <quote>raster images</quote> or
	<quote>pixels</quote> (one notable exception is pen plotters). Of course, you can look at a PostScript file in
	its textual form and you will be reading its PostScript code, the language instructions that need to be
	interpreted by a rasterizer. Rasterizers produce pixel images, which may be displayed on screen by a viewer
	program or on paper by a printer.
	</para>
	</formalpara>
	</sect2>

	<sect2 id="post-and-ghost">
	<title>PostScript and Ghostscript</title>

	<para>
	<indexterm significance="normal"><primary>PostScript</primary></indexterm>
	<indexterm significance="normal"><primary>GhostScript</primary><seealso>PostScript</seealso></indexterm>
	<indexterm significance="normal"><primary>PostScript</primary><secondary>RIP</secondary></indexterm>
<indexterm significance="normal"><primary>PostScript interpreter</primary></indexterm>
<indexterm significance="normal"><primary>raster image processor</primary><see>RIP</see></indexterm>
	So UNIX is lacking a common ground for printing on paper and displaying on screen. Despite this unfavorable
	legacy for UNIX, basic printing is fairly easy if you have PostScript printers at your disposal. The reason is
	that these devices have a built-in PostScript language <quote>interpreter,</quote> also called a raster image
	processor (RIP), (which makes them more expensive than other types of printers; throw PostScript toward them,
	and they will spit out your printed pages. The RIP does all the hard work of converting the PostScript drawing
	commands into a bitmap picture as you see it on paper, in a resolution as done by your printer. This is no
	different than PostScript printing a file from a Windows origin.
	</para>

	<note><para>
	<indexterm significance="normal"><primary>PPD</primary></indexterm>
<indexterm significance="normal"><primary>PPD-aware</primary></indexterm>
<indexterm significance="normal"><primary>PostScript Printer Description</primary><see>PPD</see></indexterm>
	Traditional UNIX programs and printing systems <?latex --- ?> while using PostScript <?latex --- ?> are largely not
	PPD-aware. PPDs are <quote>PostScript Printer Description</quote> files. They enable you to specify and
	control all options a printer supports: duplexing, stapling, and punching. Therefore, UNIX users for a long
	time couldn't choose many of the supported device and job options, unlike Windows or Apple users. But now
	there is CUPS. as illustrated in <link linkend="2small">Printing to a PostScript Printer</link>.
	</para>
	</note>

	<figure id="2small" float="0">
		<title>Printing to a PostScript Printer.</title>
		<mediaobject><imageobject role="html"><imagedata fileref="images/2small.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/2small.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/2small"/></imageobject></mediaobject>
	</figure>

	<para>
	<indexterm significance="normal"><primary>PDL</primary></indexterm>
	However, there are other types of printers out there. These do not know how to print PostScript. They use
	their own PDL, often proprietary. To print to them is much more demanding. Since your UNIX applications mostly
	produce PostScript, and since these devices do not understand PostScript, you need to convert the print files
	to a format suitable for your printer on the host before you can send it away.
	</para>

	</sect2>

	<sect2>
	<title>Ghostscript: The Software RIP for Non-PostScript Printers</title>

	<para>
	<indexterm significance="normal"><primary>GhostScript</primary></indexterm>
	Here is where Ghostscript kicks in. Ghostscript is the traditional (and quite powerful) PostScript interpreter
	used on UNIX platforms. It is a RIP in software, capable of doing a <emphasis>lot</emphasis> of file format
	conversions for a very broad spectrum of hardware devices as well as software file formats.  Ghostscript
	technology and drivers are what enable PostScript printing to non-PostScript hardware. This is shown in
	<link linkend="3small">Ghostscript as a RIP for Non-PostScript Printers</link>.
	</para>

	<figure id="3small" float="0">
		<title>Ghostscript as a RIP for Non-PostScript Printers.</title>
		<mediaobject><imageobject role="html"><imagedata fileref="images/3small.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/3small.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/3small"/></imageobject></mediaobject>
	</figure>

	<tip><para>
<indexterm significance="normal"><primary>PNG</primary></indexterm>
<indexterm significance="normal"><primary>AFPL</primary></indexterm>
<indexterm significance="normal"><primary>ESP</primary></indexterm>
	Use the <quote>gs -h</quote> command to check for all built-in <quote>devices</quote> on your Ghostscript
	version. If you specify a parameter of <parameter moreinfo="none">-sDEVICE=png256</parameter> on your Ghostscript command
	line, you are asking Ghostscript to convert the input into a PNG file. Naming a <quote>device</quote> on the
	command line is the most important single parameter to tell Ghostscript exactly how it should render the
	input. New Ghostscript versions are released at fairly regular intervals, now by artofcode LLC. They are
	initially put under the <quote>AFPL</quote> license, but re-released under the GNU GPL as soon as the next
	AFPL version appears. GNU Ghostscript is probably the version installed on most Samba systems. But it has some
	deficiencies.  <indexterm significance="normal"><primary>Ghostscript</primary><secondary>ESP</secondary><see>ESP
	GhostScript</see></indexterm> Therefore, ESP Ghostscript was developed as an enhancement over GNU Ghostscript,
	with lots of bug-fixes, additional devices, and improvements. It is jointly maintained by developers from
	CUPS, Gutenprint, MandrakeSoft, SuSE, Red Hat, and Debian. It includes the <quote>cups</quote> device
	(essential to print to non-PS printers from CUPS).
	</para></tip>

	</sect2>

	<sect2>
	<title>PostScript Printer Description (PPD) Specification</title>

	<para>
	<indexterm significance="normal"><primary>PPD</primary></indexterm>
<indexterm significance="normal"><primary>PDL</primary></indexterm>
<indexterm significance="normal"><primary>PostScript</primary></indexterm>
	While PostScript in essence is a PDL to represent the page layout in a device-independent way, real-world
	print jobs are always ending up being output on hardware with device-specific features. To take care of all
	the differences in hardware and to allow for innovations, Adobe has specified a syntax and file format for
	PostScript Printer Description (PPD) files. Every PostScript printer ships with one of these files.
	</para>

	<para>
	PPDs contain all the information about general and special features of the
	given printer model: Which different resolutions can it handle? Does
	it have a duplexing unit? How many paper trays are there? What media
	types and sizes does it take? For each item, it also names the special
	command string to be sent to the printer (mostly inside the PostScript
	file) in order to enable it.
	</para>

	<para>
	Information from these PPDs is meant to be taken into account by the
	printer drivers. Therefore, installed as part of the Windows
	PostScript driver for a given printer is the printer's PPD. Where it
	makes sense, the PPD features are presented in the drivers' UI dialogs
	to display to the user a choice of print options. In the end, the
	user selections are somehow written (in the form of special
	PostScript, PJL, JCL, or vendor-dependent commands) into the PostScript
	file created by the driver.
	</para>

	<warning><para>
	<indexterm significance="normal"><primary>PDF</primary></indexterm>
<indexterm significance="normal"><primary>PDF distilling</primary></indexterm>
	A PostScript file that was created to contain device-specific commands
	for achieving a certain print job output (e.g., duplexed, stapled, and
	punched) on a specific target machine may not print as expected, or
	may not be printable at all on other models; it also may not be fit
	for further processing by software (e.g., by a PDF distilling program).
	</para></warning>
	</sect2>

	<sect2>
	<title>Using Windows-Formatted Vendor PPDs</title>

	<para>
<indexterm significance="normal"><primary>CUPS</primary></indexterm>
<indexterm significance="normal"><primary>PPDs</primary></indexterm>
<indexterm significance="normal"><primary>PostScript</primary></indexterm>
	CUPS can handle all spec-compliant PPDs as supplied by the manufacturers for their PostScript models. Even if
	a vendor does not mention our favorite OS in his or her manuals and brochures, you can safely trust this:
	<emphasis>If you get the Windows NT version of the PPD, you can use it unchanged in CUPS</emphasis> and thus
	access the full power of your printer just like a Windows NT user could!
	</para>

	<tip><para>
	To check the spec compliance of any PPD online, go to <ulink noescape="1" url="http://www.cups.org/testppd.php">http://www.cups.org/testppd.php</ulink> and upload your PPD. You will
	see the results displayed immediately. CUPS in all versions after 1.1.19 has a much stricter internal PPD
	parsing and checking code enabled; in case of printing trouble, this online resource should be one of your
	first pit stops.
	</para></tip>

	<warning><para>
	<indexterm significance="normal"><primary>foomatic</primary></indexterm>
	<indexterm significance="normal"><primary>cupsomatic</primary></indexterm>
	For real PostScript printers, <emphasis>do not</emphasis> use the <emphasis>Foomatic</emphasis> or
	<emphasis>cupsomatic</emphasis> PPDs from Linuxprinting.org. With these devices, the original vendor-provided
	PPDs are always the first choice.
	</para></warning>

	<tip><para>
<indexterm significance="normal"><primary>W32X86/2</primary></indexterm>
	If you are looking for an original vendor-provided PPD of a specific device, and you know that an NT4 box (or
	any other Windows box) on your LAN has the PostScript driver installed, just use <literal>smbclient
	//NT4-box/print\$ -U username</literal> to access the Windows directory where all printer driver files are
	stored. First look in the <filename moreinfo="none">W32X86/2</filename> subdirectory for the PPD you are seeking.
	</para></tip>
	</sect2>

	<sect2>
	<title>CUPS Also Uses PPDs for Non-PostScript Printers</title>

	<para>
<indexterm significance="normal"><primary>non-PostScript</primary></indexterm>
<indexterm significance="normal"><primary>PPD</primary></indexterm>
<indexterm significance="normal"><primary>CUPS filtering</primary></indexterm>
	CUPS also uses specially crafted PPDs to handle non-PostScript printers. These PPDs are usually not available
	from the vendors (and no, you can't just take the PPD of a PostScript printer with the same model name and
	hope it works for the non-PostScript version too). To understand how these PPDs work for non-PS printers, we
	first need to dive deeply into the CUPS filtering and file format conversion architecture. Stay tuned.
	</para>

	</sect2>

</sect1>

<sect1>
<title>The CUPS Filtering Architecture</title>

<para>
<indexterm significance="normal"><primary>CUPS filtering</primary></indexterm>
<indexterm significance="normal"><primary>Ghostscript</primary></indexterm>
<indexterm significance="normal"><primary>MIME type</primary></indexterm>
<indexterm significance="normal"><primary>MIME recognition</primary></indexterm>
<indexterm significance="normal"><primary>MIME conversion rules</primary></indexterm>
The core of the CUPS filtering system is based on Ghostscript. In addition to Ghostscript, CUPS uses some
other filters of its own. You (or your OS vendor) may have plugged in even more filters. CUPS handles all data
file formats under the label of various MIME types. Every incoming print file is subjected to an initial
autotyping. The autotyping determines its given MIME type. A given MIME type implies zero or more possible
filtering chains relevant to the selected target printer. This section discusses how MIME types recognition
and conversion rules interact. They are used by CUPS to automatically set up a working filtering chain for any
given input data format.
</para>

<para>
If CUPS rasterizes a PostScript file natively to a bitmap, this is done in two stages:
</para>

<itemizedlist>
	<listitem><para>
<indexterm significance="normal"><primary>generic raster format</primary></indexterm>
<indexterm significance="normal"><primary>CUPS raster</primary></indexterm>
	The first stage uses a Ghostscript device named <quote>cups</quote>
	(this is since version 1.1.15) and produces a generic raster format
	called <quote>CUPS raster</quote>.
	</para></listitem>

	<listitem><para>
<indexterm significance="normal"><primary>raster driver</primary></indexterm>
	The second stage uses a <quote>raster driver</quote> that converts
	the generic CUPS raster to a device-specific raster.
	</para></listitem>
</itemizedlist>

<para>
<indexterm significance="normal"><primary>Ghostscript</primary></indexterm>
<indexterm significance="normal"><primary>GNU Ghostscript</primary></indexterm>
<indexterm significance="normal"><primary>ESP Ghostscript</primary></indexterm>
Make sure your Ghostscript version has the <quote>cups</quote> device compiled in (check with <literal>gs -h |
grep cups</literal>). Otherwise you may encounter the dreaded <computeroutput moreinfo="none">Unable to convert file
0</computeroutput> in your CUPS error_log file. To have <quote>cups</quote> as a device in your Ghostscript,
you either need to patch GNU Ghostscript and recompile or use
<indexterm significance="normal"><primary>ESP</primary><secondary>Ghostscript</secondary></indexterm><ulink url="http://www.cups.org/ghostscript.php">ESP Ghostscript</ulink>. The superior alternative is ESP
Ghostscript. It supports not just CUPS, but 300 other devices (while GNU Ghostscript supports only about 180).
Because of this broad output device support, ESP Ghostscript is the first choice for non-CUPS spoolers, too.
It is now recommended by Linuxprinting.org for all spoolers.
</para>

<para>
<indexterm significance="normal"><primary>cupsomatic</primary></indexterm>
<indexterm significance="normal"><primary>foomatic</primary></indexterm>
<indexterm significance="normal"><primary>foomatic-rip</primary></indexterm>
<indexterm significance="normal"><primary>ESP Ghostscript</primary></indexterm>
CUPS printers may be set up to use external rendering paths. One of the most common is provided by the
Foomatic/cupsomatic concept from <ulink url="http://www.linuxprinting.org/">Linuxprinting.org</ulink>. This
uses the classical Ghostscript approach, doing everything in one step.  It does not use the
<quote>cups</quote> device, but one of the many others. However, even for Foomatic/cupsomatic usage, best
results and <indexterm significance="normal"><primary>ESP</primary><secondary>Ghostscript</secondary></indexterm> broadest printer
model support is provided by ESP Ghostscript (more about Foomatic/cupsomatic, particularly the new version
called now <emphasis>foomatic-rip</emphasis>, follows).
</para>

	<sect2>
	<title>MIME Types and CUPS Filters</title>


	<para>
	<indexterm significance="normal"><primary>MIME</primary><secondary>filters</secondary></indexterm>
	<indexterm significance="normal"><primary>MIME</primary></indexterm>
<indexterm significance="normal"><primary>mime.types</primary></indexterm>
<indexterm significance="normal"><primary>application/pdf</primary></indexterm>
<indexterm significance="normal"><primary>autotyping</primary></indexterm>
	CUPS reads the file <filename moreinfo="none">/etc/cups/mime.types</filename> (and all other files carrying a
	<filename moreinfo="none">*.types</filename> suffix in the same directory) upon startup. These files contain the MIME type
	recognition rules that are applied when CUPS runs its autotyping routines. The rule syntax is explained in the
	man page for <filename moreinfo="none">mime.types</filename> and in the comments section of the
	<filename moreinfo="none">mime.types</filename> file itself. A simple rule reads like this:
	<indexterm significance="normal"><primary>application/pdf</primary></indexterm>
<programlisting format="linespecific">
application/pdf         pdf string(0,%PDF)
</programlisting>
<indexterm significance="normal"><primary>%PDF</primary></indexterm>
<indexterm significance="normal"><primary>.pdf</primary></indexterm>
	This means if a filename has a <filename moreinfo="none">.pdf</filename> suffix or if the magic string
	<emphasis>%PDF</emphasis> is right at the beginning of the file itself (offset 0 from the start), then it is a
	PDF file (<parameter moreinfo="none">application/pdf</parameter>).  Another rule is this:
<programlisting format="linespecific">
application/postscript  ai eps ps string(0,%!) string(0,&lt;04&gt;%!)
</programlisting>
<indexterm significance="normal"><primary>suffixes</primary></indexterm>
<indexterm significance="normal"><primary>.ai</primary></indexterm>
<indexterm significance="normal"><primary>.eps</primary></indexterm>
<indexterm significance="normal"><primary>.ps</primary></indexterm>
<indexterm significance="normal"><primary>generic PostScript</primary></indexterm>
<indexterm significance="normal"><primary>application/postscript</primary></indexterm>
	If the filename has one of the suffixes <filename moreinfo="none">.ai</filename>, <filename moreinfo="none">.eps</filename>,
	<filename moreinfo="none">.ps</filename>, or if the file itself starts with one of the strings <emphasis>%!</emphasis> or
	<emphasis>&lt;04&gt;%!</emphasis>, it is a generic PostScript file
	(<parameter moreinfo="none">application/postscript</parameter>).
	</para>

	<warning><para>
<indexterm significance="normal"><primary>/etc/cups/</primary></indexterm>
	Don't confuse the other mime.types files your system might be using
	with the one in the <filename moreinfo="none">/etc/cups/</filename> directory.
	</para></warning>

	<note><para>
<indexterm significance="normal"><primary>application/postscript</primary></indexterm>
<indexterm significance="normal"><primary>PostScript</primary></indexterm>
<indexterm significance="normal"><primary>filter</primary></indexterm>
<indexterm significance="normal"><primary>PPD</primary></indexterm>
<indexterm significance="normal"><primary>transformation</primary></indexterm>
	There is an important difference between two similar MIME types in CUPS: one is
	<parameter moreinfo="none">application/postscript</parameter>, the other is
	<parameter moreinfo="none">application/vnd.cups-postscript</parameter>. While <parameter moreinfo="none">application/postscript</parameter> is
	meant to be device-independent, job options for the file are still outside the PS file content, embedded in
	command line or environment variables by CUPS, <parameter moreinfo="none">application/vnd.cups-postscript</parameter> may have
	the job options inserted into the PostScript data itself (where applicable). The transformation of the generic
	PostScript (<parameter moreinfo="none">application/postscript</parameter>) to the device-specific version
	(<parameter moreinfo="none">application/vnd.cups-postscript</parameter>) is the responsibility of the CUPS
	<parameter moreinfo="none">pstops</parameter> filter. pstops uses information contained in the PPD to do the transformation.
	</para></note>

	<para>
<indexterm significance="normal"><primary>ASCII</primary></indexterm>
<indexterm significance="normal"><primary>HP-GL</primary></indexterm>
<indexterm significance="normal"><primary>PDF</primary></indexterm>
<indexterm significance="normal"><primary>PostScript</primary></indexterm>
<indexterm significance="normal"><primary>DVI</primary></indexterm>
<indexterm significance="normal"><primary>GIF</primary></indexterm>
<indexterm significance="normal"><primary>PNG</primary></indexterm>
<indexterm significance="normal"><primary>TIFF</primary></indexterm>
<indexterm significance="normal"><primary>JPEG</primary></indexterm>
<indexterm significance="normal"><primary>Photo-CD</primary></indexterm>
<indexterm significance="normal"><primary>SUN-Raster</primary></indexterm>
<indexterm significance="normal"><primary>PNM</primary></indexterm>
<indexterm significance="normal"><primary>PBM</primary></indexterm>
<indexterm significance="normal"><primary>SGI-RGB</primary></indexterm>
<indexterm significance="normal"><primary>MIME</primary></indexterm>
<indexterm significance="normal"><primary>filters</primary></indexterm>
	CUPS can handle ASCII text, HP-GL, PDF, PostScript, DVI, and
	many image formats (GIF, PNG, TIFF, JPEG, Photo-CD, SUN-Raster,
	PNM, PBM, SGI-RGB, and more) and their associated MIME types
	with its filters.
	</para>

	</sect2>

	<sect2>
	<title>MIME Type Conversion Rules</title>


	<para>
	<indexterm significance="normal"><primary>MIME</primary></indexterm>
	<indexterm significance="normal"><primary>application/pdf</primary></indexterm>
<indexterm significance="normal"><primary>/etc/cups/mime.convs</primary></indexterm>
<indexterm significance="normal"><primary>application/pdf</primary></indexterm>
<indexterm significance="normal"><primary>application/postscript</primary></indexterm>
	CUPS reads the file <filename moreinfo="none">/etc/cups/mime.convs</filename>
	(and all other files named with a <filename moreinfo="none">*.convs</filename>
	suffix in the same directory) upon startup. These files contain
	lines naming an input MIME type, an output MIME type, a format
	conversion filter that can produce the output from the input type,
	and virtual costs associated with this conversion. One example line
	reads like this:
<programlisting format="linespecific">
application/pdf         application/postscript   33   pdftops
</programlisting>
<indexterm significance="normal"><primary>pdftops</primary></indexterm>
	This means that the <parameter moreinfo="none">pdftops</parameter> filter will take
	<parameter moreinfo="none">application/pdf</parameter> as input and produce
	<parameter moreinfo="none">application/postscript</parameter> as output; the virtual
	cost of this operation is 33 CUPS-$. The next filter is more
	expensive, costing 66 CUPS-$:
	<indexterm significance="normal"><primary>pdf</primary></indexterm>
<programlisting format="linespecific">
application/vnd.hp-HPGL application/postscript   66   hpgltops
</programlisting>
<indexterm significance="normal"><primary>hpgltops</primary></indexterm>
	This is the <parameter moreinfo="none">hpgltops</parameter>, which processes HP-GL
	plotter files to PostScript.
	<indexterm significance="normal"><primary>application/octet-stream</primary></indexterm>
<programlisting format="linespecific">
application/octet-stream
</programlisting>
	Here are two more examples: 
	<indexterm significance="normal"><primary>text/plain</primary></indexterm>
<indexterm significance="normal"><primary>application/x-shell</primary></indexterm>
<indexterm significance="normal"><primary>text/plain</primary></indexterm>
<indexterm significance="normal"><primary>texttops</primary></indexterm>
<programlisting format="linespecific">
application/x-shell     application/postscript   33    texttops
text/plain              application/postscript   33    texttops
</programlisting>
<indexterm significance="normal"><primary>application/x-shell</primary></indexterm>
	The last two examples name the <parameter moreinfo="none">texttops</parameter> filter to work on
	<parameter moreinfo="none">text/plain</parameter> as well as on <parameter moreinfo="none">application/x-shell</parameter>. (Hint: This
	differentiation is needed for the syntax highlighting feature of <parameter moreinfo="none">texttops</parameter>).
	</para>
	</sect2>

	<sect2>
	<title>Filtering  Overview</title>

	<para>
	<indexterm significance="normal"><primary>MIME</primary></indexterm>
	There are many more combinations named in <filename moreinfo="none">mime.convs</filename>. However, you are not limited to use
	the ones predefined there. You can plug in any filter you like to the CUPS framework. It must meet, or must be
	made to meet, some minimal requirements. If you find (or write) a cool conversion filter of some kind, make
	sure it complies with what CUPS needs and put in the right lines in <filename moreinfo="none">mime.types</filename> and
	<filename moreinfo="none">mime.convs</filename>; then it will work seamlessly inside CUPS.
	</para>

	<sect3>
	<title>Filter Requirements</title>

	<para>
	The <quote>CUPS requirements</quote> for filters are simple. Take filenames or <filename moreinfo="none">stdin</filename> as
	input and write to <filename moreinfo="none">stdout</filename>. They should take these arguments:
	</para>

	<variablelist>
		<varlistentry><term>printer</term>
			<listitem><para>
			The name of the printer queue (normally this is the name of the filter being run).
			</para></listitem>
		</varlistentry>

		<varlistentry><term>job</term>
			<listitem><para>
			The numeric job ID for the job being printed.
			</para></listitem>
		</varlistentry>

		<varlistentry><term>user</term>
			<listitem><para>
			The string from the originating-user-name attribute.
			</para></listitem>
		</varlistentry>

		<varlistentry><term>title</term>
			<listitem><para>
			The string from the job-name attribute.
			</para></listitem>
		</varlistentry>

		<varlistentry><term>copies</term>
			<listitem><para>
			The numeric value from the number-copies attribute.
			</para></listitem>
		</varlistentry>

		<varlistentry><term>options</term>
			<listitem><para>
			The job options.
			</para></listitem>
		</varlistentry>

		<varlistentry><term>filename</term>
			<listitem><para>
			(optionally) The print request file (if missing, filters expect data
			fed through <filename moreinfo="none">stdin</filename>). In most cases, it is easy to
			write a simple wrapper script around existing filters to make them work with CUPS.
			</para></listitem>
		</varlistentry>
	</variablelist>

	</sect3>

	</sect2>

	<sect2>
	<title>Prefilters</title>

	<para>
	<indexterm significance="normal"><primary>PostScript</primary></indexterm>
<indexterm significance="normal"><primary>non-PostScript printers</primary></indexterm>
<indexterm significance="normal"><primary>raster</primary></indexterm>
	As previously stated, PostScript is the central file format to any UNIX-based
	printing system. From PostScript, CUPS generates raster data to feed
	non-PostScript printers.
	</para>

	<para>
<indexterm significance="normal"><primary>prefilters</primary></indexterm>
<indexterm significance="normal"><primary>PostScript</primary></indexterm>
<indexterm significance="normal"><primary>ASCII text</primary></indexterm>
<indexterm significance="normal"><primary>PDF</primary></indexterm>
<indexterm significance="normal"><primary>DVI</primary></indexterm>
<indexterm significance="normal"><primary>HP-GL.</primary></indexterm>
<indexterm significance="normal"><primary>MIME type</primary></indexterm>
<indexterm significance="normal"><primary>application/postscript</primary></indexterm>
<indexterm significance="normal"><primary>pstops</primary></indexterm>
<indexterm significance="normal"><primary>application/vnd.cups-postscript</primary></indexterm>
	But what happens if you send one of the supported non-PS formats to print? Then CUPS runs
	<quote>prefilters</quote> on these input formats to generate PostScript first. There are prefilters to create
	PostScript from ASCII text, PDF, DVI, or HP-GL. The outcome of these filters is always of MIME type
	<parameter moreinfo="none">application/postscript</parameter> (meaning that any device-specific print options are not yet
	embedded into the PostScript by CUPS and that the next filter to be called is pstops). Another prefilter is
	running on all supported image formats, the <parameter moreinfo="none">imagetops</parameter> filter. Its outcome is always of
	MIME type <parameter moreinfo="none">application/vnd.cups-postscript</parameter> (not application/postscript), meaning it has
	the print options already embedded into the file. This is shown in <link linkend="4small">Prefiltering in
	CUPS to Form PostScript</link>.
	</para>

	<figure id="4small" float="0">
		<title>Prefiltering in CUPS to Form PostScript.</title>
		<mediaobject><imageobject role="html"><imagedata fileref="images/4small.png" scale="25" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/4small.png" scale="25" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/4small"/></imageobject></mediaobject>
	</figure>

	</sect2>

	<sect2>
	<title>pstops</title>

	<para>
<indexterm significance="normal"><primary>pstops</primary></indexterm>
<indexterm significance="normal"><primary>application/postscript</primary></indexterm>
<indexterm significance="normal"><primary>application/vnd.cups-postscript</primary></indexterm>
<indexterm significance="normal"><primary>output duplexing</primary></indexterm>
<indexterm significance="normal"><primary>stapling</primary></indexterm>
<indexterm significance="normal"><primary>punching</primary></indexterm>
<indexterm significance="normal"><primary>PostScript</primary></indexterm>
	<emphasis>pstops</emphasis> is a filter that is used to convert <parameter moreinfo="none">application/postscript</parameter> to
	<parameter moreinfo="none">application/vnd.cups-postscript</parameter>. As stated earlier, this filter inserts all
	device-specific print options (commands to the printer to ask for the duplexing of output, or stapling and
	punching it, and so on) into the PostScript file. An example is illustrated in <link linkend="5small">Adding Device-Specific Print Options</link>.
	</para>

	<figure id="5small" float="0">
		<title>Adding Device-Specific Print Options.</title>
		<mediaobject><imageobject role="html"><imagedata fileref="images/5small.png" scale="25" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/5small.png" scale="25" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/5small"/></imageobject></mediaobject>
	</figure>

	<para>
	This is not all. Other tasks performed by it are:
	</para>

	<itemizedlist>
		<listitem><para>
		Selecting the range of pages to be printed (e.g., you can choose to
		print only pages <quote>3, 6, 8-11, 16, and 19-21</quote>, or only odd-numbered
		pages).
		</para></listitem>

		<listitem><para>
		Putting two or more logical pages on one sheet of paper (the
		so-called <quote>number-up</quote> function).
		</para></listitem>

		<listitem><para>Counting the pages of the job to insert the accounting
		information into the <filename moreinfo="none">/var/log/cups/page_log</filename>.
		</para></listitem>
	</itemizedlist>
	</sect2>

	<sect2>
	<title>pstoraster</title>

	<para>
<indexterm significance="normal"><primary>pstoraster</primary></indexterm>
<indexterm significance="normal"><primary>rasterization</primary></indexterm>
<indexterm significance="normal"><primary>raster drivers</primary></indexterm>
	<parameter moreinfo="none">pstoraster</parameter> is at the core of the CUPS filtering system. It is responsible for the first
	stage of the rasterization process. Its input is of MIME type application/vnd.cups-postscript; its output is
	application/vnd.cups-raster. This output format is not yet meant to be printable. Its aim is to serve as a
	general-purpose input format for more specialized <emphasis>raster drivers</emphasis> that are able to
	generate device-specific printer data. This is shown in <link linkend="cups-raster">the PostScript to
	Intermediate Raster Format diagram</link>.
	</para>

	<figure id="cups-raster" float="0">
		<title>PostScript to Intermediate Raster Format.</title>
		<mediaobject><imageobject role="html"><imagedata fileref="images/6small.png" scale="25" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/6small.png" scale="25" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/6small"/></imageobject></mediaobject>
	</figure>

	<para>
<indexterm significance="normal"><primary>CUPS raster</primary></indexterm>
<indexterm significance="normal"><primary>generic raster</primary></indexterm>
<indexterm significance="normal"><primary>IANA</primary></indexterm>
<indexterm significance="normal"><primary>raster drivers</primary></indexterm>
	CUPS raster is a generic raster format with powerful features. It is able to include per-page information,
	color profiles, and more, to be used by the downstream raster drivers. Its MIME type is registered with IANA
	and its specification is, of course, completely open. It is designed to make it quite easy and inexpensive for
	manufacturers to develop Linux and UNIX raster drivers for their printer models should they choose to do so.
	CUPS always takes care of the first stage of rasterization so these vendors do not need to care about
	Ghostscript complications (in fact, there are currently more than one vendor financing the development of CUPS
	raster drivers). This is illustrated in <link linkend="cups-raster2">the CUPS-Raster Production Using
	Ghostscript illustration</link>.
	</para>

	<figure id="cups-raster2" float="0">
		<title>CUPS-Raster Production Using Ghostscript.</title>
		<mediaobject><imageobject role="html"><imagedata fileref="images/7small.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/7small.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/7small"/></imageobject></mediaobject>
	</figure>

	<para>
<indexterm significance="normal"><primary>pstoraster</primary></indexterm>
<indexterm significance="normal"><primary>GNU Ghostscript</primary></indexterm>
<indexterm significance="normal"><primary>AFPL Ghostscript</primary></indexterm>
<indexterm significance="normal"><primary>standalone filter</primary></indexterm>
	CUPS versions before version 1.1.15 shipped a binary (or source code) standalone filter, named
	<parameter moreinfo="none">pstoraster</parameter>. <parameter moreinfo="none">pstoraster</parameter>, which was derived from GNU Ghostscript
	5.50 and could be installed instead of and in addition to any GNU or AFPL Ghostscript package without
	conflicting.
	</para>

	<para>
	Since version 1.1.15, this feature has changed. The functions for this filter have been integrated back
	into Ghostscript (now based on GNU Ghostscript version 7.05). The <parameter moreinfo="none">pstoraster</parameter> filter is
	now a simple shell script calling <literal>gs</literal> with the <literal>-sDEVICE=cups</literal> parameter.
	If your Ghostscript fails when this command is executed: <literal>gs -h |grep cups</literal>, you might not 
	be able to print, update your Ghostscript.
	</para>
	</sect2>

	<sect2>
	<title>imagetops and imagetoraster</title>

	<para>
<indexterm significance="normal"><primary>prefilter</primary></indexterm>
<indexterm significance="normal"><primary>imagetoraster</primary></indexterm>
	In the section about prefilters, we mentioned the prefilter
	that generates PostScript from image formats. The <parameter moreinfo="none">imagetoraster</parameter>
	filter is used to convert directly from image to raster, without the
	intermediate PostScript stage. It is used more often than the previously
	mentioned prefilters. We summarize in a flowchart the image file
	filtering in <link linkend="small8">the Image Format to CUPS-Raster Format Conversion illustration</link>.
	</para>

	<figure id="small8" float="0">
		<title>Image Format to CUPS-Raster Format Conversion.</title>
		<mediaobject><imageobject role="html"><imagedata fileref="images/8small.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/8small.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/8small"/></imageobject></mediaobject>
	</figure>

	</sect2>

	<sect2>
	<title>rasterto [printers specific]</title>

	<para>
<indexterm significance="normal"><primary>rastertoalps</primary></indexterm>
<indexterm significance="normal"><primary>rastertobj</primary></indexterm>
<indexterm significance="normal"><primary>rastertoepson</primary></indexterm>
<indexterm significance="normal"><primary>rastertoescp</primary></indexterm>
<indexterm significance="normal"><primary>rastertopcl</primary></indexterm>
<indexterm significance="normal"><primary>rastertoturboprint</primary></indexterm>
<indexterm significance="normal"><primary>rastertoescp</primary></indexterm>
<indexterm significance="normal"><primary>rastertohp</primary></indexterm>
<indexterm significance="normal"><primary>rastertoprinter</primary></indexterm>
<indexterm significance="normal"><primary>rastertoprinter</primary></indexterm>
<indexterm significance="normal"><primary>Gutenprint</primary></indexterm>
	CUPS ships with quite a variety of raster drivers for processing CUPS raster. On my system, I find in
	/usr/lib/cups/filter/ the following: <parameter moreinfo="none">rastertoalps</parameter>, <parameter moreinfo="none">rastertobj</parameter>,
	<parameter moreinfo="none">rastertoepson</parameter>, <parameter moreinfo="none">rastertoescp</parameter>, <parameter moreinfo="none">rastertopcl</parameter>,
	<parameter moreinfo="none">rastertoturboprint</parameter>, <parameter moreinfo="none">rastertoapdk</parameter>,
	<parameter moreinfo="none">rastertodymo</parameter>, <parameter moreinfo="none">rastertoescp</parameter>, <parameter moreinfo="none">rastertohp</parameter>,
	and <parameter moreinfo="none">rastertoprinter</parameter>. Don't worry if you have fewer drivers than this; some of these are
	installed by commercial add-ons to CUPS (like <parameter moreinfo="none">rastertoturboprint</parameter>), and others (like
	<parameter moreinfo="none">rastertoprinter</parameter>) by third-party driver development projects (such as Gutenprint)
	wanting to cooperate as closely as possible with CUPS. See <link linkend="small9">the Raster to
	Printer-Specific Formats illustration</link>.
	</para>

		<figure id="small9" float="0">
			<title>Raster to Printer-Specific Formats.</title>
			<mediaobject><imageobject role="html"><imagedata fileref="images/9small.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/9small.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/9small"/></imageobject></mediaobject>
		</figure>
	</sect2>

	<sect2>
	<title>CUPS Backends</title>

	<para>
<indexterm significance="normal"><primary>CUPS filtering chain</primary></indexterm>
<indexterm significance="normal"><primary>print queue</primary></indexterm>
	The last part of any CUPS filtering chain is a backend. Backends
	are special programs that send the print-ready file to the final
	device. There is a separate backend program for any transfer
	protocol for sending print jobs over the network, and one for every local
	interface. Every CUPS print queue needs to have a CUPS <quote>device-URI</quote>
	associated with it. The device URI is the way to encode the backend
	used to send the job to its destination. Network device-URIs use
	two slashes in their syntax, local device URIs only one, as you can
	see from the following list. Keep in mind that local interface names
	may vary greatly from my examples, if your OS is not Linux:
	</para>

	<variablelist>
		<varlistentry><term>usb</term>
		<listitem><para>
		This backend sends print files to USB-connected printers. An
		example for the CUPS device-URI to use is
		<filename moreinfo="none">usb:/dev/usb/lp0</filename>.
		</para></listitem></varlistentry>

		<varlistentry><term>serial</term>
		<listitem><para>
		This backend sends print files to serially connected printers.
		An example for the CUPS device-URI to use is
		<filename moreinfo="none">serial:/dev/ttyS0?baud=11500</filename>.
		</para></listitem></varlistentry>

		<varlistentry><term>parallel</term>
		<listitem><para>
		This backend sends print files to printers connected to the
		parallel port. An example for the CUPS device-URI to use is
		<filename moreinfo="none">parallel:/dev/lp0</filename>.
		</para></listitem></varlistentry>

		<varlistentry><term>SCSI</term>
		<listitem><para>
		This backend sends print files to printers attached to the
		SCSI interface. An example for the CUPS device-URI to use is
		<filename moreinfo="none">scsi:/dev/sr1</filename>.
		</para></listitem></varlistentry>

		<varlistentry><term>lpd</term>
		<listitem><para>
		This backend sends print files to LPR/LPD-connected network
		printers. An example for the CUPS device-URI to use is
		<filename moreinfo="none">lpd://remote_host_name/remote_queue_name</filename>.
		</para></listitem></varlistentry>

		<varlistentry><term>AppSocket/HP JetDirect</term>
		<listitem><para>
		This backend sends print files to AppSocket (a.k.a., HP
		JetDirect) connected network printers. An example for the CUPS
		device-URI to use is
		<filename moreinfo="none">socket://10.11.12.13:9100</filename>.
		</para></listitem></varlistentry>

		<varlistentry><term>ipp</term>
		<listitem><para>
		This backend sends print files to IPP-connected network
		printers (or to other CUPS servers). Examples for CUPS device-URIs
		to use are
		<filename moreinfo="none">ipp:://192.193.194.195/ipp</filename>
		(for many HP printers) and
		<filename moreinfo="none">ipp://remote_cups_server/printers/remote_printer_name</filename>.
		</para></listitem></varlistentry>

		<varlistentry><term>http</term>
		<listitem><para>
		This backend sends print files to HTTP-connected printers.
		(The http:// CUPS backend is only a symlink to the ipp:// backend.)
		Examples for the CUPS device-URIs to use are
		<filename moreinfo="none">http:://192.193.194.195:631/ipp</filename>
		(for many HP printers) and
		<filename moreinfo="none">http://remote_cups_server:631/printers/remote_printer_name</filename>.
		</para></listitem></varlistentry>

		<varlistentry><term>smb</term>
		<listitem><para>
		This backend sends print files to printers shared by a Windows
		host. Examples of CUPS device-URIs that may be used includes:
		</para>

		<para>
		<simplelist type="vert">
		<member><filename moreinfo="none">smb://workgroup/server/printersharename</filename></member>
		<member><filename moreinfo="none">smb://server/printersharename</filename></member>
		<member><filename moreinfo="none">smb://username:password@workgroup/server/printersharename</filename></member>
		<member><filename moreinfo="none">smb://username:password@server/printersharename</filename></member>
		</simplelist>
		</para>

		<para>
		The smb:// backend is a symlink to the Samba utility
		<parameter moreinfo="none">smbspool</parameter> (does not ship with CUPS). If the
		symlink is not present in your CUPS backend directory, have your
		root user create it: <literal>ln -s `which smbspool'
		/usr/lib/cups/backend/smb</literal>.
		</para></listitem></varlistentry>
	</variablelist>

	<para>
	It is easy to write your own backends as shell or Perl scripts if you
	need any modification or extension to the CUPS print system. One
	reason could be that you want to create <quote>special</quote> printers that send
	the print jobs as email (through a <quote>mailto:/</quote> backend), convert them to
	PDF (through a <quote>pdfgen:/</quote> backend) or dump them to <quote>/dev/null</quote>. (In
	fact, I have the systemwide default printer set up to be connected to
	a devnull:/ backend: there are just too many people sending jobs
	without specifying a printer, and scripts and programs that do not name
	a printer. The systemwide default deletes the job and sends a polite
	email back to the $USER asking him or her to always specify the correct
	printer name.)
	</para>

	<para>
<indexterm significance="normal"><primary>lpinfo</primary></indexterm>
<indexterm significance="normal"><primary>CUPS backends</primary></indexterm>
	Not all of the mentioned backends may be present on your system or
	usable (depending on your hardware configuration). One test for all
	available CUPS backends is provided by the <emphasis>lpinfo</emphasis>
	utility. Used with the <option>-v</option> parameter, it lists
	all available backends:
	</para>

	<para><screen format="linespecific">
	<prompt moreinfo="none">$ </prompt><userinput moreinfo="none">lpinfo -v</userinput>
	</screen></para>
	</sect2>

	<sect2>
	<title>The Role of <parameter moreinfo="none">cupsomatic/foomatic</parameter></title>

	<para>
	<indexterm significance="normal"><primary>cupsomatic</primary></indexterm>
	<indexterm significance="normal"><primary>foomatic</primary></indexterm>
<indexterm significance="normal"><primary>PPDs</primary></indexterm>
<indexterm significance="normal"><primary>Foomatic Printer</primary></indexterm>
<indexterm significance="normal"><primary>Linuxprinting.org</primary></indexterm>
	<parameter moreinfo="none">cupsomatic</parameter> filters may be the most widely used on CUPS
	installations. You must be clear that these were not
	developed by the CUPS people. They are a third-party add-on to
	CUPS. They utilize the traditional Ghostscript devices to render jobs
	for CUPS. When troubleshooting, you should know about the
	difference. Here the whole rendering process is done in one stage,
	inside Ghostscript, using an appropriate device for the target
	printer. <parameter moreinfo="none">cupsomatic</parameter> uses PPDs that are generated from the Foomatic
	Printer &amp; Driver Database at Linuxprinting.org.
	</para>

	<para>
	You can recognize these PPDs from the line calling the
	<parameter moreinfo="none">cupsomatic</parameter> filter:
<programlisting format="linespecific">
*cupsFilter: "application/vnd.cups-postscript  0  cupsomatic"
</programlisting>
	You may find this line among the first 40 or so lines of the PPD
	file. If you have such a PPD installed, the printer shows up in the
	CUPS Web interface with a <parameter moreinfo="none">foomatic</parameter> namepart for
	the driver description. <parameter moreinfo="none">cupsomatic</parameter> is a Perl script that runs
	Ghostscript with all the complicated command line options
	autoconstructed from the selected PPD and command line options given to
	the print job.
	</para>

	<para>
	<indexterm significance="normal"><primary>point'n'print</primary></indexterm>
<indexterm significance="normal"><primary>foomatic-rip</primary></indexterm>
<indexterm significance="normal"><primary>Adobe specifications</primary></indexterm>
<indexterm significance="normal"><primary>hi-res photo</primary></indexterm>
<indexterm significance="normal"><primary>normal color</primary></indexterm>
<indexterm significance="normal"><primary>grayscale</primary></indexterm>
<indexterm significance="normal"><primary>draft</primary></indexterm>
<indexterm significance="normal"><primary>media type</primary></indexterm>
<indexterm significance="normal"><primary>resolution</primary></indexterm>
<indexterm significance="normal"><primary>inktype</primary></indexterm>
<indexterm significance="normal"><primary>dithering algorithm</primary></indexterm>
	However, <parameter moreinfo="none">cupsomatic</parameter> is now deprecated. Its PPDs (especially the first
	generation of them, still in heavy use out there) are not meeting the
	Adobe specifications. You might also suffer difficulties when you try
	to download them with <quote>Point'n'Print</quote> to Windows clients. A better
	and more powerful successor is now available: it is called <parameter moreinfo="none">foomatic-rip</parameter>. To use
	<parameter moreinfo="none">foomatic-rip</parameter> as a filter with CUPS, you need the new type of PPDs, which
	have a similar but different line:
<programlisting format="linespecific">
*cupsFilter: "application/vnd.cups-postscript  0  foomatic-rip"
</programlisting>
	The PPD-generating engine at Linuxprinting.org has been revamped.
	The new PPDs comply with the Adobe spec. They also provide a
	new way to specify different quality levels (hi-res photo, normal
	color, grayscale, and draft) with a single click, whereas before you
	could have required five or more different selections (media type,
	resolution, inktype, and dithering algorithm). There is support for
	custom-size media built in. There is support to switch
	print options from page to page in the middle of a job. And the
	best thing is that the new <constant>foomatic-rip</constant> works seamlessly with all
	legacy spoolers too (like LPRng, BSD-LPD, PDQ, PPR, and so on), providing
	for them access to use PPDs for their printing.
	</para>
	</sect2>

	<sect2>
	<title>The Complete Picture</title>

	<para>
	If you want to see an overview of all the filters and how they
	relate to each other, the complete picture of the puzzle is at the end
	of this chapter.
	</para>
	</sect2>

	<sect2>
	<title><filename moreinfo="none">mime.convs</filename></title>

	<para>
	CUPS autoconstructs all possible filtering chain paths for any given
	MIME type and every printer installed. But how does it decide in
	favor of or against a specific alternative?  (There may be cases
	where there is a choice of two or more possible filtering chains for
	the same target printer.) Simple. You may have noticed the figures in
	the third column of the mime.convs file. They represent virtual costs
	assigned to this filter. Every possible filtering chain will sum up to
	a total <quote>filter cost.</quote> CUPS decides for the most <quote>inexpensive</quote> route.
	</para>

	<tip><para>
<indexterm significance="normal"><primary>cupsd.conf</primary></indexterm>
<indexterm significance="normal"><primary>FilterLimit</primary></indexterm>
	Setting <parameter moreinfo="none">FilterLimit 1000</parameter> in
	<filename moreinfo="none">cupsd.conf</filename> will not allow more filters to
	run concurrently than will consume a total of 1000 virtual filter
	cost. This is an efficient way to limit the load of any CUPS
	server by setting an appropriate <quote>FilterLimit</quote> value. A FilterLimit of
	200 allows roughly one job at a time, while a FilterLimit of 1000 allows
	approximately five jobs maximum at a time.
	</para></tip>
	</sect2>

	<sect2>
	<title><quote>Raw</quote> Printing</title>

	<para>
<indexterm significance="normal"><primary>PPD</primary></indexterm>
<indexterm significance="normal"><primary>lpadmin</primary></indexterm>
<indexterm significance="normal"><primary>rawprinter</primary></indexterm>
	You can tell CUPS to print (nearly) any file <quote>raw</quote>. <quote>Raw</quote> means it will not be
	filtered. CUPS will send the file to the printer <quote>as is</quote> without bothering if the printer is able
	to digest it. Users need to take care themselves that they send sensible data formats only. Raw printing can
	happen on any queue if the <quote><parameter moreinfo="none">-o raw</parameter></quote> option is specified on the command
	line. You can also set up raw-only queues by simply not associating any PPD with it. This command:
<screen format="linespecific">
<prompt moreinfo="none">$ </prompt><userinput moreinfo="none">lpadmin -P rawprinter -v socket://11.12.13.14:9100 -E</userinput>
</screen>
	sets up a queue named <quote>rawprinter</quote>, connected via the <quote>socket</quote> protocol (a.k.a.
	<quote>HP JetDirect</quote>) to the device at IP address 11.12.1.3.14, using port 9100. (If you had added a
	PPD with <literal>-P /path/to/PPD</literal> to this command line, you would have installed a
	<quote>normal</quote> print queue.)
	</para>

	<para>
	CUPS will automatically treat each job sent to a queue as a <quote>raw</quote> one
	if it can't find a PPD associated with the queue. However, CUPS will
	only send known MIME types (as defined in its own mime.types file) and
	refuse others.
	</para>
	</sect2>

	<sect2>
	<title>application/octet-stream Printing</title>

	<para>
<indexterm significance="normal"><primary>/etc/cups/mime.types</primary></indexterm>
<indexterm significance="normal"><primary>application/octet-stream</primary></indexterm>
	Any MIME type with no rule in the <filename moreinfo="none">/etc/cups/mime.types</filename> file is regarded as unknown
	or <parameter moreinfo="none">application/octet-stream</parameter> and will not be
	sent. Because CUPS refuses to print unknown MIME types by default,
	you will probably have experienced that print jobs originating
	from Windows clients were not printed. You may have found an error
	message in your CUPS logs like:
	</para>

	<para><computeroutput moreinfo="none">
	 Unable to convert file 0 to printable format for job
	</computeroutput></para>

	<para>
	To enable the printing of <parameter moreinfo="none">application/octet-stream</parameter> files, edit
	these two files:
	</para>

	<itemizedlist>
		<listitem><para><filename moreinfo="none">/etc/cups/mime.convs</filename></para></listitem>

		<listitem><para><filename moreinfo="none">/etc/cups/mime.types</filename></para></listitem>
	</itemizedlist>

	<para>
<indexterm significance="normal"><primary>raw mode</primary></indexterm>
	Both contain entries (at the end of the respective files) that must be uncommented to allow raw mode
	operation for <parameter moreinfo="none">application/octet-stream</parameter>. In <filename moreinfo="none">/etc/cups/mime.types</filename>
	make sure this line is present:
	<indexterm significance="normal"><primary>application/octet-stream</primary></indexterm>
<programlisting format="linespecific">
application/octet-stream
</programlisting>
	This line (with no specific autotyping rule set) makes all files
	not otherwise auto-typed a member of <parameter moreinfo="none">application/octet-stream</parameter>. In
	<filename moreinfo="none">/etc/cups/mime.convs</filename>, have this
	line: 
<programlisting format="linespecific">
application/octet-stream   application/vnd.cups-raw   0   -
</programlisting>
	<indexterm significance="normal"><primary>MIME</primary></indexterm>
	This line tells CUPS to use the <emphasis>Null Filter</emphasis>
	(denoted as <quote>-</quote>, doing nothing at all) on
	<parameter moreinfo="none">application/octet-stream</parameter>, and tag the result as
	<parameter moreinfo="none">application/vnd.cups-raw</parameter>. This last one is
	always a green light to the CUPS scheduler to now hand the file over
	to the backend connecting to the printer and sending it over.
	</para>

	<note><para>
	Editing the <filename moreinfo="none">mime.convs</filename> and the <filename moreinfo="none">mime.types</filename> file does not
	<emphasis>enforce</emphasis> <quote>raw</quote> printing, it only <emphasis>allows</emphasis> it.
	</para></note>

	<formalpara>
	<title>Background</title>

	<para>
<indexterm significance="normal"><primary>security-aware</primary></indexterm>
<indexterm significance="normal"><primary>MIME type</primary></indexterm>
<indexterm significance="normal"><primary>/etc/cups/mime.types</primary></indexterm>
<indexterm significance="normal"><primary>/etc/cups/mime.convs</primary></indexterm>
	That CUPS is a more security-aware printing system than traditional ones
	does not by default allow one to send deliberate (possibly binary)
	data to printing devices. (This could be easily abused to launch a
	Denial of Service attack on your printer(s), causing at least the loss
	of a lot of paper and ink.) <quote>Unknown</quote> data are regarded by CUPS
	as <emphasis>MIME type</emphasis> <emphasis>application/octet-stream</emphasis>. While you
	<emphasis>can</emphasis> send data <quote>raw</quote>, the MIME type for these must
	be one that is known to CUPS and allowed by it. The file
	<filename moreinfo="none">/etc/cups/mime.types</filename> defines the <quote>rules</quote> of how CUPS
	recognizes MIME types. The file <filename moreinfo="none">/etc/cups/mime.convs</filename> decides which file
	conversion filter(s) may be applied to which MIME types.
	</para>
	</formalpara>
	</sect2>

	<sect2>
	<title>PostScript Printer Descriptions for Non-PostScript Printers</title>

	<para>
	<indexterm significance="normal"><primary>PPD</primary></indexterm>
<indexterm significance="normal"><primary>non-PostScript</primary></indexterm>
<indexterm significance="normal"><primary>PostScript</primary></indexterm>
<indexterm significance="normal"><primary>RIP</primary></indexterm>
<indexterm significance="normal"><primary>Ghostscript</primary></indexterm>
<indexterm significance="normal"><primary>device-specific commands</primary></indexterm>
	Originally PPDs were meant to be used for PostScript printers
	only. Here, they help to send device-specific commands and settings
	to the RIP, which processes the job file. CUPS has extended this
	scope for PPDs to cover non-PostScript printers too. This was not
	difficult, because it is a standardized file format. In a way
	it was logical too: CUPS handles PostScript and uses a PostScript
	RIP (Ghostscript) to process the job files. The only difference is that
	a PostScript printer has the RIP built-in, for other types of
	printers the Ghostscript RIP runs on the host computer.
	</para>

	<para>
	PPDs for a non-PostScript printer have a few lines that are unique to
	CUPS. The most important one looks similar to this:
	<indexterm significance="normal"><primary>application/vnd.cups-raster</primary></indexterm>
<programlisting format="linespecific">
*cupsFilter: application/vnd.cups-raster  66   rastertoprinter
</programlisting>
	It is the last piece in the CUPS filtering puzzle. This line tells the
	CUPS daemon to use as a last filter <parameter moreinfo="none">rastertoprinter</parameter>. This filter
	should be served as input an <parameter moreinfo="none">application/vnd.cups-raster</parameter> MIME type
	file. Therefore, CUPS should autoconstruct a filtering chain, which
	delivers as its last output the specified MIME type. This is then
	taken as input to the specified <parameter moreinfo="none">rastertoprinter</parameter> filter. After
	the last filter has done its work (<parameter moreinfo="none">rastertoprinter</parameter> is a Gutenprint
	filter), the file should go to the backend, which sends it to the
	output device.
	</para>

	<para>
	CUPS by default ships only a few generic PPDs, but they are good for
	several hundred printer models. You may not be able to control
	different paper trays, or you may get larger margins than your
	specific model supports. See Table 21.1<link linkend="cups-ppds"/> for summary information.
	</para>

	<table frame="all" id="cups-ppds">
		<title>PPDs Shipped with CUPS</title>
		<tgroup cols="2" align="left">
			<colspec align="left"/>
			<colspec align="justify" colwidth="1*"/>
			<thead><row><entry>PPD file</entry><entry>Printer type</entry></row></thead>
			<tbody>
			<row><entry>deskjet.ppd</entry><entry>older HP inkjet printers and compatible</entry></row>

			<row><entry>deskjet2.ppd</entry> <entry>newer HP inkjet printers and compatible </entry> </row>

			<row><entry>dymo.ppd</entry> <entry>label printers </entry> </row>

			<row><entry>epson9.ppd</entry> <entry>Epson 24-pin impact printers and compatible </entry> </row>

			<row><entry>epson24.ppd</entry> <entry>Epson 24-pin impact printers and compatible </entry> </row>

			<row><entry>okidata9.ppd</entry> <entry>Okidata 9-pin impact printers and compatible </entry> </row>

			<row><entry>okidat24.ppd</entry> <entry>Okidata 24-pin impact printers and compatible </entry> </row>

			<row><entry>stcolor.ppd</entry> <entry>older Epson Stylus Color printers </entry> </row>

			<row><entry>stcolor2.ppd</entry> <entry>newer Epson Stylus Color printers </entry> </row>

			<row><entry>stphoto.ppd</entry> <entry>older Epson Stylus Photo printers </entry> </row>

			<row><entry>stphoto2.ppd</entry> <entry>newer Epson Stylus Photo printers </entry> </row>

			<row><entry>laserjet.ppd</entry> <entry>all PCL printers </entry> </row>

			</tbody>
		</tgroup>
	</table>

	</sect2>

	<sect2>
	<title><emphasis>cupsomatic/foomatic-rip</emphasis> Versus <emphasis>Native CUPS</emphasis> Printing</title>

	<para>
	<indexterm significance="normal"><primary>cupsomatic</primary></indexterm>
	<indexterm significance="normal"><primary>foomatic-rip</primary></indexterm>
	Native CUPS rasterization works in two steps:
	</para>

	<itemizedlist>
		<listitem><para>
<indexterm significance="normal"><primary>pstoraster</primary></indexterm>
		First is the <parameter moreinfo="none">pstoraster</parameter> step. It uses the special CUPS
		<indexterm significance="normal"><primary>ESP</primary><secondary>Ghostscript</secondary></indexterm>
		device from ESP Ghostscript 7.05.x as its tool.
		</para></listitem>

		<listitem><para>
		Second is the <parameter moreinfo="none">rasterdriver</parameter> step. It uses various
		device-specific filters; there are several vendors who provide good
		quality filters for this step. Some are free software, some are
		shareware, and some are proprietary.
		</para></listitem>
	</itemizedlist>

	<para>
	Often this produces better quality (and has several more advantages) than other methods.
	This is shown in <link linkend="cupsomatic-dia"> the cupsomatic/foomatic Processing Versus Native CUPS
	illustration</link>.
	</para>

	<figure id="cupsomatic-dia" float="0">
		<title>cupsomatic/foomatic Processing Versus Native CUPS.</title>
		<mediaobject><imageobject role="html"><imagedata fileref="images/10small.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/10small.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/10small"/></imageobject></mediaobject>
	</figure>

	<para>
	One other method is the <parameter moreinfo="none">cupsomatic/foomatic-rip</parameter>
	way. Note that <parameter moreinfo="none">cupsomatic</parameter> is <emphasis>not</emphasis> made by the CUPS
	developers. It is an independent contribution to printing development,
	made by people from Linuxprinting.org.<footnote><para>See also <ulink noescape="1" url="http://www.cups.org/cups-help.html">http://www.cups.org/cups-help.html</ulink></para></footnote>
	<parameter moreinfo="none">cupsomatic</parameter> is no longer developed, maintained, or supported. It now been
	replaced by <parameter moreinfo="none">foomatic-rip</parameter>. <parameter moreinfo="none">foomatic-rip</parameter> is a complete rewrite
	of the old <parameter moreinfo="none">cupsomatic</parameter> idea, but very much improved and generalized to
	other (non-CUPS) spoolers. An upgrade to <parameter moreinfo="none">foomatic-rip</parameter> is strongly
	advised, especially if you are upgrading to a recent version of CUPS,
	too.
	</para>

	<para>
	<indexterm significance="normal"><primary>cupsomatic</primary></indexterm>
	<indexterm significance="normal"><primary>foomatic</primary></indexterm>
	Like the old <parameter moreinfo="none">cupsomatic</parameter> method, the <parameter moreinfo="none">foomatic-rip</parameter> (new) method
	from Linuxprinting.org uses the traditional Ghostscript print file processing, doing everything in a single
	step. It therefore relies on all the other devices built into Ghostscript. The quality is as good (or bad) as
	Ghostscript rendering is in other spoolers. The advantage is that this method supports many printer models not
	supported (yet) by the more modern CUPS method.
	</para>

	<para>
	Of course, you can use both methods side by side on one system (and even for one printer, if you set up
	different queues) and find out which works best for you.
	</para>

	<para>
<indexterm significance="normal"><primary>cupsomatic</primary></indexterm>
<indexterm significance="normal"><primary>pstoraster</primary></indexterm>
<indexterm significance="normal"><primary>rastertosomething</primary></indexterm>
<indexterm significance="normal"><primary>rasterization</primary></indexterm>
<indexterm significance="normal"><primary>Foomatic/cupsomatic</primary></indexterm>
<indexterm significance="normal"><primary>rendering</primary></indexterm>
	<parameter moreinfo="none">cupsomatic</parameter> kidnaps the print file after the
	<parameter moreinfo="none">application/vnd.cups-postscript</parameter> stage and deviates it through the CUPS-external,
	systemwide Ghostscript installation. Therefore, the print file bypasses the <parameter moreinfo="none">pstoraster</parameter>
	filter (and also bypasses the CUPS raster drivers <parameter moreinfo="none">rastertosomething</parameter>). After Ghostscript
	finished its rasterization, <parameter moreinfo="none">cupsomatic</parameter> hands the rendered file directly to the CUPS
	backend. <link linkend="cupsomatic-dia">cupsomatic/foomatic Processing Versus Native
	CUPS</link>, illustrates the difference between native CUPS rendering and the
	<parameter moreinfo="none">Foomatic/cupsomatic</parameter> method.
	</para>
	</sect2>

	<sect2>
	<title>Examples for Filtering Chains</title>

	<para>
	Here are a few examples of commonly occurring filtering chains to
	illustrate the workings of CUPS.
	</para>

	<para>
<indexterm significance="normal"><primary>HP JetDirect</primary></indexterm>
<indexterm significance="normal"><primary>PostScript</primary></indexterm>
<indexterm significance="normal"><primary>two-up</primary></indexterm>
<indexterm significance="normal"><primary>duplex</primary></indexterm>
	Assume you want to print a PDF file to an HP JetDirect-connected
	PostScript printer, but you want to print pages 3-5, 7, and 11-13
	only, and you want to print them <quote>two-up</quote> and <quote>duplex</quote>:
	</para>

	<itemizedlist>
	<listitem><para>Your print options (page selection as required, two-up,
	duplex) are passed to CUPS on the command line.</para></listitem>

	<listitem><para>The (complete) PDF file is sent to CUPS and autotyped as
	<parameter moreinfo="none">application/pdf</parameter>.</para></listitem>

	<listitem><para>The file therefore must first pass the
	<parameter moreinfo="none">pdftops</parameter> prefilter, which produces PostScript
	MIME type <parameter moreinfo="none">application/postscript</parameter> (a preview here
	would still show all pages of the original PDF).</para></listitem>

	<listitem><para>The file then passes the <parameter moreinfo="none">pstops</parameter>
	filter that applies the command line options: it selects pages
	2-5, 7, and 11-13, creates the imposed layout <quote>two pages on one sheet</quote>, and
	inserts the correct <quote>duplex</quote> command (as defined in the printer's
	PPD) into the new PostScript file; the file is now of PostScript MIME
	type
	<parameter moreinfo="none">application/vnd.cups-postscript</parameter>.</para></listitem>

	<listitem><para>The file goes to the <parameter moreinfo="none">socket</parameter>
	backend, which transfers the job to the printers.</para></listitem>
	</itemizedlist>

	<para>
	The resulting filter chain, therefore, is as shown in <link linkend="pdftosocket">the PDF to socket chain
	illustration</link>.
	</para>

<indexterm significance="normal"><primary>pdftosocket</primary></indexterm>
	<figure id="pdftosocket" float="0">
		<title>PDF to Socket Chain.</title>
		<mediaobject><imageobject role="html"><imagedata fileref="images/pdftosocket.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/pdftosocket.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/pdftosocket"/></imageobject></mediaobject>
	</figure>

	<para>
<indexterm significance="normal"><primary>USB</primary></indexterm>
<indexterm significance="normal"><primary>Epson Stylus</primary></indexterm>
<indexterm significance="normal"><primary>stphoto2.ppd</primary></indexterm>
	Assume you want to print the same filter to an USB-connected Epson Stylus Photo Printer installed with the CUPS
	<filename moreinfo="none">stphoto2.ppd</filename>. The first few filtering stages are nearly the same:
	</para>

	<itemizedlist>
		<listitem><para>
		Your print options (page selection as required, two-up,
		duplex) are passed to CUPS on the command line.
		</para></listitem>

		<listitem><para>
		The (complete) PDF file is sent to CUPS and autotyped as
		<parameter moreinfo="none">application/pdf</parameter>.
		</para></listitem>

		<listitem><para>
<indexterm significance="normal"><primary>pdftops</primary></indexterm>
<indexterm significance="normal"><primary>PDF</primary></indexterm>
		The file must first pass the <parameter moreinfo="none">pdftops</parameter> prefilter, which produces PostScript
		MIME type <parameter moreinfo="none">application/postscript</parameter> (a preview here would still show all
		pages of the original PDF).
		</para></listitem>

		<listitem><para>
<indexterm significance="normal"><primary>pstops</primary></indexterm>
<indexterm significance="normal"><primary>duplex printing</primary></indexterm>
		The file then passes the <quote>pstops</quote> filter that applies
		the command line options: it selects the pages 2-5, 7, and 11-13,
		creates the imposed layout <quote>two pages on one sheet,</quote> and inserts the
		correct <quote>duplex</quote> command (oops <?latex --- ?> this printer and PPD
		do not support duplex printing at all, so this option will
		be ignored) into the new PostScript file; the file is now of PostScript
		MIME type <parameter moreinfo="none">application/vnd.cups-postscript</parameter>.
		</para></listitem>

		<listitem><para>
		The file then passes the <parameter moreinfo="none">pstoraster</parameter> stage and becomes MIME type
		<parameter moreinfo="none">application/cups-raster</parameter>.
		</para></listitem>

		<listitem><para>
<indexterm significance="normal"><primary>rastertoepson</primary></indexterm>
		Finally, the <parameter moreinfo="none">rastertoepson</parameter> filter
		does its work (as indicated in the printer's PPD), creating the
		printer-specific raster data and embedding any user-selected
		print options into the print data stream.
		</para></listitem>

		<listitem><para>
		The file goes to the <parameter moreinfo="none">usb</parameter> backend, which transfers the job to the printers.
		</para></listitem>
	</itemizedlist>

	<para>
	The resulting filter chain therefore is as shown in <link linkend="pdftoepsonusb">the PDF to USB Chain
	illustration</link>.
	</para>

	<figure id="pdftoepsonusb" float="0">
		<title>PDF to USB Chain.</title>
		<mediaobject><imageobject role="html"><imagedata fileref="images/pdftoepsonusb.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/pdftoepsonusb.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/pdftoepsonusb"/></imageobject></mediaobject>
	</figure>
	</sect2>

	<sect2>
	<title>Sources of CUPS Drivers/PPDs</title>

	<para>
	On the Internet you can now find many thousands of CUPS-PPD files
	(with their companion filters), in many national languages
	supporting more than 1,000 non-PostScript models.
	</para>

	<itemizedlist>
		<indexterm significance="normal"><primary>ESP</primary><secondary>Print Pro</secondary></indexterm>
		<indexterm significance="normal"><primary>PrintPro</primary><see>ESP Print Pro</see></indexterm>
		<listitem><para>
		<ulink url="http://www.easysw.com/printpro/">ESP PrintPro</ulink>
		(commercial, non-free) is packaged with more than 3,000 PPDs, ready for
		successful use <quote>out of the box</quote> on Linux, Mac OS X, IBM-AIX,
		HP-UX, Sun-Solaris, SGI-IRIX, Compaq Tru64, Digital UNIX, and
		other commercial Unices (it is written by the CUPS developers
		themselves and its sales help finance the further development of
		CUPS, as they feed their creators).
		</para></listitem>

		<listitem><para>
		The <ulink url="http://gimp-print.sourceforge.net/">Gutenprint Project</ulink>
		(GPL, free software) provides around 140 PPDs (supporting nearly 400 printers, many driven
		to photo quality output), to be used alongside the Gutenprint CUPS filters.
		</para></listitem>

		<listitem><para>
		<ulink url="http://www.turboprint.de/english.html/">TurboPrint </ulink> (shareware, non-free) supports
		roughly the same number of printers in excellent quality.
		</para></listitem>

		<listitem><para>
		<ulink url="http://www-124.ibm.com/developerworks/oss/linux/projects/omni/">OMNI </ulink>
		(LPGL, free) is a package made by IBM, now containing support for more
		than 400 printers, stemming from the inheritance of IBM OS/2 know-how
		ported over to Linux (CUPS support is in a beta stage at present).
		</para></listitem>

		<listitem><para>
		<ulink url="http://hpinkjet.sourceforge.net/">HPIJS </ulink> (BSD-style licenses, free)
		supports approximately 150 of HP's own printers and also provides
		excellent print quality now (currently available only via the Foomatic path).
		</para></listitem>

		<listitem><para>
		<ulink url="http://www.linuxprinting.org/">Foomatic/cupsomatic </ulink>
		(LPGL, free) from Linuxprinting.org provide PPDs for practically every Ghostscript
		filter known to the world (including Omni, Gutenprint, and HPIJS).
		</para></listitem>
	</itemizedlist>

	</sect2>

	<sect2>
	<title>Printing with Interface Scripts</title>

	<para>
<indexterm significance="normal"><primary>PCL</primary></indexterm>
<indexterm significance="normal"><primary>lpadmin</primary></indexterm>
	CUPS also supports the use of <quote>interface scripts</quote> as known from
	System V AT&amp;T printing systems. These are often used for PCL
	printers, from applications that generate PCL print jobs. Interface
	scripts are specific to printer models. They have a role similar to
	PPDs for PostScript printers. Interface scripts may inject the Escape
	sequences as required into the print data stream if the user, for example, selects
	a certain paper tray, or changes paper orientation, or uses A3
	paper. Interface scripts are practically unknown in the Linux
	realm. On HP-UX platforms they are more often used. You can use any
	working interface script on CUPS too. Just install the printer with
	the <literal>-i</literal> option:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">lpadmin -p pclprinter -v socket://11.12.13.14:9100 \
          -i /path/to/interface-script</userinput>
</screen></para>

	<para>
	Interface scripts might be the <quote>unknown animal</quote> to many. However,
	with CUPS they provide the easiest way to plug in your own custom-written filtering
	script or program into one specific print queue (some information about the traditional
	use of interface scripts is found at
	<ulink noescape="1" url="http://playground.sun.com/printing/documentation/interface.html">
	http://playground.sun.com/printing/documentation/interface.html</ulink>).
	</para>
	</sect2>
</sect1>

<sect1>
<title>Network Printing (Purely Windows)</title>

<para>
Network printing covers a lot of ground. To understand what exactly
goes on with Samba when it is printing on behalf of its Windows
clients, let's first look at a <quote>purely Windows</quote> setup: Windows clients
with a Windows NT print server.
</para>

<sect2>
<title>From Windows Clients to an NT Print Server</title>

<para>
Windows clients printing to an NT-based print server have two
options. They may:
<indexterm significance="normal"><primary>GDI</primary></indexterm>
<indexterm significance="normal"><primary>EMF</primary></indexterm>
</para>


<itemizedlist>
	<listitem><para>Execute the driver locally and render the GDI output
			(EMF) into the printer-specific format on their own.
	</para></listitem>

	<listitem><para>Send the GDI output (EMF) to the server, where the
	driver is executed to render the printer-specific output.
	</para></listitem>
</itemizedlist>

<para>
Both print paths are shown in the flowcharts in <link linkend="small11">
Print Driver Execution on the Client</link>, and
<link linkend="small12">Print Driver Execution on the Server</link>.
</para>
</sect2>

<sect2>
<title>Driver Execution on the Client</title>

<para>
In the first case, the print server must spool the file as raw, meaning it shouldn't touch the job file and try
to convert it in any way. This is what a traditional UNIX-based print server can do too, and at a better
performance and more reliably than an NT print server. This is what most Samba administrators probably are
familiar with. One advantage of this setup is that this <quote>spooling-only</quote> print server may be used
even if no driver(s) for UNIX is available. It is sufficient to have the Windows client drivers available and
installed on the clients. This is illustrated in <link linkend="small11">the Print Driver Execution on the
Client diagram</link>.
</para>

<figure id="small11" float="0">
	<title>Print Driver Execution on the Client.</title>
	<mediaobject><imageobject role="html"><imagedata fileref="images/11small.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/11small.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/11small"/></imageobject></mediaobject>
</figure>

</sect2>

<sect2>
<title>Driver Execution on the Server</title>


<para>
<indexterm significance="normal"><primary>PostScript</primary></indexterm>
<indexterm significance="normal"><primary>PCL</primary></indexterm>
<indexterm significance="normal"><primary>ESC/P</primary></indexterm>
<indexterm significance="normal"><primary>EMF</primary></indexterm>
<indexterm significance="normal"><primary>GDI</primary></indexterm>
The other path executes the printer driver on the server. The client transfers print files in EMF format to
the server. The server uses the PostScript, PCL, ESC/P, or other driver to convert the EMF file into the
printer-specific language. It is not possible for UNIX to do the same. Currently, there is no program or
method to convert a Windows client's GDI output on a UNIX server into something a printer could understand.
This is illustrated in <link linkend="small12">the Print Driver Execution on the Server diagram</link>.
</para>

	<figure id="small12" float="0">
		<title>Print Driver Execution on the Server.</title>
		<mediaobject><imageobject role="html"><imagedata fileref="images/12small.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/12small.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/12small"/></imageobject></mediaobject>
	</figure>

<para>
However, something similar is possible with CUPS, so read on.
</para>
</sect2>
</sect1>

<sect1>
<title>Network Printing (Windows Clients and UNIX/Samba Print
Servers)</title>

<para>
Since UNIX print servers <emphasis>cannot</emphasis> execute the Win32
program code on their platform, the picture is somewhat
different. However, this does not limit your options all that
much. On the contrary, you may have a way here to implement printing
features that are not possible otherwise.
</para>

<sect2>
<title>From Windows Clients to a CUPS/Samba Print Server</title>

<para>
Here is a simple recipe showing how you can take advantage of CUPS's
powerful features for the benefit of your Windows network printing
clients:
</para>

<itemizedlist>
	<listitem><para>Let the Windows clients send PostScript to the CUPS
	server.</para></listitem>

	<listitem><para>Let the CUPS server render the PostScript into device-specific raster format.</para></listitem>
</itemizedlist>

<para>
This requires the clients to use a PostScript driver (even if the
printer is a non-PostScript model. It also requires that you have a
driver on the CUPS server.
</para>

<para>
First, to enable CUPS-based printing through Samba, the following options should be set in your <filename moreinfo="none">smb.conf</filename>
file <parameter moreinfo="none">[global]</parameter> section:
</para>

<simplelist>
<member><indexterm>printing</indexterm><parameter>printing = cups</parameter></member>
<member><indexterm>printcap</indexterm><parameter>printcap = cups</parameter></member>
</simplelist>

<para>
When these parameters are specified, all manually set print directives (like <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTCOMMAND" xlink:href="smb.conf.5.html#PRINTCOMMAND">print command</link> or <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LPPAUSECOMMAND" xlink:href="smb.conf.5.html#LPPAUSECOMMAND">lppause command</link>) in <filename moreinfo="none">smb.conf</filename> (as well as in Samba itself) will be
ignored. Instead, Samba will directly interface with CUPS through its application program interface (API), as
long as Samba has been compiled with CUPS library (libcups) support. If Samba has not been compiled with CUPS
support, and if no other print commands are set up, then printing will use the <emphasis>System V</emphasis>
AT&amp;T command set, with the -oraw option automatically passing through (if you want your own defined print
commands to work with a Samba server that has CUPS support compiled in, simply use <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="CLASSICALPRINTING" xlink:href="smb.conf.5.html#CLASSICALPRINTING">classicalprinting = sysv</link>). This is illustrated in <link linkend="13small">the Printing via
CUPS/Samba Server diagram</link>.
</para>

	<figure id="13small" float="0">
		<title>Printing via CUPS/Samba Server.</title>
		<mediaobject><imageobject role="html"><imagedata fileref="images/13small.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/13small.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/13small"/></imageobject></mediaobject>
	</figure>
</sect2>

<sect2>
<title>Samba Receiving Job-Files and Passing Them to CUPS</title>

<para>
Samba <emphasis>must</emphasis> use its own spool directory (it is set by a line similar to <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PATH" xlink:href="smb.conf.5.html#PATH">path = /var/spool/samba</link>, in the <parameter>[printers]</parameter> or <parameter>[printername]</parameter> section of <filename moreinfo="none">smb.conf</filename>). Samba receives the job in its own spool space and passes it
into the spool directory of CUPS (the CUPS spool directory is set by the <parameter moreinfo="none">RequestRoot</parameter>
directive in a line that defaults to <parameter moreinfo="none">RequestRoot /var/spool/cups</parameter>). CUPS checks the
access rights of its spool directory and resets it to healthy values with every restart. We have seen quite a
few people who used a common spooling space for Samba and CUPS, and struggled for weeks with this
<quote>problem.</quote>
</para>

<para>
A Windows user authenticates only to Samba (by whatever means is
configured). If Samba runs on the same host as CUPS, you only need to
allow <quote>localhost</quote> to print. If it runs on different machines, you
need to make sure the Samba host gets access to printing on CUPS.
</para>
</sect2>
</sect1>

<sect1>
<title>Network PostScript RIP</title>

<para>
This section discusses the use of CUPS filters on the server <?latex --- ?> configuration where
clients make use of a PostScript driver with CUPS-PPDs.
</para>


<para>
<indexterm significance="normal"><primary>PostScript</primary></indexterm>
<indexterm significance="normal"><primary>PCL</primary></indexterm>
<indexterm significance="normal"><primary>PJL</primary></indexterm>
PPDs can control all print device options. They are usually provided by the manufacturer <?latex --- ?> if you own
a PostScript printer, that is. PPD files are always a component of PostScript printer drivers on MS Windows or
Apple Mac OS systems. They are ASCII files containing user-selectable print options, mapped to appropriate
PostScript, PCL, or PJL commands for the target printer. Printer driver GUI dialogs translate these options
<quote>on the fly</quote> into buttons and drop-down lists for the user to select.
</para>

<para>
CUPS can load, without any conversions, the PPD file from any Windows (NT is recommended) PostScript driver
and handle the options. There is a Web browser interface to the print options (select <ulink noescape="1" url="http://localhost:631/printers/">http://localhost:631/printers/</ulink> and click on one
<guibutton moreinfo="none">Configure Printer</guibutton> button to see it) or a command line interface (see <literal>man
lpoptions</literal> or see if you have <literal>lphelp</literal> on your system). There are also some
different GUI front-ends on Linux/UNIX, which can present PPD options to users. PPD options are normally meant
to be evaluated by the PostScript RIP on the real PostScript printer.
</para>

<sect2>
<title>PPDs for Non-PS Printers on UNIX</title>


<para>
<indexterm significance="normal"><primary>PPD</primary></indexterm>
CUPS does not limit itself to <quote>real</quote> PostScript printers in its use of PPDs. The CUPS developers
have extended the scope of the PPD concept to also describe available device and driver options for
non-PostScript printers through CUPS-PPDs.
</para>

<para>
This is logical, because CUPS includes a fully featured PostScript interpreter (RIP). This RIP is based on
Ghostscript. It can process all received PostScript (and additionally many other file formats) from clients.
All CUPS-PPDs geared to non-PostScript printers contain an additional line, starting with the keyword
<parameter moreinfo="none">*cupsFilter</parameter>. This line tells the CUPS print system which printer-specific filter to use
for the interpretation of the supplied PostScript. Thus CUPS lets all its printers appear as PostScript
devices to its clients, because it can act as a PostScript RIP for those printers, processing the received
PostScript code into a proper raster print format.
</para>
</sect2>

<sect2>
<title>PPDs for Non-PS Printers on Windows</title>

<para>
<indexterm significance="normal"><primary>PPD</primary></indexterm>
CUPS-PPDs can also be used on Windows clients, on top of a <quote>core</quote> PostScript driver (now
recommended is the CUPS PostScript Driver for Windows NT/200x/XP; you can also use the Adobe one, with
limitations). This feature enables CUPS to do a few tricks no other spooler can do:
</para>

<itemizedlist>
	<listitem><para>
	Act as a networked PostScript RIP handling print files from all client platforms in a uniform way.
	</para></listitem>

	<listitem><para>
	Act as a central accounting and billing server, since all files are passed through the pstops filter and are therefore
	logged in the CUPS <filename moreinfo="none">page_log</filename> file.  <emphasis>Note:</emphasis> this cannot happen with
	<quote>raw</quote> print jobs, which always remain unfiltered per definition.
	</para></listitem>

	<listitem><para>
	Enable clients to consolidate on a single PostScript driver, even for many different target printers.
	</para></listitem>
</itemizedlist>

<para>
Using CUPS PPDs on Windows clients enables them to control all print job settings just as a UNIX client can do.
</para>
</sect2>
</sect1>

<sect1>
<title>Windows Terminal Servers (WTS) as CUPS Clients</title>

<para>
This setup may be of special interest to people experiencing major problems in WTS environments. WTS often
need a multitude of non-PostScript drivers installed to run their clients' variety of different printer
models. This often imposes the price of much increased instability.
</para>

<sect2>
<title>Printer Drivers Running in <quote>Kernel Mode</quote> Cause Many
Problems</title>

<para>
Windows NT printer drivers, which run in <quote>kernel mode</quote>, introduce a high risk for the stability
of the system if the driver is not really stable and well-tested. And there are a lot of bad drivers out
there! Especially notorious is the example of the PCL printer driver that had an additional sound module
running to notify users via soundcard of their finished jobs. Do I need to say that this one was also reliably
causing <quote>blue screens of death</quote> on a regular basis?
</para>

<para>
PostScript drivers are generally well-tested. They are not known to cause any problems, even though they also
run in kernel mode. This might be because until now there have been only two different PostScript drivers: the
one from Adobe and the one from Microsoft. Both are well-tested and are as stable as you can imagine on
Windows. The CUPS driver is derived from the Microsoft one.
</para>
</sect2>

<sect2>
<title>Workarounds Impose Heavy Limitations</title>

<para>
In an attempt to work around problems, site administrators have resorted to restricting the
allowed drivers installed on their WTS to one generic PCL and one PostScript driver. This, however, restricts
the number of printer options available for clients to use. Often they can't get out more than simplex
prints from one standard paper tray, while their devices could do much better if driven by a different driver!
</para>
</sect2>

<sect2>
<title>CUPS: A <quote>Magical Stone</quote>?</title>

<para>
<indexterm significance="normal"><primary>PPD</primary></indexterm>
<indexterm significance="normal"><primary>PostScript</primary></indexterm>
Using a PostScript driver, enabled with a CUPS-PPD, seems to be a very elegant way to overcome all these
shortcomings. There are, depending on the version of Windows OS you use, up to three different PostScript
drivers now available: Adobe, Microsoft, and CUPS PostScript drivers. None of them is known to cause major
stability problems on WTS (even if used with many different PPDs). The clients will be able to (again) choose
paper trays, duplex printing, and other settings. However, there is a certain price for this too: a CUPS
server acting as a PostScript RIP for its clients requires more CPU and RAM than when just acting as a
<quote>raw spooling</quote> device. Plus, this setup is not yet widely tested, although the first feedbacks
look very promising.
</para>
</sect2>

<sect2>
<title>PostScript Drivers with No Major Problems, Even in Kernel
Mode</title>

<para>
<indexterm significance="normal"><primary>DDK</primary></indexterm>
<indexterm significance="normal"><primary>W32X86</primary></indexterm>
<indexterm significance="normal"><primary>PostScript</primary></indexterm>
<indexterm significance="normal"><primary>Visual Studio</primary></indexterm>
<indexterm significance="normal"><primary>Microsoft driver</primary></indexterm>
<indexterm significance="normal"><primary>Adobe</primary></indexterm>
More recent printer drivers on W200x and XP no longer run in kernel mode (unlike Windows NT). However, both
operating systems can still use the NT drivers, running in kernel mode (you can roughly tell which is which as
the drivers in subdirectory <quote>2</quote> of <quote>W32X86</quote> are <quote>old</quote> ones). As was
said before, the Adobe as well as the Microsoft PostScript drivers are not known to cause any stability
problems. The CUPS driver is derived from the Microsoft one. There is a simple reason for this: the MS DDK
(Device Development Kit) for Windows NT (which used to be available at no cost to licensees of Visual Studio)
includes the source code of the Microsoft driver, and licensees of Visual Studio are allowed to use and modify
it for their own driver development efforts. This is what the CUPS people have done. The license does not
allow them to publish the whole of the source code.  However, they have released the <quote>diff</quote> under
the GPL, and if you are the owner of an <quote>MS DDK for Windows NT,</quote> you can check the driver
yourself.
</para>
</sect2>
</sect1>

<sect1>
<title>Configuring CUPS for Driver Download</title>

<para>
As we have said before, all previously known methods to prepare client printer drivers on the Samba server for
download and Point'n'Print convenience of Windows workstations are working with CUPS, too. These methods were
described in <link linkend="classicalprinting">Classical Printing</link>. In reality, this is a pure Samba
business and relates only to the Samba-Windows client relationship.
</para>

<sect2>
<title><emphasis>cupsaddsmb</emphasis>: The Unknown Utility</title>


<para>
<indexterm significance="normal"><primary>cupsaddsmb</primary></indexterm>
The <parameter moreinfo="none">cupsaddsmb</parameter> utility (shipped with all current CUPS versions) is an alternative
method to transfer printer drivers into the Samba <parameter>[print$]</parameter> share. Remember, this
share is where clients expect drivers deposited and set up for download and installation. It makes the sharing
of any (or all) installed CUPS printers quite easy. <literal>cupsaddsmb</literal> can use the Adobe PostScript
driver as well as the newly developed CUPS PostScript driver for Windows NT/200x/XP.
<parameter moreinfo="none">cupsaddsmb</parameter> does <emphasis>not</emphasis> work with arbitrary vendor printer drivers,
but only with the <emphasis>exact</emphasis> driver files that are named in its man page.
</para>

<para>
The CUPS printer driver is available from the CUPS download site. Its package name is
<filename moreinfo="none">cups-samba-[version].tar.gz</filename>. It is preferred over the Adobe drivers because it has a
number of advantages:
</para>

<itemizedlist>
	<listitem><para>It supports a much more accurate page accounting.</para></listitem>

	<listitem><para>It supports banner pages and page labels on all printers.</para></listitem>

	<listitem><para>It supports the setting of a number of job IPP attributes
	(such as job priority, page label, and job billing).</para></listitem>
</itemizedlist>

<para>
However, currently only Windows NT, 2000, and XP are supported by the
CUPS drivers. You will also need to get the respective part of the Adobe driver
if you need to support Windows 95, 98, and Me clients.
</para>
</sect2>

<sect2>
<title>Prepare Your <filename moreinfo="none">smb.conf</filename> for <literal>cupsaddsmb</literal></title>

<para>
Prior to running <literal>cupsaddsmb</literal>, you need the settings in
<filename moreinfo="none">smb.conf</filename> as shown in <link linkend="cupsadd-ex">the <filename moreinfo="none">smb.conf</filename> for cupsaddsmb Usage</link>.
</para>

<example id="cupsadd-ex">
<title>smb.conf for cupsaddsmb Usage</title>
<simplelist>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>load printers</indexterm><parameter>load printers = yes</parameter></member>
<member><indexterm>printing</indexterm><parameter>printing = cups</parameter></member>
<member><indexterm>printcap name</indexterm><parameter>printcap name = cups</parameter></member>

<member> </member><member><parameter>[printers]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = All Printers</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /var/spool/samba</parameter></member>
<member><indexterm>browseable</indexterm><parameter>browseable = no</parameter></member>
<member># setting depends on your requirements</member>
<member><indexterm>guest ok</indexterm><parameter>guest ok = yes</parameter></member>
<member><indexterm>writable</indexterm><parameter>writable = no</parameter></member>
<member><indexterm>printable</indexterm><parameter>printable = yes</parameter></member>
<member><indexterm>printer admin</indexterm><parameter>printer admin = root</parameter></member>
 <member> </member><member><parameter>[print$]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = Printer Drivers</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /etc/samba/drivers</parameter></member>
<member><indexterm>browseable</indexterm><parameter>browseable = yes</parameter></member>
<member><indexterm>guest ok</indexterm><parameter>guest ok = no</parameter></member>
<member><indexterm>read only</indexterm><parameter>read only = yes</parameter></member>
<member><indexterm>write list</indexterm><parameter>write list = root, @smbprintadm</parameter></member>
</simplelist>
</example>
</sect2>

<sect2>
<title>CUPS <quote>PostScript Driver for Windows NT/200x/XP</quote></title>

<para>
<indexterm significance="normal"><primary>PostScript</primary></indexterm>
CUPS users may get the exact same package from <ulink noescape="1" url="http://www.cups.org/software.html">http://www.cups.org/software.html</ulink>.  It is a separate package
from the CUPS-based software files, tagged as CUPS 1.1.x Windows NT/200x/XP Printer Driver for Samba (tar.gz,
192k). The filename to download is <filename moreinfo="none">cups-samba-1.1.x.tar.gz</filename>. Upon untar and unzipping, it
will reveal these files:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">tar xvzf cups-samba-1.1.19.tar.gz</userinput>
cups-samba.install
cups-samba.license
cups-samba.readme
cups-samba.remove
cups-samba.ss
</screen></para>

<para>
<indexterm significance="normal"><primary>ESP</primary><secondary>meta packager</secondary></indexterm>
<indexterm significance="normal"><primary>EPM</primary><see>ESP meta packager</see></indexterm>
These have been packaged with the ESP meta-packager software EPM. The <filename moreinfo="none">*.install</filename> and
<filename moreinfo="none">*.remove</filename> files are simple shell scripts, which untar the <filename moreinfo="none">*.ss</filename> (the
<filename moreinfo="none">*.ss</filename> is nothing else but a tar archive, which can be untarred by <quote>tar</quote> too).
Then it puts the content into <filename moreinfo="none">/usr/share/cups/drivers/</filename>. This content includes three
files:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">tar tv cups-samba.ss</userinput>
cupsdrvr.dll
cupsui.dll
cups.hlp  
</screen></para>

<para>
The <parameter moreinfo="none">cups-samba.install</parameter> shell scripts are easy to
handle:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">./cups-samba.install</userinput>
[....]
Installing software...
Updating file permissions...
Running post-install commands...
Installation is complete.       
</screen></para>

<para>
The script should automatically put the driver files into the
<filename moreinfo="none">/usr/share/cups/drivers/</filename> directory:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">cp /usr/share/drivers/cups.hlp /usr/share/cups/drivers/</userinput>
</screen></para>

<warning><para>
Due to a bug, one recent CUPS release puts the <filename moreinfo="none">cups.hlp</filename> driver file
into<filename moreinfo="none">/usr/share/drivers/</filename> instead of <filename moreinfo="none">/usr/share/cups/drivers/</filename>. To work
around this, copy/move the file (after running the <literal>./cups-samba.install</literal> script) manually to
the correct place.
</para></warning>

<para>
<indexterm significance="normal"><primary>DDK</primary></indexterm>
This new CUPS PostScript driver is currently binary only, but free of charge. No complete source code is
provided (yet). The reason is that it has been developed with the help of the Microsoft DDK and compiled with
Microsoft Visual Studio 6. Driver developers are not allowed to distribute the whole of the source code as
free software. However, CUPS developers released the <quote>diff</quote> in source code under the GPL, so
anybody with a license for Visual Studio and a DDK will be able to compile for himself or herself.
</para>
</sect2>

<sect2>
<title>Recognizing Different Driver Files</title>

<para>
The CUPS drivers do not support the older Windows 95/98/Me, but only the Windows NT/2000/XP client.
</para>

<para>Windows NT, 2000, and XP are supported by:</para>

<itemizedlist>
	<listitem><para>cups.hlp</para></listitem>
	<listitem><para>cupsdrvr.dll</para></listitem>
	<listitem><para>cupsui.dll</para></listitem>
</itemizedlist>

<para>
Adobe drivers are available for the older Windows 95/98/Me as well as
for Windows NT/2000/XP clients. The set of files is different from the
different platforms.
</para>

<para>Windows 95, 98, and ME are supported by:</para>

<itemizedlist>
	<listitem><para>ADFONTS.MFM</para></listitem>
	<listitem><para>ADOBEPS4.DRV</para></listitem>
	<listitem><para>ADOBEPS4.HLP</para></listitem>
	<listitem><para>DEFPRTR2.PPD</para></listitem>
	<listitem><para>ICONLIB.DLL</para></listitem>
	<listitem><para>PSMON.DLL</para></listitem>
</itemizedlist>

<para>Windows NT, 2000, and XP are supported by:</para>

<itemizedlist>
	<listitem><para>ADOBEPS5.DLL</para></listitem>
	<listitem><para>ADOBEPSU.DLL</para></listitem>
	<listitem><para>ADOBEPSU.HLP</para></listitem>
</itemizedlist>

<note><para>
<indexterm significance="normal"><primary>Adobe driver files</primary></indexterm>
If both the Adobe driver files and the CUPS driver files for the support of Windows NT/200x/XP are presently
installed on the server, the Adobe files will be ignored and the CUPS files will be used. If you prefer
<?latex --- ?> for whatever reason <?latex --- ?> to use Adobe-only drivers, move away the three CUPS driver files.
The Windows 9x/Me clients use the Adobe drivers in any case.
</para></note>
</sect2>

<sect2>
<title>Acquiring the Adobe Driver Files</title>

<para>
Acquiring the Adobe driver files seems to be unexpectedly difficult for many users. They are not available on
the Adobe Web site as single files, and the self-extracting and/or self-installing Windows-.exe is not easy to
locate either. You probably need to use the included native installer and run the installation process on one
client once. This will install the drivers (and one generic PostScript printer) locally on the client. When
they are installed, share the generic PostScript printer. After this, the client's <parameter>[print$]</parameter> share holds the Adobe files, which you can get with smbclient from the CUPS host.
</para>
</sect2>

<sect2>
<title>ESP Print Pro PostScript Driver for Windows NT/200x/XP</title>

<para>
<indexterm significance="normal"><primary>ESP</primary><secondary>Print Pro</secondary></indexterm>
Users of the ESP Print Pro software are able to install the ESP print drivers package as an alternative to the
Adobe PostScript drivers.  To do so, retrieve the driver files from the normal download area of the ESP Print
Pro software at <ulink noescape="1" url="http://www.easysw.com/software.html">Easy Software</ulink> web site.
You need to locate the link labeled <quote>SAMBA</quote> among the <guilabel moreinfo="none">Download Printer Drivers for ESP
Print Pro 4.x</guilabel> area and download the package. Once installed, you can prepare any driver by simply
highlighting the printer in the Printer Manager GUI and selecting <guilabel moreinfo="none">Export Driver...</guilabel> from
the menu. Of course, you need to have prepared Samba beforehand to handle the driver files; that is, set up
the <parameter>[print$]</parameter> share, and so on. The ESP Print Pro package includes the CUPS driver
files as well as a (licensed) set of Adobe drivers for the Windows 95/98/Me client family.
</para>
</sect2>

<sect2>
<title>Caveats to Be Considered</title>


<para>
<indexterm significance="normal"><primary>cupsaddsmb</primary></indexterm>
<indexterm significance="normal"><primary>cups.hlp</primary></indexterm>
<indexterm significance="normal"><primary>WIN40</primary></indexterm>
<indexterm significance="normal"><primary>W32X86</primary></indexterm>
Once you have run the install script (and possibly manually moved the <filename moreinfo="none">cups.hlp</filename> file to
<filename moreinfo="none">/usr/share/cups/drivers/</filename>), the driver is ready to be put into Samba's <parameter>[print$]</parameter> share (which often maps to <filename moreinfo="none">/etc/samba/drivers/</filename> and contains a
subdirectory tree with <emphasis>WIN40</emphasis> and <emphasis>W32X86</emphasis> branches). You do this by
running <literal>cupsaddsmb</literal> (see also <literal>man cupsaddsmb</literal> for CUPS since release
1.1.16).
</para>

<tip><para>
<indexterm significance="normal"><primary>Single Sign-On</primary></indexterm>
<indexterm significance="normal"><primary>Domain Controller</primary></indexterm>
You may need to put root into the smbpasswd file by running <literal>smbpasswd</literal>; this is especially
important if you should run this whole procedure for the first time and are not working in an environment
where everything is configured for <emphasis>single sign-on</emphasis> to a Windows Domain Controller.
</para></tip>

<para>
Once the driver files are in the <parameter>[print$]</parameter> share and are initialized, they are ready
to be downloaded and installed by the Windows NT/200x/XP clients.
</para>

<note><para>
Win 9x/Me clients will not work with the CUPS PostScript driver. For these you still need to use the
<filename moreinfo="none">ADOBE*.*</filename> drivers, as previously stated.
</para></note>

<note>
<para>
It is not harmful if you still have the <filename moreinfo="none">ADOBE*.*</filename> driver files from previous installations
in the <filename moreinfo="none">/usr/share/cups/drivers/</filename> directory. The new <literal>cupsaddsmb</literal> (from
1.1.16) will automatically prefer its own drivers if it finds both.
</para></note>

<note><para>
<indexterm significance="normal"><primary>"Printers" folder</primary></indexterm>
<indexterm significance="normal"><primary>Adobe PostScript</primary></indexterm>
Should your Windows clients have had the old <filename moreinfo="none">ADOBE*.*</filename> files for the Adobe PostScript
driver installed, the download and installation of the new CUPS PostScript driver for Windows NT/200x/XP will
fail at first. You need to wipe the old driver from the clients first. It is not enough to
<quote>delete</quote> the printer, because the driver files will still be kept by the clients and re-used if
you try to re-install the printer. To really get rid of the Adobe driver files on the clients, open the
<guilabel moreinfo="none">Printers</guilabel> folder (possibly via <guilabel moreinfo="none">Start -&gt; Settings -&gt; Control Panel -&gt;
Printers</guilabel>), right-click on the folder background, and select <guimenuitem moreinfo="none">Server
Properties</guimenuitem>. When the new dialog opens, select the <guilabel moreinfo="none">Drivers</guilabel> tab. On the list
select the driver you want to delete and click the <guilabel moreinfo="none">Delete</guilabel> button. This will only work if
there is not one single printer left that uses that particular driver. You need to <quote>delete</quote> all
printers using this driver in the <guilabel moreinfo="none">Printers</guilabel> folder first. You will need Administrator
privileges to do this.
</para></note>

<note><para>
<indexterm significance="normal"><primary>rpcclient</primary><secondary>setdriver</secondary></indexterm>
<indexterm significance="normal"><primary>CUPS PostScript</primary></indexterm>
Once you have successfully downloaded the CUPS PostScript driver to a client, you can easily switch all
printers to this one by proceeding as described in <link linkend="classicalprinting">Classical Printing
Support</link>. Either change a driver for an existing printer by running the <guilabel moreinfo="none">Printer
Properties</guilabel> dialog, or use <literal>rpcclient</literal> with the <literal>setdriver</literal>
subcommand.
</para></note>
</sect2>

<sect2>
<title>Windows CUPS PostScript Driver Versus Adobe Driver</title>

<para>
Are you interested in a comparison between the CUPS and the Adobe PostScript drivers? For our purposes, these
are the most important items that weigh in favor of CUPS:
</para>

<itemizedlist>
	<listitem><para>No hassle with the Adobe EULA.</para></listitem>

	<listitem><para>No hassle with the question, <quote>Where do I
	get the ADOBE*.* driver files?</quote></para></listitem>

	<listitem><para>
	<indexterm significance="normal"><primary>PJL</primary></indexterm>
	The Adobe drivers (on request of the printer PPD associated with them) often put a PJL header in front of the
	main PostScript part of the print file. Thus, the print file starts with <parameter moreinfo="none">&lt;1B
	&gt;%-12345X</parameter> or <parameter moreinfo="none">&lt;escape&gt;%-12345X</parameter> instead of
	<parameter moreinfo="none">%!PS</parameter>. This leads to the CUPS daemon autotyping the incoming file as a print-ready file,
	not initiating a pass through the <parameter moreinfo="none">pstops</parameter> filter (to speak more technically, it is not
	regarded as the generic MIME-type <indexterm significance="normal"><primary>application/postscript</primary></indexterm>
	<parameter moreinfo="none">application/postscript</parameter>, but as the more special MIME type
	<indexterm significance="normal"><primary>application/cups.vnd-postscript</primary></indexterm>
	<parameter moreinfo="none">application/cups.vnd-postscript</parameter>), which therefore also leads to the page accounting in
	<parameter moreinfo="none">/var/log/cups/page_log</parameter> not receiving the exact number of pages; instead the dummy page
	number of <quote>1</quote> is logged in a standard setup).
	</para></listitem>

	<listitem><para>The Adobe driver has more options to misconfigure the
<indexterm significance="normal"><primary>Adobe driver</primary></indexterm>
	PostScript generated by it (like setting it inadvertently to
	<guilabel moreinfo="none">Optimize for Speed</guilabel> instead of
	<guilabel moreinfo="none">Optimize for Portability</guilabel>, which
	could lead to CUPS being unable to process it).</para></listitem>

	<listitem><para>The CUPS PostScript driver output sent by Windows
<indexterm significance="normal"><primary>CUPS PostScript driver</primary></indexterm>
	clients to the CUPS server is guaranteed to autotype 
	as the generic MIME type <parameter moreinfo="none">application/postscript</parameter>,
	thus passing through the CUPS <parameter moreinfo="none">pstops</parameter> filter and logging the
	correct number of pages in the <filename moreinfo="none">page_log</filename> for
	accounting and quota purposes.</para></listitem>

	<listitem><para>
	<indexterm significance="normal"><primary>banner pages</primary></indexterm>
	The CUPS PostScript driver supports the sending of additional standard (IPP) print options by Windows
	NT/200x/XP clients. Such additional print options are naming the CUPS standard <emphasis>banner
	pages</emphasis> (or the custom ones, should they be installed at the time of driver download), using the CUPS
	page-label option, setting a job priority, and setting the scheduled time of printing (with the option to
	support additional useful IPP job attributes in the future).
	</para></listitem>

	<listitem><para>The CUPS PostScript driver supports the inclusion of
	the new <parameter moreinfo="none">*cupsJobTicket</parameter> comments at the
	beginning of the PostScript file (which could be used in the future
	for all sorts of beneficial extensions on the CUPS side, but which will
	not disturb any other applications because they will regard it as a comment
	and simply ignore it).</para></listitem>

	<listitem><para>The CUPS PostScript driver will be the heart of the
	fully fledged CUPS IPP client for Windows NT/200x/XP to be released soon
	(probably alongside the first beta release for CUPS 1.2).</para></listitem>
</itemizedlist>

</sect2>

<sect2>
<title>Run cupsaddsmb (Quiet Mode)</title>


<para>
<indexterm significance="normal"><primary>cupsaddsmb</primary></indexterm>
<indexterm significance="normal"><primary>point 'n' print</primary></indexterm>
The <literal>cupsaddsmb</literal> command copies the needed files into your <parameter>[print$]</parameter>
share. Additionally, the PPD associated with this printer is copied from <filename moreinfo="none">/etc/cups/ppd/</filename>
to <parameter>[print$]</parameter>. There the files wait for convenient Windows client installations via
Point'n'Print. Before we can run the command successfully, we need to be sure that we can authenticate toward
Samba. If you have a small network, you are probably using user-level security (<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SECURITY" xlink:href="smb.conf.5.html#SECURITY">security = user</link>).
</para>

<para>
Here is an example of a successfully run <literal>cupsaddsmb</literal> command: 
<indexterm significance="normal"><primary>banner pages</primary></indexterm>
<indexterm significance="normal"><primary>cupsaddsmb</primary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">cupsaddsmb -U root infotec_IS2027</userinput>
Password for root required to access localhost via Samba: <userinput moreinfo="none">['secret']</userinput>
</screen></para>

<para>
<indexterm significance="normal"><primary>cupsaddsmb</primary></indexterm>
To share <emphasis>all</emphasis> printers and drivers, use the
<option>-a</option> parameter instead of a printer name. Since
<literal>cupsaddsmb</literal> <quote>exports</quote> the printer drivers to Samba, it should be
obvious that it only works for queues with a CUPS driver associated.
</para>
</sect2>

<sect2>
<title>Run cupsaddsmb with Verbose Output</title>


<para>
<indexterm significance="normal"><primary>cupsaddsmb</primary></indexterm>
Probably you want to see what's going on. Use the
<option>-v</option> parameter to get a more verbose output. The
output below was edited for better readability: all <quote>\</quote> at the end of
a line indicate that I inserted an artificial line break plus some
indentation here:
<indexterm significance="normal"><primary>rpcclient</primary><secondary>adddriver</secondary></indexterm>
<indexterm significance="normal"><primary>rpcclient</primary><secondary>setdriver</secondary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">cupsaddsmb -U root -v infotec_2105</userinput>
Password for root required to access localhost via GANDALF:
Running command: smbclient //localhost/print\$ -N -U'root%secret' \
    -c 'mkdir W32X86; \
    put /var/spool/cups/tmp/3e98bf2d333b5 W32X86/infotec_2105.ppd; \
	put /usr/share/cups/drivers/cupsdrvr.dll W32X86/cupsdrvr.dll; \
    put /usr/share/cups/drivers/cupsui.dll W32X86/cupsui.dll; \
    put /usr/share/cups/drivers/cups.hlp W32X86/cups.hlp'
added interface ip=10.160.51.60 bcast=10.160.51.255 nmask=255.255.252.0
Domain=[CUPS-PRINT] OS=[UNIX] Server=[Samba 2.2.7a]
NT_STATUS_OBJECT_NAME_COLLISION making remote directory \W32X86
putting file /var/spool/cups/tmp/3e98bf2d333b5 as \W32X86/infotec_2105.ppd
putting file /usr/share/cups/drivers/cupsdrvr.dll as \W32X86/cupsdrvr.dll
putting file /usr/share/cups/drivers/cupsui.dll as \W32X86/cupsui.dll
putting file /usr/share/cups/drivers/cups.hlp as \W32X86/cups.hlp
  
Running command: rpcclient localhost -N -U'root%secret' 
   -c 'adddriver "Windows NT x86"   \
   "infotec_2105:cupsdrvr.dll:infotec_2105.ppd:cupsui.dll:cups.hlp:NULL: \
    RAW:NULL"'
cmd = adddriver "Windows NT x86" \
   "infotec_2105:cupsdrvr.dll:infotec_2105.ppd:cupsui.dll:cups.hlp:NULL: \
	RAW:NULL"
Printer Driver infotec_2105 successfully installed.
  
Running command: smbclient //localhost/print\$ -N -U'root%secret' \
-c 'mkdir WIN40; \
    put /var/spool/cups/tmp/3e98bf2d333b5 WIN40/infotec_2105.PPD; \
	put /usr/share/cups/drivers/ADFONTS.MFM WIN40/ADFONTS.MFM;   \
    put /usr/share/cups/drivers/ADOBEPS4.DRV WIN40/ADOBEPS4.DRV; \
    put /usr/share/cups/drivers/ADOBEPS4.HLP WIN40/ADOBEPS4.HLP; \
    put /usr/share/cups/drivers/DEFPRTR2.PPD WIN40/DEFPRTR2.PPD; \
	put /usr/share/cups/drivers/ICONLIB.DLL WIN40/ICONLIB.DLL; \
	put /usr/share/cups/drivers/PSMON.DLL WIN40/PSMON.DLL;'
  added interface ip=10.160.51.60 bcast=10.160.51.255 nmask=255.255.252.0
  Domain=[CUPS-PRINT] OS=[UNIX] Server=[Samba 2.2.7a]
  NT_STATUS_OBJECT_NAME_COLLISION making remote directory \WIN40
  putting file /var/spool/cups/tmp/3e98bf2d333b5 as \WIN40/infotec_2105.PPD
  putting file /usr/share/cups/drivers/ADFONTS.MFM as \WIN40/ADFONTS.MFM
  putting file /usr/share/cups/drivers/ADOBEPS4.DRV as \WIN40/ADOBEPS4.DRV
  putting file /usr/share/cups/drivers/ADOBEPS4.HLP as \WIN40/ADOBEPS4.HLP
  putting file /usr/share/cups/drivers/DEFPRTR2.PPD as \WIN40/DEFPRTR2.PPD
  putting file /usr/share/cups/drivers/ICONLIB.DLL as \WIN40/ICONLIB.DLL
  putting file /usr/share/cups/drivers/PSMON.DLL as \WIN40/PSMON.DLL
  
  Running command: rpcclient localhost -N -U'root%secret' \
   -c 'adddriver "Windows 4.0"      \
   "infotec_2105:ADOBEPS4.DRV:infotec_2105.PPD:NULL:ADOBEPS4.HLP: \
   PSMON.DLL:RAW:ADOBEPS4.DRV,infotec_2105.PPD,ADOBEPS4.HLP,PSMON.DLL, \
    ADFONTS.MFM,DEFPRTR2.PPD,ICONLIB.DLL"'
	cmd = adddriver "Windows 4.0" "infotec_2105:ADOBEPS4.DRV:\
	infotec_2105.PPD:NULL:ADOBEPS4.HLP:PSMON.DLL:RAW:ADOBEPS4.DRV,\
	infotec_2105.PPD,ADOBEPS4.HLP,PSMON.DLL,ADFONTS.MFM,DEFPRTR2.PPD,\
	ICONLIB.DLL"
  Printer Driver infotec_2105 successfully installed.
  
  Running command: rpcclient localhost -N -U'root%secret'  \
   -c 'setdriver infotec_2105 infotec_2105'
  cmd = setdriver infotec_2105 infotec_2105
  Successfully set infotec_2105 to driver infotec_2105.
</screen></para>

<warning><para>
You will see the root password for the Samba account printed on screen. 
</para></warning>

<para>
If you look closely, you'll discover your root password was transferred unencrypted over the wire, so beware!
Also, if you look further, you may discover error messages like NT_STATUS_OBJECT_NAME_COLLISION in the output.
This will occur when the directories WIN40 and W32X86 already existed in the <parameter>[print$]</parameter>
driver download share (from a previous driver installation). These are harmless warning messages.
</para>
</sect2>

<sect2>
<title>Understanding cupsaddsmb</title>

<para>
<indexterm significance="normal"><primary>cupsaddsmb</primary></indexterm>
What has happened? What did <literal>cupsaddsmb</literal> do? There are five stages of the procedure:
</para>

<orderedlist inheritnum="ignore" continuation="restarts">
	<listitem><para>
	<indexterm significance="normal"><primary>IPP</primary></indexterm>
	Call the CUPS server via IPP and request the driver files and the PPD file for the named printer.</para></listitem>

	<listitem><para>Store the files temporarily in the local TEMPDIR (as defined in <filename moreinfo="none">cupsd.conf</filename>).</para></listitem>

	<listitem><para>Connect via smbclient to the Samba server's <parameter>[print$]</parameter> share and put the files into the
	 share's WIN40 (for Windows 9x/Me) and W32X86 (for Windows NT/200x/XP) subdirectories.</para></listitem>

	<listitem><para>
	<indexterm significance="normal"><primary>rpcclient</primary><secondary>adddriver</secondary></indexterm>
	Connect via rpcclient to the Samba server and execute the <literal>adddriver</literal> command with the correct parameters.
	</para></listitem>

	<listitem><para>
	<indexterm significance="normal"><primary>rpcclient</primary><secondary>setdriver</secondary></indexterm>
	Connect via rpcclient to the Samba server a second time and execute the <literal>setdriver</literal> command.</para></listitem>
</orderedlist>

<note>
<para>
You can run the <literal>cupsaddsmb</literal> utility with parameters to specify one remote host as Samba host
and a second remote host as CUPS host. Especially if you want to get a deeper understanding, it is a good idea
to try it and see more clearly what is going on (though in real life most people will have their CUPS and
Samba servers run on the same host):
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">cupsaddsmb -H sambaserver -h cupsserver -v printer</userinput>
</screen>
</para></note>

</sect2>

<sect2>
<title>How to Recognize If cupsaddsmb Completed Successfully</title>

<para>
You <emphasis>must</emphasis> always check if the utility completed
successfully in all fields. You need at minimum these three messages
among the output:
</para>

<orderedlist inheritnum="ignore" continuation="restarts">
	<listitem><para><emphasis>Printer Driver infotec_2105 successfully
	installed.</emphasis> # (for the W32X86 == Windows NT/200x/XP
	architecture).</para></listitem>

	<listitem><para><emphasis>Printer Driver infotec_2105 successfully
	installed.</emphasis> # (for the WIN40 == Windows 9x/Me
	architecture).</para></listitem>

	<listitem><para><emphasis>Successfully set [printerXPZ] to driver
	[printerXYZ].</emphasis></para></listitem>
</orderedlist>

<para>
These messages are probably not easily recognized in the general
output. If you run <literal>cupsaddsmb</literal> with the <option>-a</option>
parameter (which tries to prepare <emphasis>all</emphasis> active CUPS
printer drivers for download), you might miss if individual printer
drivers had problems installing properly. A redirection of the
output will help you analyze the results in retrospective.
</para>

<para>
If you get:
<screen format="linespecific">
SetPrinter call failed!
result was WERR_ACCESS_DENIED
</screen>
it means that you might have set <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="USECLIENTDRIVER" xlink:href="smb.conf.5.html#USECLIENTDRIVER">use client driver = yes</link> for this printer. 
Setting it to <quote>no</quote> will solve the problem. Refer to the <filename moreinfo="none">smb.conf</filename> man page for explanation of 
the <parameter moreinfo="none">use client driver</parameter>.
</para>

<note><para>
It is impossible to see any diagnostic output if you do not run <literal>cupsaddsmb</literal> in verbose mode.
Therefore, we strongly recommend against use of the default quiet mode. It will hide any problems from you that
might occur.
</para></note>
</sect2>

<sect2>
<title>cupsaddsmb with a Samba PDC</title>

<para>
<indexterm significance="normal"><primary>cupsaddsmb</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
Can't get the standard <literal>cupsaddsmb</literal> command to run on a Samba PDC?  Are you asked for the
password credential again and again, and the command just will not take off at all? Try one of these
variations:
</para>

<para><screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">cupsaddsmb -U MIDEARTH\\root -v printername</userinput>
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">cupsaddsmb -H SAURON -U MIDEARTH\\root -v printername</userinput>
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">cupsaddsmb -H SAURON -U MIDEARTH\\root -h cups-server -v printername</userinput>
</screen></para>

<para>
(Note the two backslashes: the first one is required to <quote>escape</quote> the second one).
</para>
</sect2>

<sect2>
<title>cupsaddsmb Flowchart</title>

<para>
<indexterm significance="normal"><primary>cupsaddsmb</primary></indexterm>
<indexterm significance="normal"><primary>raw print</primary></indexterm>
<link linkend="small14">The cupsaddsmb Flowchart</link> shows a chart about the procedures, command flows, and
data flows of the <literal>cupaddsmb</literal> command. Note again: cupsaddsmb is
not intended to, and does not work with, raw print queues!
</para>

	<figure id="small14" float="0">
		<title>cupsaddsmb Flowchart.</title>
		<mediaobject><imageobject role="html"><imagedata fileref="images/14small.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/14small.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/14small"/></imageobject></mediaobject></figure>
</sect2>

<sect2>
<title>Installing the PostScript Driver on a Client</title>

<para>
<indexterm significance="normal"><primary>point'n'print</primary></indexterm>
<indexterm significance="normal"><primary>cupsaddsmb</primary></indexterm>
After <literal>cupsaddsmb</literal> is completed, your driver is prepared for the clients to use. Here are the
steps you must perform to download and install it via Point'n'Print. From a Windows client, browse to the
CUPS/Samba server:
</para>

<itemizedlist>

	<listitem><para>
	<indexterm significance="normal"><primary>"Printers" folder</primary></indexterm>
	Open the <guilabel moreinfo="none">Printers</guilabel> share of Samba in Network Neighborhood.</para></listitem>

	<listitem><para>Right-click on the printer in question.</para></listitem>

	<listitem><para>From the opening context menu select
	<guimenuitem moreinfo="none">Install...</guimenuitem> or 
	<guimenuitem moreinfo="none">Connect...</guimenuitem> (depending on the Windows version you use).</para></listitem>
</itemizedlist>

<para>
After a few seconds, there should be a new printer in your client's <emphasis>local</emphasis>
<guilabel moreinfo="none">Printers</guilabel> folder. On Windows XP it will follow a naming convention of
<emphasis>PrinterName on SambaServer</emphasis>. (In my current case it is infotec_2105 on kde-bitshop). If
you want to test it and send your first job from an application like Microsoft Word,
the new printer appears in a
<filename moreinfo="none">\\SambaServer\PrinterName</filename> entry in the drop-down list of available printers.
</para>

<para>
<indexterm significance="normal"><primary>PPD</primary></indexterm>
<indexterm significance="normal"><primary>Adobe PostScript driver</primary></indexterm>
<indexterm significance="normal"><primary>net use lpt1:</primary></indexterm>
<literal>cupsaddsmb</literal> will only reliably work with CUPS version 1.1.15 or higher and with Samba
version 2.2.4, or later. If it does not work, or if the automatic printer driver download to the clients does
not succeed, you can still manually install the CUPS printer PPD on top of the Adobe PostScript driver on
clients. Then point the client's printer queue to the Samba printer share for a UNC type of connection:
<screen format="linespecific">
<prompt moreinfo="none">C:\&gt; </prompt><userinput moreinfo="none">net use lpt1: \\sambaserver\printershare /user:ntadmin</userinput>
</screen>
should you desire to use the CUPS networked PostScript RIP functions. (Note that user <quote>ntadmin</quote>
needs to be a valid Samba user with the required privileges to access the printershare.) This sets up the
printer connection in the traditional LanMan way (not using MS-RPC).
</para>
</sect2>

<sect2 id="cups-avoidps1">
<title>Avoiding Critical PostScript Driver Settings on the Client</title>

<para>
Printing works, but there are still problems. Most jobs print well, some do not print at all. Some jobs have
problems with fonts, which do not look very good. Some jobs print fast and some are dead-slow. Many of these
problems can be greatly reduced or even completely eliminated if you follow a few guidelines. Remember, if
your print device is not PostScript-enabled, you are treating your Ghostscript installation on your CUPS host
with the output your client driver settings produce. Treat it well:
</para>

<itemizedlist>
	<listitem><para>
	Avoid the PostScript Output Option: Optimize for Speed setting. Use the Optimize for Portability instead
	(Adobe PostScript driver).</para></listitem>

	<listitem><para>
	Don't use the Page Independence: NO setting. Instead, use Page Independence: YES (CUPS PostScript Driver).
	</para></listitem>

	<listitem><para>
	Recommended is the True Type Font Downloading Option: Native True Type over Automatic and Outline; 
	you should by all means avoid Bitmap (Adobe PostScript Driver).</para></listitem>

	<listitem><para>
	Choose True Type Font: Download as Softfont into Printer over the default Replace by Device
	Font (for exotic fonts, you may need to change it back to get a printout at all; Adobe).</para></listitem>

	<listitem><para>
	Sometimes you can choose PostScript Language Level: in case of problems try 2
	instead of 3 (the latest ESP Ghostscript package handles Level 3 PostScript very well; Adobe).
	</para></listitem>

	<listitem><para>
	Say Yes to PostScript Error Handler (Adobe).</para></listitem>
</itemizedlist>

</sect2>
</sect1>

<sect1>
<title>Installing PostScript Driver Files Manually Using rpcclient</title>

<para>
Of course, you can run all the commands that are embedded into the
cupsaddsmb convenience utility yourself, one by one, and upload
and prepare the driver files for future client downloads.
</para>

<orderedlist inheritnum="ignore" continuation="restarts">
	<listitem><para>Prepare Samba (a CUPS print queue with the name of the
	printer should be there. We are providing the driver now).</para></listitem>

	<listitem><para>Copy all files to <parameter>[print$]</parameter>.</para></listitem>

	<listitem><para>
	<indexterm significance="normal"><primary>rpcclient</primary><secondary>adddriver</secondary></indexterm>
	Run <literal>rpcclient adddriver</literal>
	(for each client architecture you want to support).</para></listitem>

	<listitem><para>
	<indexterm significance="normal"><primary>rpcclient</primary><secondary>setdriver</secondary></indexterm>
	Run <literal>rpcclient setdriver.</literal></para></listitem>
</orderedlist>

<para>
<indexterm significance="normal"><primary>rpcclient</primary><secondary>enumports</secondary></indexterm>
<indexterm significance="normal"><primary>rpcclient</primary><secondary>enumprinters</secondary></indexterm>
<indexterm significance="normal"><primary>rpcclient</primary><secondary>enumdrivers</secondary></indexterm>
<indexterm significance="normal"><primary>rpcclient</primary><secondary>setdriver</secondary></indexterm>
<indexterm significance="normal"><primary>rpcclient</primary><secondary>adddriver</secondary></indexterm>
We are going to do this now. First, read the man page on <parameter moreinfo="none">rpcclient</parameter> to get a first idea.
Look at all the printing-related subcommands: <literal>enumprinters</literal>, <literal>enumdrivers</literal>,
<literal>enumports</literal>, <literal>adddriver</literal>, and <literal>setdriver</literal> are among the
most interesting ones. <parameter moreinfo="none">rpcclient</parameter> implements an important part of the MS-RPC protocol.
You can use it to query (and command) a Windows NT (or 200x/XP) PC, too. MS-RPC is used by Windows clients,
among other things, to benefit from the Point'n'Print features. Samba can now mimic this as well.
</para>

<sect2>
<title>A Check of the rpcclient man Page</title>

<para>
First let's check the <parameter moreinfo="none">rpcclient</parameter> man page. Here are two relevant passages:
</para>

<para>
<indexterm significance="normal"><primary>adddriver</primary></indexterm>
<indexterm significance="normal"><primary>AddPrinterDriver()</primary></indexterm>
<indexterm significance="normal"><primary>getdriverdir</primary></indexterm>
<literal>adddriver &lt;arch&gt; &lt;config&gt;</literal> Execute an <literal>AddPrinterDriver()</literal> RPC
to install the printer driver information on the server. The driver files should already exist in the
directory returned by <literal>getdriverdir</literal>. Possible values for <parameter moreinfo="none">arch</parameter> are the
same as those for the <literal>getdriverdir</literal> command. The <parameter moreinfo="none">config</parameter> parameter is
defined as follows:
<screen format="linespecific">
Long Printer Name:\
Driver File Name:\
Data File Name:\
Config File Name:\
Help File Name:\
Language Monitor Name:\
Default Data Type:\
Comma Separated list of Files
</screen></para>

<para>
Any empty fields should be entered as the string <quote>NULL</quote>. 
</para>

<para>
Samba does not need to support the concept of print monitors, since these only apply to local printers whose
drivers can use a bidirectional link for communication. This field should be <quote>NULL</quote>.  On a remote
NT print server, the print monitor for a driver must already be installed before adding the driver or else the
RPC will fail.
</para>

<para>
<indexterm significance="normal"><primary>setdriver</primary></indexterm>
<indexterm significance="normal"><primary>SetPrinter()</primary></indexterm>
<literal>setdriver &lt;printername&gt; &lt;drivername&gt;</literal> Execute a <literal>SetPrinter()</literal>
command to update the printer driver associated with an installed printer. The printer driver must already be
correctly installed on the print server.
</para>

<para>
<indexterm significance="normal"><primary>enumprinters</primary></indexterm>
<indexterm significance="normal"><primary>enumdrivers</primary></indexterm>
See also the <literal>enumprinters</literal> and <literal>enumdrivers</literal> commands to
obtain a list of installed printers and drivers.
</para>

</sect2>

<sect2>
<title>Understanding the rpcclient man Page</title>

<para>
<indexterm significance="normal"><primary>rpcclient</primary><secondary>adddriver</secondary></indexterm>
The <emphasis>exact</emphasis> format isn't made too clear by the man page, since you have to deal with some
parameters containing spaces. Here is a better description for it. We have line-broken the command and
indicated the breaks with <quote>\</quote>. Usually you would type the command in one line without the line
breaks:
<screen format="linespecific">
adddriver "Architecture" \
   "LongPrinterName:DriverFile:DataFile:ConfigFile:HelpFile:\
   LanguageMonitorFile:DataType:ListOfFiles,Comma-separated"
</screen></para>

<para>
What the man pages denote as a simple <parameter moreinfo="none">&lt;config&gt;</parameter> keyword in reality consists of
eight colon-separated fields. The last field may take multiple (in some very insane cases, even 20 different
additional) files. This might sound confusing at first.  What the man pages call the
<quote>LongPrinterName</quote> in reality should be called the <quote>Driver Name</quote>. You can name it
anything you want, as long as you use this name later in the <literal>rpcclient ... setdriver</literal>
command. For practical reasons, many name the driver the same as the printer.
</para>

<para>
It isn't simple at all. I hear you asking: <quote>How do I know which files are Driver File</quote>,
<quote>Data File</quote>, <quote>Config File</quote>, <quote>Help File</quote> and <quote>Language Monitor
File in each case?</quote> For an answer, you may want to have a look at how a Windows NT box with a shared
printer presents the files to us. Remember that this whole procedure has to be developed by the Samba Team by
listening to the traffic caused by Windows computers on the wire. We may as well turn to a Windows box now and
access it from a UNIX workstation. We will query it with <literal>rpcclient</literal> to see what it tells us
and try to understand the man page more clearly.
</para>
</sect2>

<sect2>
<title>Producing an Example by Querying a Windows Box</title>

<para>
<indexterm significance="normal"><primary>rpcclient</primary><secondary>getdriver</secondary></indexterm>
<indexterm significance="normal"><primary>rpcclient</primary><secondary>getprinter</secondary></indexterm>
We could run <literal>rpcclient</literal> with a <literal>getdriver</literal> or a
<literal>getprinter</literal> subcommand (in level 3 verbosity) against it. Just sit down at a UNIX or Linux
workstation with the Samba utilities installed, then type the following command:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">rpcclient -U'user%secret' NT-SERVER -c 'getdriver printername 3'</userinput>
</screen></para>

<para>
From the result it should become clear which is which. Here is an example from my installation:
<indexterm significance="normal"><primary>rpcclient</primary><secondary>getdriver</secondary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">rpcclient -U'Danka%xxxx' W200xSERVER \
    -c'getdriver "DANKA InfoStream Virtual Printer" 3'</userinput>
    cmd = getdriver "DANKA InfoStream Virtual Printer" 3

 [Windows NT x86]
 Printer Driver Info 3:
         Version: [2]
         Driver Name: [DANKA InfoStream]
         Architecture: [Windows NT x86]
         Driver Path: [C:\WINNT\System32\spool\DRIVERS\W32X86\2\PSCRIPT.DLL]
         Datafile: [C:\WINNT\System32\spool\DRIVERS\W32X86\2\INFOSTRM.PPD]
         Configfile: [C:\WINNT\System32\spool\DRIVERS\W32X86\2\PSCRPTUI.DLL]
         Helpfile: [C:\WINNT\System32\spool\DRIVERS\W32X86\2\PSCRIPT.HLP]
 
         Dependentfiles: []
         Dependentfiles: []
         Dependentfiles: []
         Dependentfiles: []
         Dependentfiles: []
         Dependentfiles: []
         Dependentfiles: []
 
         Monitorname: []
         Defaultdatatype: []
</screen></para>

<para>
Some printer drivers list additional files under the label <parameter moreinfo="none">Dependentfiles</parameter>, and these
would go into the last field <parameter moreinfo="none">ListOfFiles,Comma-separated</parameter>. For the CUPS PostScript
drivers, we do not need any (nor would we for the Adobe PostScript driver); therefore, the field will get a
<quote>NULL</quote> entry.
</para>
</sect2>

<sect2>
<title>Requirements for adddriver and setdriver to Succeed</title>

<para>
<indexterm significance="normal"><primary>rpcclient</primary><secondary>adddriver</secondary></indexterm>
<indexterm significance="normal"><primary>cupsaddsmb</primary></indexterm>
<indexterm significance="normal"><primary>setdriver</primary></indexterm>
From the man page (and from the quoted output of <literal>cupsaddsmb</literal> above) it becomes clear that
you need to have certain conditions in order to make the manual uploading and initializing of the driver files
succeed. The two <literal>rpcclient</literal> subcommands (<literal>adddriver</literal> and
<literal>setdriver</literal>) need to encounter the following preconditions to complete successfully:
</para>

<itemizedlist>
	<listitem><para>You are connected as <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTERADMIN" xlink:href="smb.conf.5.html#PRINTERADMIN">printer admin</link> or root (this is
	<emphasis>not</emphasis> the <quote>Printer Operators</quote> group in NT, but the <emphasis>printer
	admin</emphasis> group as defined in the <parameter>[global]</parameter> section of <filename moreinfo="none">smb.conf</filename>).
	</para></listitem>

	<listitem><para>Copy all required driver files to <filename moreinfo="none">\\SAMBA\print$\w32x86</filename> and
	<filename moreinfo="none">\\SAMBA\print$\win40</filename> as appropriate. They will end up in the <quote>0</quote> respective
	<quote>2</quote> subdirectories later. For now, <emphasis>do not</emphasis> put them there; they'll be
	automatically used by the <literal>adddriver</literal> subcommand. (If you use <literal>smbclient</literal> to
	put the driver files into the share, note that you need to escape the <quote>$</quote>: <literal>smbclient
	//sambaserver/print\$ -U root.</literal>)</para></listitem>

	<listitem><para>The user you're connecting as must be able to write to
	the <parameter>[print$]</parameter> share and create
	subdirectories.</para></listitem>

	<listitem><para>The printer you are going to set up for the Windows
	clients needs to be installed in CUPS already.</para></listitem>

	<listitem><para>
	<indexterm significance="normal"><primary>rpcclient</primary><secondary>setdriver</secondary></indexterm>
	<indexterm significance="normal"><primary>rpcclient</primary><secondary>enumprinters</secondary></indexterm>
	The CUPS printer must be known to Samba; otherwise the <literal>setdriver</literal> subcommand fails with an
	NT_STATUS_UNSUCCESSFUL error. To check if the printer is known by Samba, you may use the
	<literal>enumprinters</literal> subcommand to <literal>rpcclient</literal>. A long-standing bug prevented a
	proper update of the printer list until every smbd process had received a SIGHUP or was restarted. Remember
	this in case you've created the CUPS printer just recently and encounter problems: try restarting Samba.
	</para></listitem>
</itemizedlist>
</sect2>

<sect2>
<title>Manual Driver Installation in 15 Steps</title>

<para>
We are going to install a printer driver now by manually executing all
required commands. Because this may seem a rather complicated process at
first, we go through the procedure step by step, explaining every
single action item as it comes up.
</para>

<procedure>
<title>Manual Driver Installation</title>

	<step performance="required">
	<title>Install the printer on CUPS.</title>

	<para><screen format="linespecific">
	<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">lpadmin -p mysmbtstprn -v socket://10.160.51.131:9100 -E \
				-P canonIR85.ppd</userinput>
	</screen></para>

	<para>
	This installs a printer with the name <parameter moreinfo="none">mysmbtstprn</parameter>
	to the CUPS system. The printer is accessed via a socket
	(a.k.a. JetDirect or Direct TCP/IP) connection. You need to be root
	for this step.
	</para>
	</step>

	<step performance="required">
	<title>(Optional.) Check if the printer is recognized by Samba.</title>

	<para>
	<indexterm significance="normal"><primary>rpcclient</primary><secondary>enumprinters</secondary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">rpcclient -Uroot%xxxx -c 'enumprinters' localhost \
  | grep -C2 mysmbtstprn</userinput>
flags:[0x800000]
name:[\\kde-bitshop\mysmbtstprn]
description:[\\kde-bitshop\mysmbtstprn,,mysmbtstprn]
comment:[mysmbtstprn]
</screen>
	</para>

	<para>
	This should show the printer in the list. If not, stop and restart the Samba daemon (smbd) or send a HUP signal: 
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">kill -HUP `pidof smbd`</userinput>
</screen>
	Check again. Troubleshoot and repeat until successful. Note the <quote>empty</quote> field between the two
	commas in the <quote>description</quote> line. The driver name would appear here if there was one already. You
	need to know root's Samba password (as set by the <literal>smbpasswd</literal> command) for this step and most
	of the following steps. Alternatively, you can authenticate as one of the users from the <quote>write
	list</quote> as defined in <filename moreinfo="none">smb.conf</filename> for <parameter>[print$]</parameter>.
	</para>
	</step>

	<step performance="required">
	<title>(Optional.) Check if Samba knows a driver for the printer.</title>

	<para>
	<indexterm significance="normal"><primary>rpcclient</primary><secondary>getprinter</secondary></indexterm>
	<indexterm significance="normal"><primary>rpcclient</primary><secondary>getdriver</secondary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">rpcclient -Uroot%xxxx -c 'getprinter mysmbtstprn 2'\
 localhost | grep driver </userinput>

drivername:[]

<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">rpcclient -Uroot%xxxx -c 'getprinter mysmbtstprn 2' \
 localhost | grep -C4 driv</userinput>

servername:[\\kde-bitshop]
printername:[\\kde-bitshop\mysmbtstprn]
sharename:[mysmbtstprn]
portname:[Samba Printer Port]
drivername:[]
comment:[mysmbtstprn]
location:[]
sepfile:[]
printprocessor:[winprint]
 
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">rpcclient -U root%xxxx -c 'getdriver mysmbtstprn' localhost</userinput>
 result was WERR_UNKNOWN_PRINTER_DRIVER
</screen></para>

<para>
None of the three commands shown above should show a driver.
This step was done for the purpose of demonstrating this condition. An
attempt to connect to the printer at this stage will prompt a
message along the lines of, <quote>The server does not have the required printer
driver installed.</quote>
</para>
</step>

<step performance="required">
<title>Put all required driver files into Samba's
[print$].</title>

<para><screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">smbclient //localhost/print\$ -U 'root%xxxx' \
	-c 'cd W32X86; \
	put /etc/cups/ppd/mysmbtstprn.ppd mysmbtstprn.PPD; \ 
	put /usr/share/cups/drivers/cupsui.dll cupsui.dll; \
	put /usr/share/cups/drivers/cupsdrvr.dll cupsdrvr.dll; \
	put /usr/share/cups/drivers/cups.hlp cups.hlp'</userinput>
</screen></para>

<para>
(This command should be entered in one long single line. Line breaks and the line ends indicated by
<quote>\</quote> have been inserted for readability reasons.) This step is <emphasis>required</emphasis> for
the next one to succeed. It makes the driver files physically present in the <parameter>[print$]</parameter>
share. However, clients would still not be able to install them, because Samba does not yet treat them as
driver files. A client asking for the driver would still be presented with a <quote>not installed here</quote>
message.
</para>
</step>

<step performance="required">
<title>Verify where the driver files are now.</title>

<para><screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">ls -l /etc/samba/drivers/W32X86/</userinput>
total 669
drwxr-sr-x    2 root     ntadmin       532 May 25 23:08 2
drwxr-sr-x    2 root     ntadmin       670 May 16 03:15 3
-rwxr--r--    1 root     ntadmin     14234 May 25 23:21 cups.hlp
-rwxr--r--    1 root     ntadmin    278380 May 25 23:21 cupsdrvr.dll
-rwxr--r--    1 root     ntadmin    215848 May 25 23:21 cupsui.dll
-rwxr--r--    1 root     ntadmin    169458 May 25 23:21 mysmbtstprn.PPD
</screen></para>

<para>
The driver files now are in the W32X86 architecture <quote>root</quote> of
<parameter>[print$]</parameter>.
</para>
</step>

<step performance="required">
<title>Tell Samba that these are driver files (<literal>adddriver</literal>).</title>

<para>
<indexterm significance="normal"><primary>rpcclient</primary><secondary>adddriver</secondary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">rpcclient -Uroot%xxxx -c 'adddriver "Windows NT x86" \
	"mydrivername:cupsdrvr.dll:mysmbtstprn.PPD: \
  cupsui.dll:cups.hlp:NULL:RAW:NULL"' \
  localhost</userinput>
Printer Driver mydrivername successfully installed.
</screen></para>

<para>
You cannot repeat this step if it fails. It could fail even as a result of a simple typo. It will most likely
have moved a part of the driver files into the <quote>2</quote> subdirectory. If this step fails, you need to
go back to the fourth step and repeat it before you can try this one again. In this step, you need to choose a
name for your driver. It is normally a good idea to use the same name as is used for the printer name;
however, in big installations you may use this driver for a number of printers that obviously have different
names, so the name of the driver is not fixed.
</para>
</step>

<step performance="required">
<title>Verify where the driver files are now.</title>

<para><screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">ls -l /etc/samba/drivers/W32X86/</userinput>
total 1
drwxr-sr-x    2 root     ntadmin       532 May 25 23:22 2
drwxr-sr-x    2 root     ntadmin       670 May 16 03:15 3

<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">ls -l /etc/samba/drivers/W32X86/2</userinput>
total 5039
[....]
-rwxr--r--    1 root     ntadmin     14234 May 25 23:21 cups.hlp
-rwxr--r--    1 root     ntadmin    278380 May 13 13:53 cupsdrvr.dll
-rwxr--r--    1 root     ntadmin    215848 May 13 13:53 cupsui.dll
-rwxr--r--    1 root     ntadmin    169458 May 25 23:21 mysmbtstprn.PPD
</screen></para>

<para>
Notice how step 6 also moved the driver files to the appropriate
subdirectory. Compare this with the situation after step 5.
</para>
</step>

<step performance="required">
<title>(Optional.) Verify if Samba now recognizes the driver.</title>

<para>
<indexterm significance="normal"><primary>rpcclient</primary><secondary>enumdrivers</secondary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">rpcclient -Uroot%xxxx -c 'enumdrivers 3' \
	localhost | grep -B2 -A5 mydrivername</userinput>
Printer Driver Info 3:
Version: [2]
Driver Name: [mydrivername]
Architecture: [Windows NT x86]
Driver Path: [\\kde-bitshop\print$\W32X86\2\cupsdrvr.dll]
Datafile: [\\kde-bitshop\print$\W32X86\2\mysmbtstprn.PPD]
Configfile: [\\kde-bitshop\print$\W32X86\2\cupsui.dll]
Helpfile: [\\kde-bitshop\print$\W32X86\2\cups.hlp]
</screen></para>

<para>
Remember, this command greps for the name you chose for the
driver in step 6. This command must succeed before you can proceed.
</para>
</step>

<step performance="required">
<title>Tell Samba which printer should use these driver files (<literal>setdriver</literal>).</title>

<para>
<indexterm significance="normal"><primary>rpcclient</primary><secondary>setdriver</secondary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">rpcclient -Uroot%xxxx -c 'setdriver mysmbtstprn mydrivername' \
	localhost</userinput>
Successfully set mysmbtstprn to driver mydrivername
</screen></para>

<para>
Since you can bind any printer name (print queue) to any driver, this is a convenient way to set up many
queues that use the same driver. You do not need to repeat all the previous steps for the setdriver command to
succeed. The only preconditions are that <literal>enumdrivers</literal> must find the driver and
<literal>enumprinters</literal> must find the printer.
</para>
</step>

<step performance="required">
<title>(Optional) Verify if Samba has recognized this association.</title>

<para>
<indexterm significance="normal"><primary>rpcclient</primary><secondary>getprinter</secondary></indexterm>
<indexterm significance="normal"><primary>rpcclient</primary><secondary>getdriver</secondary></indexterm>
<indexterm significance="normal"><primary>rpcclient</primary><secondary>enumprinters</secondary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">rpcclient -Uroot%xxxx -c 'getprinter mysmbtstprn 2' localhost \
  | grep driver</userinput>
drivername:[mydrivername]
 
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">rpcclient -Uroot%xxxx -c 'getprinter mysmbtstprn 2' localhost \
  | grep -C4 driv</userinput>
servername:[\\kde-bitshop]
printername:[\\kde-bitshop\mysmbtstprn]
sharename:[mysmbtstprn]
portname:[Done]
drivername:[mydrivername]
comment:[mysmbtstprn]
location:[]
sepfile:[]
printprocessor:[winprint]
 
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">rpcclient -U root%xxxx -c 'getdriver mysmbtstprn' localhost</userinput>
[Windows NT x86]
Printer Driver Info 3:
     Version: [2]
     Driver Name: [mydrivername]
     Architecture: [Windows NT x86]
     Driver Path: [\\kde-bitshop\print$\W32X86\2\cupsdrvr.dll]
     Datafile: [\\kde-bitshop\print$\W32X86\2\mysmbtstprn.PPD]
     Configfile: [\\kde-bitshop\print$\W32X86\2\cupsui.dll]
     Helpfile: [\\kde-bitshop\print$\W32X86\2\cups.hlp]
     Monitorname: []
     Defaultdatatype: [RAW]
     Monitorname: []
     Defaultdatatype: [RAW]
 
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">rpcclient -Uroot%xxxx -c 'enumprinters' localhost \
	| grep mysmbtstprn</userinput>
     name:[\\kde-bitshop\mysmbtstprn]
     description:[\\kde-bitshop\mysmbtstprn,mydrivername,mysmbtstprn]
     comment:[mysmbtstprn]

</screen></para>

<para>
<indexterm significance="normal"><primary>rpcclient</primary><secondary>enumprinters</secondary></indexterm>
Compare these results with the ones from steps 2 and 3. Every one of these commands show the driver is installed. Even
the <literal>enumprinters</literal> command now lists the driver
on the <quote>description</quote> line.
</para>
</step>

<step performance="required">
<title>(Optional.) Tickle the driver into a correct
device mode.</title>

<para>
<indexterm significance="normal"><primary>"Printers" folder</primary></indexterm>
You certainly know how to install the driver on the client. In case
you are not particularly familiar with Windows, here is a short
recipe: Browse the Network Neighborhood, go to the Samba server, and look
for the shares. You should see all shared Samba printers.
Double-click on the one in question. The driver should get
installed and the network connection set up. Another way is to
open the <guilabel moreinfo="none">Printers (and Faxes)</guilabel> folder, right-click on the printer in
question, and select <guilabel moreinfo="none">Connect</guilabel> or <guilabel moreinfo="none">Install</guilabel>. As a result, a new printer
should appear in your client's local <guilabel moreinfo="none">Printers (and Faxes)</guilabel>
folder, named something like <guilabel moreinfo="none">printersharename on Sambahostname</guilabel>.
</para>

<para>
It is important that you execute this step as a Samba printer admin
(as defined in <filename moreinfo="none">smb.conf</filename>). Here is another method
to do this on Windows XP. It uses a command line, which you may type
into the <quote>DOS box</quote> (type root's smbpassword when prompted):
</para>

<para><screen format="linespecific">
<prompt moreinfo="none">C:\&gt; </prompt><userinput moreinfo="none">runas /netonly /user:root "rundll32 printui.dll,PrintUIEntry \
	/in /n \\sambaserver\mysmbtstprn"</userinput>
</screen></para>

<para>
Change any printer setting once (like changing <emphasis><guilabel moreinfo="none">portrait</guilabel> to
<guilabel moreinfo="none">landscape</guilabel></emphasis>), click on <guibutton moreinfo="none">Apply</guibutton>, and change the setting back.
</para>
</step>

<step performance="required">
<title>Install the printer on a client (Point'n'Print).</title>

<para>
<indexterm significance="preferred"><primary>point 'n' print</primary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">C:\&gt; </prompt><userinput moreinfo="none">rundll32 printui.dll,PrintUIEntry /in /n "\\sambaserver\mysmbtstprn"</userinput>
</screen>
If it does not work, it could be a permissions problem with the <parameter>[print$]</parameter> share.
</para>
</step>

<step performance="required">
<title>(Optional) Print a test page.</title>

<indexterm significance="normal"><primary>rundll32</primary></indexterm>
<para><screen format="linespecific">
<prompt moreinfo="none">C:\&gt; </prompt><userinput moreinfo="none">rundll32 printui.dll,PrintUIEntry /p /n "\\sambaserver\mysmbtstprn"</userinput>
</screen></para>

<para>
Then hit [TAB] five times, [ENTER] twice, [TAB] once, and [ENTER] again, and march to the printer.
</para>
</step>

<step performance="required">
<title>(Recommended.) Study the test page.</title>

<para>
Hmmm. Just kidding! By now you know everything about printer installations and you do not need to read a word.
Just put it in a frame and bolt it to the wall with the heading "MY FIRST RPCCLIENT-INSTALLED PRINTER"
<?latex --- ?> why not just throw it away!
</para>
</step>

<step performance="required">
<title>(Obligatory.) Enjoy. Jump. Celebrate your success.</title>

<para><screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">echo "Cheeeeerioooooo! Success..." &gt;&gt; /var/log/samba/log.smbd</userinput>
</screen></para>
</step>
</procedure>
</sect2>

<sect2>
<title>Troubleshooting Revisited</title>

<para>
<indexterm significance="normal"><primary>adddriver</primary></indexterm>
The setdriver command will fail if in Samba's mind the queue is not
already there. A successful installation displys the promising message that the:
<screen format="linespecific">
Printer Driver ABC successfully installed.
</screen>
following the <literal>adddriver</literal> parts of the procedure.  But you may also see
a disappointing message like this one:
<computeroutput moreinfo="none">
result was NT_STATUS_UNSUCCESSFUL
</computeroutput></para>

<para>
<indexterm significance="normal"><primary>lpstat</primary></indexterm>
<indexterm significance="normal"><primary>rpcclient</primary></indexterm>
It is not good enough that you can see the queue in CUPS, using the <literal>lpstat -p ir85wm</literal>
command. A bug in most recent versions of Samba prevents the proper update of the queue list. The recognition
of newly installed CUPS printers fails unless you restart Samba or send a HUP to all smbd processes. To verify
if this is the reason why Samba does not execute the <literal>setdriver</literal> command successfully, check
if Samba <quote>sees</quote> the printer:
<indexterm significance="normal"><primary>rpcclient</primary><secondary>enumprinters</secondary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">rpcclient transmeta -N -U'root%xxxx' -c 'enumprinters 0'|grep ir85wm</userinput>
        printername:[ir85wm]
</screen></para>

<para>
An alternate command could be this: 
<indexterm significance="normal"><primary>rpcclient</primary><secondary>getprinter</secondary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">rpcclient transmeta -N -U'root%secret' -c 'getprinter ir85wm' </userinput>
        cmd = getprinter ir85wm
        flags:[0x800000]
        name:[\\transmeta\ir85wm]
        description:[\\transmeta\ir85wm,ir85wm,DPD]
        comment:[CUPS PostScript-Treiber for Windows NT/200x/XP]
</screen></para>

<para>
By the way, you can use these commands, plus a few more, of course, to install drivers on remote Windows NT print servers too!
</para>
</sect2>
</sect1>

<sect1>
<title>The Printing <filename moreinfo="none">*.tdb</filename> Files</title>

<para>
<indexterm significance="normal"><primary>TDB</primary></indexterm>
<indexterm significance="normal"><primary>connections.tdb</primary><seealso>TDB</seealso></indexterm>
<indexterm significance="normal"><primary>printing.tdb</primary><seealso>TDB</seealso></indexterm>
<indexterm significance="normal"><primary>share_info.tdb</primary><seealso>TDB</seealso></indexterm>
<indexterm significance="normal"><primary>ntdrivers.tdb</primary><seealso>TDB</seealso></indexterm>
<indexterm significance="normal"><primary>unexpected.tdb</primary><seealso>TDB</seealso></indexterm>
<indexterm significance="normal"><primary>brlock.tdb</primary><seealso>TDB</seealso></indexterm>
<indexterm significance="normal"><primary>locking.tdb</primary><seealso>TDB</seealso></indexterm>
<indexterm significance="normal"><primary>ntforms.tdb</primary><seealso>TDB</seealso></indexterm>
<indexterm significance="normal"><primary>messages.tdb</primary><seealso>TDB</seealso></indexterm>
<indexterm significance="normal"><primary>ntprinters.tdb</primary><seealso>TDB</seealso></indexterm>
<indexterm significance="normal"><primary>sessionid.tdb</primary><seealso>TDB</seealso></indexterm>
<indexterm significance="normal"><primary>secrets.tdb</primary><seealso>TDB</seealso></indexterm>
Some mystery is associated with the series of files with a tdb suffix appearing in every Samba installation.
They are <filename moreinfo="none">connections.tdb</filename>, <filename moreinfo="none">printing.tdb</filename>,
<filename moreinfo="none">share_info.tdb</filename>, <filename moreinfo="none">ntdrivers.tdb</filename>, <filename moreinfo="none">unexpected.tdb</filename>,
<filename moreinfo="none">brlock.tdb</filename>, <filename moreinfo="none">locking.tdb</filename>, <filename moreinfo="none">ntforms.tdb</filename>,
<filename moreinfo="none">messages.tdb</filename> , <filename moreinfo="none">ntprinters.tdb</filename>, <filename moreinfo="none">sessionid.tdb</filename>,
and <filename moreinfo="none">secrets.tdb</filename>. What is their purpose?
</para>

<sect2>
<title>Trivial Database Files</title>

<para>
<indexterm significance="normal"><primary>TDB</primary></indexterm>
A Windows NT (print) server keeps track of all information needed to serve its duty toward its clients by
storing entries in the Windows registry. Client queries are answered by reading from the registry,
Administrator or user configuration settings that are saved by writing into the registry. Samba and UNIX
obviously do not have such a Registry. Samba instead keeps track of all client-related information in a series
of <filename moreinfo="none">*.tdb</filename> files. (TDB stands for trivial data base). These are often located in
<filename moreinfo="none">/var/lib/samba/</filename> or <filename moreinfo="none">/var/lock/samba/</filename>. The printing-related files are
<filename moreinfo="none">ntprinters.tdb</filename>, <filename moreinfo="none">printing.tdb</filename>,<filename moreinfo="none">ntforms.tdb</filename>, and
<filename moreinfo="none">ntdrivers.tdb</filename>.
</para>
</sect2>

<sect2>
<title>Binary Format</title>

<para>
<filename moreinfo="none">*.tdb</filename> files are not human readable. They are written in a binary format. <quote>Why not
ASCII?</quote>, you may ask. <quote>After all, ASCII configuration files are a good and proven tradition on
UNIX.</quote> The reason for this design decision by the Samba Team is mainly performance. Samba needs to be
fast; it runs a separate <literal>smbd</literal> process for each client connection, in some environments many
thousands of them. Some of these <literal>smbds</literal> might need to write-access the same
<filename moreinfo="none">*.tdb</filename> file <emphasis>at the same time</emphasis>. The file format of Samba's
<filename moreinfo="none">*.tdb</filename> files allows for this provision. Many smbd processes may write to the same
<filename moreinfo="none">*.tdb</filename> file at the same time. This wouldn't be possible with pure ASCII files.
</para>
</sect2>

<sect2>
<title>Losing <filename moreinfo="none">*.tdb</filename> Files</title>

<para>
It is very important that all <filename moreinfo="none">*.tdb</filename> files remain consistent over all write and read
accesses. However, it may happen that these files <emphasis>do</emphasis> get corrupted. (A <literal>kill -9
`pidof smbd'</literal> while a write access is in progress could do the damage, as could a power interruption,
etc.). In cases of trouble, a deletion of the old printing-related <filename moreinfo="none">*.tdb</filename> files may be the
only option. After that, you need to re-create all print-related setups unless you have made a backup of the
<filename moreinfo="none">*.tdb</filename> files in time.
</para>
</sect2>

<sect2>
<title>Using <literal>tdbbackup</literal></title>

<para>
<indexterm significance="normal"><primary>TDB</primary><secondary>backing up</secondary><see>tdbbackup</see></indexterm>
<indexterm significance="normal"><primary>tdbbackup</primary></indexterm>
Samba ships with a little utility that helps the root user of your system to backup your
<filename moreinfo="none">*.tdb</filename> files. If you run it with no argument, it prints a usage message:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">tdbbackup</userinput>
 Usage: tdbbackup [options] &lt;fname...&gt;
 
 Version:3.0a
   -h            this help message
   -s suffix     set the backup suffix
   -v            verify mode (restore if corrupt)
</screen></para>

<para>
Here is how I backed up my <filename moreinfo="none">printing.tdb</filename> file:
</para>

<para><screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">ls</userinput>
.              browse.dat     locking.tdb     ntdrivers.tdb printing.tdb
..             share_info.tdb connections.tdb messages.tdb  ntforms.tdb
printing.tdbkp unexpected.tdb brlock.tdb      gmon.out      namelist.debug  
ntprinters.tdb sessionid.tdb
 
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">tdbbackup -s .bak printing.tdb</userinput>
 printing.tdb : 135 records
 
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">ls -l printing.tdb*</userinput>
 -rw-------    1 root     root        40960 May  2 03:44 printing.tdb
 -rw-------    1 root     root        40960 May  2 03:44 printing.tdb.bak

</screen></para>
</sect2>
</sect1>

<sect1>
<title>CUPS Print Drivers from Linuxprinting.org</title>

<para>
<indexterm significance="normal"><primary>Linuxprinting.org</primary></indexterm>
CUPS ships with good support for HP LaserJet-type printers. You can install the generic driver as follows:
<indexterm significance="normal"><primary>lpadmin</primary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">lpadmin -p laserjet4plus -v parallel:/dev/lp0 -E -m laserjet.ppd</userinput>
</screen></para>

<para>
The <option>-m</option> switch will retrieve the <filename moreinfo="none">laserjet.ppd</filename> from the standard
repository for not-yet-installed PPDs, which CUPS typically stores in
<filename moreinfo="none">/usr/share/cups/model</filename>. Alternatively, you may use <option>-P /path/to/your.ppd</option>.
</para>

<para>
The generic <filename moreinfo="none">laserjet.ppd,</filename> however, does not support every special option for every
LaserJet-compatible model. It constitutes a sort of <quote>least common denominator</quote> of all the models.
If for some reason you must pay for the commercially available ESP Print Pro drivers, your first move should
be to consult the database on the <ulink noescape="1" url="http://www.linuxprinting.org/printer_list.cgi">Linuxprinting</ulink> Web site.  Linuxprinting.org has
excellent recommendations about which driver is best used for each printer. Its database is kept current by
the tireless work of Till Kamppeter from Mandrakesoft, who is also the principal author of the
<literal>foomatic-rip</literal> utility.
</para>

<note><para>
<indexterm significance="normal"><primary>foomatic-rip</primary></indexterm>
<indexterm significance="normal"><primary>cupsomatic</primary></indexterm>
<indexterm significance="normal"><primary>Adobe PPD</primary></indexterm>
The former <literal>cupsomatic</literal> concept is now being replaced by the new successor, a much more
powerful <literal>foomatic-rip</literal>.  <literal>cupsomatic</literal> is no longer maintained. Here is the
new URL to the <ulink noescape="1" url="http://www.linuxprinting.org/driver_list.cgi">Foomatic-3.0</ulink>
database.  If you upgrade to <literal>foomatic-rip</literal>, remember to also upgrade to the new-style PPDs
for your Foomatic-driven printers. foomatic-rip will not work with PPDs generated for the old
<literal>cupsomatic</literal>. The new-style PPDs are 100% compliant with the Adobe PPD specification. They
are also intended to be used by Samba and the cupsaddsmb utility, to provide the driver files for the Windows
clients!
</para></note>

<sect2>
<title>foomatic-rip and Foomatic Explained</title>


<para>
<indexterm significance="preferred"><primary>foomatic</primary></indexterm>
<indexterm significance="preferred"><primary>foomatic-rip</primary></indexterm>
Nowadays, most Linux distributions rely on the utilities from the <ulink url="http://www.linuxprinting.org/">Linuxprinting.org</ulink> to create their printing-related software
(which, by the way, works on all UNIXes and on Mac OS X and Darwin, too).  The utilities from this sire have a
very end-user-friendly interface that allows for an easy update of drivers and PPDs for all supported models,
all spoolers, all operating systems, and all package formats (because there is none). Its history goes back a
few years.
</para>

<para>
Recently, Foomatic has achieved the astonishing milestone of <ulink url="http://www.linuxprinting.org/printer_list.cgi?make=Anyone">1,000 listed</ulink> printer models.
Linuxprinting.org keeps all the important facts about printer drivers, supported models, and which options are
available for the various driver/printer combinations in its <ulink url="http://www.linuxprinting.org/foomatic.html">Foomatic</ulink> database. Currently there are <ulink url="http://www.linuxprinting.org/driver_list.cgi">245 drivers</ulink> in the database. Many drivers support
various models, and many models may be driven by different drivers <?latex --- ?> its your choice!
</para>

<sect3>
<title>690 <quote>Perfect</quote> Printers</title>

<para>
<indexterm significance="normal"><primary>Windows PPD</primary></indexterm>
At present, there are 690 devices dubbed as working perfectly: 181 are <emphasis>mostly</emphasis> perfect, 96
are <emphasis>partially</emphasis> perfect, and 46 are paperweights. Keeping in mind that most of these are
non-PostScript models (PostScript printers are automatically supported by CUPS to perfection by using their
own manufacturer-provided Windows PPD), and that a multifunctional device never qualifies as working perfectly
if it does not also scan and copy and fax under GNU/Linux <?latex --- ?> then this is a truly astonishing
achievement! Three years ago the number was not more than 500, and Linux or UNIX printing at the time wasn't
anywhere near the quality it is today.
</para>
</sect3>

<sect3>
<title>How the Printing HOWTO Started It All</title>

<para>
A few years ago <ulink url="http://www2.picante.com/">Grant Taylor</ulink> started it all. The
roots of today's Linuxprinting.org are in the first <ulink url="http://www.linuxprinting.org/foomatic2.9/howto/">Linux Printing HOWTO</ulink> that he authored. As a
side-project to this document, which served many Linux users and admins to guide their first steps in this
complicated and delicate setup (to a scientist, printing is <quote>applying a structured deposition of
distinct patterns of ink or toner particles on paper substrates</quote>), he started to build in a little
Postgres database with information about the hardware and driver zoo that made up Linux printing of the time.
This database became the core component of today's Foomatic collection of tools and data. In the meantime, it
has moved to an XML representation of the data.
</para>
</sect3>

<sect3>
<title>Foomatic's Strange Name</title>


<para>
<indexterm significance="normal"><primary>foomatic</primary></indexterm>
<quote>Why the funny name?</quote> you ask. When it really took off, around spring 2000, CUPS was far less
popular than today, and most systems used LPD, LPRng, or even PDQ to print. CUPS shipped with a few generic
drivers (good for a few hundred different printer models). These didn't support many device-specific options.
CUPS also shipped with its own built-in rasterization filter (<parameter moreinfo="none">pstoraster</parameter>, derived from
Ghostscript). On the other hand, CUPS provided brilliant support for <emphasis>controlling</emphasis> all
printer options through standardized and well-defined PPD files.  Plus, CUPS was designed to be easily
extensible.
</para>

<para>
Taylor already had in his database a respectable compilation of facts about many more printers and the
Ghostscript <quote>drivers</quote> they run with. His idea, to generate PPDs from the database information and
use them to make standard Ghostscript filters work within CUPS, proved to work very well. It also killed
several birds with one stone:
</para>

<itemizedlist>
	<listitem><para>It made all current and future Ghostscript filter
	developments available for CUPS.</para></listitem>

	<listitem><para>It made available a lot of additional printer models
	to CUPS users (because often the traditional Ghostscript way of
	printing was the only one available).</para></listitem>

	<listitem><para>It gave all the advanced CUPS options (Web interface,
	GUI driver configurations) to users wanting (or needing) to use
	Ghostscript filters.</para></listitem>
</itemizedlist>
</sect3>

<sect3>
<title>cupsomatic, pdqomatic, lpdomatic, directomatic</title>

<para>
<indexterm significance="normal"><primary>cupsomatic</primary></indexterm>
<indexterm significance="normal"><primary>CUPS-PPD</primary></indexterm>
<indexterm significance="normal"><primary>PPD</primary><secondary>CUPS</secondary><see>CUPS-PPD</see></indexterm>
CUPS worked through a quickly hacked-up filter script named <ulink url="http://www.linuxprinting.org/download.cgi?filename=cupsomatic&amp;show=0">cupsomatic</ulink>.  cupsomatic
ran the printfile through Ghostscript, constructing automatically the rather complicated command line needed.
It just needed to be copied into the CUPS system to make it work. To configure the way cupsomatic controls the
Ghostscript rendering process, it needs a CUPS-PPD. This PPD is generated directly from the contents of the
database. For CUPS and the respective printer/filter combo, another Perl script named CUPS-O-Matic did the PPD
generation. After that was working, Taylor implemented within a few days a similar thing for two other
spoolers. Names chosen for the config-generator scripts were <ulink url="http://www.linuxprinting.org/download.cgi?filename=lpdomatic&amp;show=0">PDQ-O-Matic</ulink> (for PDQ)
and <ulink url="http://www.linuxprinting.org/download.cgi?filename=lpdomatic&amp;show=0">LPD-O-Matic</ulink>
(for <?latex --- ?> you guessed it <?latex --- ?> LPD); the configuration here didn't use PPDs but other
spooler-specific files.
</para>

<para>
From late summer of that year, <ulink url="http://www.linuxprinting.org/till/">Till Kamppeter</ulink> started
to put work into the database. Kamppeter had been newly employed by <ulink url="http://www.mandrakesoft.com/">Mandrakesoft</ulink> to convert its printing system over to CUPS, after
they had seen his <ulink url="http://www.fltk.org/">FLTK</ulink>-based <ulink url="http://cups.sourceforge.net/xpp/">XPP</ulink> (a GUI front-end to the CUPS lp-command). He added a huge
amount of new information and new printers. He also developed the support for other spoolers, like <ulink url="http://ppr.sourceforge.net/">PPR</ulink> (via ppromatic), <ulink url="http://sourceforge.net/projects/lpr/">GNUlpr</ulink>, and <ulink url="http://www.lprng.org/">LPRng</ulink> (both via an extended lpdomatic) and spooler-less printing (<ulink url="http://www.linuxprinting.org/download.cgi?filename=directomatic&amp;show=0">directomatic</ulink>).
</para>

<para>
So, to answer your question, <quote>Foomatic</quote> is the general name for all the overlapping code and data
behind the <quote>*omatic</quote> scripts.  Foomatic, up to versions 2.0.x, required (ugly) Perl data
structures attached to Linuxprinting.org PPDs for CUPS. It had a different <quote>*omatic</quote> script for
every spooler, as well as different printer configuration files.
</para>
</sect3>

<sect3>
<title>The <emphasis>Grand Unification</emphasis> Achieved</title>

<para>
<indexterm significance="normal"><primary>foomatic-rip</primary></indexterm>
This has all changed in Foomatic versions 2.9 (beta) and released as <quote>stable</quote> 3.0. It has now
achieved the convergence of all *omatic scripts and is called the <ulink url="http://www.linuxprinting.org/foomatic2.9/download.cgi?filename=foomatic-rip&amp;show=0">foomatic-rip</ulink>.
This single script is the unification of the previously different spooler-specific *omatic scripts.
foomatic-rip is used by all the different spoolers alike, and because it can read PPDs (both the original
PostScript printer PPDs and the Linuxprinting.org-generated ones), all of a sudden all supported spoolers can
have the power of PPDs at their disposal. Users only need to plug foomatic-rip into their system. For users
there is improved media type and source support <?latex --- ?> paper sizes and trays are easier to configure.
</para>

<para>
<indexterm significance="normal"><primary>PPDs</primary></indexterm>
<indexterm significance="normal"><primary>Foomatic tutorial</primary></indexterm>
<indexterm significance="normal"><primary>LinuxKongress2002</primary></indexterm>
Also, the new generation of Linuxprinting.org PPDs no longer contains Perl data structures. If you are a
distro maintainer and have used the previous version of Foomatic, you may want to give the new one a spin, but
remember to generate a new-version set of PPDs via the new <ulink url="http://www.linuxprinting.org/download/foomatic/foomatic-db-engine-3.0.0beta1.tar.gz">foomatic-db-engine!</ulink>.
Individual users just need to generate a single new PPD specific to their model by <ulink url="http://www.linuxprinting.org/kpfeifle/LinuxKongress2002/Tutorial/II.Foomatic-User/II.tutorial-handout-foomatic-user.html">following
the steps</ulink> outlined in the Foomatic tutorial or in this chapter. This new development is truly amazing.
</para>

<para>
<indexterm significance="normal"><primary>foomatic-rip</primary></indexterm>
<indexterm significance="normal"><primary>Adobe</primary></indexterm>
<indexterm significance="normal"><primary>printer drivers</primary></indexterm>
foomatic-rip is a very clever wrapper around the need to run Ghostscript with a different syntax, options,
device selections, and/or filters for each different printer or spooler. At the same time, it can read the PPD
associated with a print queue and modify the print job according to the user selections. Together with this
comes the 100% compliance of the new Foomatic PPDs with the Adobe spec. Some innovative features of the
Foomatic concept may surprise users. It will support custom paper sizes for many printers and will support
printing on media drawn from different paper trays within the same job (in both cases, even where there is no
support for this from Windows-based vendor printer drivers).
</para>
</sect3>

<sect3>
<title>Driver Development Outside</title>

<para>
<indexterm significance="normal"><primary>Linuxprinting.org</primary></indexterm>
Most driver development itself does not happen within Linuxprinting.org. Drivers are written by independent
maintainers.  Linuxprinting.org just pools all the information and stores it in its database. In addition, it
also provides the Foomatic glue to integrate the many drivers into any modern (or legacy) printing system
known to the world.
</para>

<para>
Speaking of the different driver development groups, most of the work is currently done in three projects:
</para>

<itemizedlist>
	<listitem><para>
<indexterm significance="normal"><primary>Omni</primary></indexterm>
	<ulink url="http://www-124.ibm.com/developerworks/oss/linux/projects/omni/">Omni</ulink>
	<?latex --- ?> a free software project by IBM that tries to convert its printer
	driver knowledge from good-ol' OS/2 times into a modern, modular,
	universal driver architecture for Linux/UNIX (still beta). This
	currently supports 437 models.</para></listitem>

	<listitem><para>
<indexterm significance="normal"><primary>HPIJS</primary></indexterm>
	<ulink url="http://hpinkjet.sf.net/">HPIJS</ulink> <?latex --- ?>
	a free software project by HP to provide the support for its own
	range of models (very mature, printing in most cases is perfect and
	provides true photo quality). This currently supports 369
	models.</para></listitem>

	<listitem><para>
<indexterm significance="normal"><primary>Gutenprint</primary></indexterm>
	<ulink url="http://gimp-print.sourceforge.net/">Gutenprint</ulink> <?latex --- ?> a free software
	effort, started by Michael Sweet (also lead developer for CUPS), now
	directed by Robert Krawitz, which has achieved an amazing level of
	photo print quality (many Epson users swear that its quality is
	better than the vendor drivers provided by Epson for the Microsoft
	platforms). This currently supports 522 models.</para></listitem>
</itemizedlist>
</sect3>

<sect3>
<title>Forums, Downloads, Tutorials, Howtos (Also for Mac OS X and Commercial UNIX)</title>

<para>
Linuxprinting.org today is the one-stop shop to download printer drivers. Look for printer information and
<ulink url="http://www.linuxprinting.org//kpfeifle/LinuxKongress2002/Tutorial/">tutorials</ulink> or solve
printing problems in its popular <ulink url="http://www.linuxprinting.org/newsportal/">forums</ulink>. This
forum is not just for GNU/Linux users, but admins of <ulink url="http://www.linuxprinting.org/macosx/">
commercial UNIX systems</ulink> are also going there, and the relatively new
<ulink url="http://www.linuxprinting.org/newsportal/thread.php3?name=linuxprinting.macosx.general">Mac OS X
forum</ulink> has turned out to be one of the most frequented forums after only a few weeks.
</para>

<para>
<indexterm significance="normal"><primary>Mandriva</primary></indexterm>
<indexterm significance="normal"><primary>Mandrake</primary></indexterm>
<indexterm significance="normal"><primary>Conectiva</primary></indexterm>
Linuxprinting.org and the Foomatic driver wrappers around Ghostscript are now a standard tool-chain for
printing on all the important distros. Most of them also have CUPS underneath. While in recent years most
printer data had been added by Kamppeter, many additional contributions came from engineers with SuSE, Red
Hat, Conectiva, Debian, and others. Vendor-neutrality is an important goal of the Foomatic project. Mandrake
and Conectiva have merged and are now called Mandriva.
</para>

<note><para>
Till Kamppeter from Mandrakesoft is doing an excellent job in his spare time to maintain Linuxprinting.org and
Foomatic. So if you use it often, please send him a note showing your appreciation.
</para></note>
</sect3>

<sect3>
<title>Foomatic Database-Generated PPDs</title>

<para>
<indexterm significance="normal"><primary>Foomatic database</primary></indexterm>
<indexterm significance="normal"><primary>XML-based datasets</primary></indexterm>
<indexterm significance="normal"><primary>kprinter</primary></indexterm>
<indexterm significance="normal"><primary>gtklp</primary></indexterm>
<indexterm significance="normal"><primary>xpp</primary></indexterm>
<indexterm significance="normal"><primary>HP Photosmart</primary></indexterm>
<indexterm significance="normal"><primary>Epson Stylus inkjet</primary></indexterm>
<indexterm significance="normal"><primary>non-PostScript printers</primary></indexterm>
<indexterm significance="normal"><primary>raster</primary></indexterm>
The Foomatic database is an amazing piece of ingenuity in itself. Not only does it keep the printer and driver
information, but it is organized in a way that it can generate PPD files on the fly from its internal
XML-based datasets. While these PPDs are modeled to the Adobe specification of PPDs, the
Linuxprinting.org/Foomatic-PPDs do not normally drive PostScript printers. They are used to describe all the
bells and whistles you could ring or blow on an Epson Stylus inkjet, or an HP Photosmart, or what-have-you.
The main trick is one little additional line, not envisaged by the PPD specification, starting with the
<parameter moreinfo="none">*cupsFilter</parameter> keyword. It tells the CUPS daemon how to proceed with the PostScript print
file (old-style Foomatic-PPDs named the cupsomatic filter script, while the new-style PPDs are now call
foomatic-rip). This filter script calls Ghostscript on the host system (the recommended variant is ESP
Ghostscript) to do the rendering work. foomatic-rip knows which filter or internal device setting it should
ask from Ghostscript to convert the PostScript print job into a raster format ready for the target device.
This usage of PPDs to describe the options of non-PostScript printers was the invention of the CUPS
developers. The rest is easy.  GUI tools (like KDE's marvelous <ulink url="http://printing.kde.org/overview/kprinter.phtml">kprinter</ulink> or the GNOME <ulink url="http://gtklp.sourceforge.net/">gtklp</ulink> xpp and the CUPS Web interface) read the PPD as well and use
this information to present the available settings to the user as an intuitive menu selection.
</para>
</sect3>
</sect2>

<sect2>
<title>foomatic-rip and Foomatic PPD Download and Installation</title>

<para>
Here are the steps to install a foomatic-rip-driven LaserJet 4 Plus-compatible
printer in CUPS (note that recent distributions of SuSE, UnitedLinux and
Mandrake may ship with a complete package of Foomatic-PPDs plus the
<literal>foomatic-rip</literal> utility. Going directly to
Linuxprinting.org ensures that you get the latest driver/PPD files).
</para>

<itemizedlist>
	<listitem><para>Open your browser at the Linuxprinting.org printer list <ulink url="http://www.linuxprinting.org/printer_list.cgi">page.</ulink>
	</para></listitem>

	<listitem><para>Check the complete list of printers in the 
	<ulink url="http://www.linuxprinting.org/printer_list.cgi?make=Anyone">database.</ulink>.
	</para></listitem>

	<listitem><para>Select your model and click on the link.
	</para></listitem>

	<listitem><para>You'll arrive at a page listing all drivers working with this
	model (for all printers, there will always be <emphasis>one</emphasis>
	recommended driver. Try this one first).
	</para></listitem>

	<listitem><para>In our case (HP LaserJet 4 Plus), we'll arrive at the default driver for the
	<ulink url="http://www.linuxprinting.org/show_printer.cgi?recnum=HP-LaserJet_4_Plus">HP-LaserJet 4 Plus.</ulink>
	</para></listitem>

	<listitem><para>The recommended driver is ljet4.</para></listitem>

	<listitem><para>Several links are provided here. You should visit them all if you
	are not familiar with the Linuxprinting.org database.
	</para></listitem>

	<listitem><para>There is a link to the database page for the
	<ulink url="http://www.linuxprinting.org/show_driver.cgi?driver=ljet4">ljet4</ulink>.
	On the driver's page, you'll find important and detailed information
	about how to use that driver within the various available
	spoolers.</para></listitem>

	<listitem><para>Another link may lead you to the home page of the
	author of the driver.</para></listitem>

	<listitem><para>Important links are the ones that provide hints with
	setup instructions for <ulink noescape="1" url="http://www.linuxprinting.org/cups-doc.html">CUPS</ulink>;
	<ulink url="http://www.linuxprinting.org/pdq-doc.html">PDQ</ulink>;
	<ulink url="http://www.linuxprinting.org/lpd-doc.html">LPD, LPRng, and GNUlpr</ulink>);
	as well as <ulink url="http://www.linuxprinting.org/ppr-doc.html">PPR</ulink>
	or <quote>spoolerless</quote> <ulink url="http://www.linuxprinting.org/direct-doc.html">printing</ulink>.
	</para></listitem>

	<listitem><para>You can view the PPD in your browser through this link:
	<ulink noescape="1" url="http://www.linuxprinting.org/ppd-o-matic.cgi?driver=ljet4&amp;printer=HP-LaserJet_4_Plus&amp;show=1">http://www.linuxprinting.org/ppd-o-matic.cgi?driver=ljet4&amp;printer=HP-LaserJet_4_Plus&amp;show=1</ulink>
	</para></listitem> <listitem><para>Most importantly, you can also generate and download
	the <ulink url="http://www.linuxprinting.org/ppd-o-matic.cgi?driver=ljet4&amp;printer=HP-LaserJet_4_Plus&amp;show=0">PPD</ulink>.
	</para></listitem>

	<listitem><para>The PPD contains all the information needed to use our
	model and the driver; once installed, this works transparently
	for the user. Later you'll only need to choose resolution, paper size,
	and so on, from the Web-based menu, or from the print dialog GUI, or from
	the command line.</para></listitem>

	<listitem><para>If you ended up on the drivers
	<ulink url="http://www.linuxprinting.org/show_driver.cgi?driver=ljet4">page</ulink>,
	you can choose to use the <quote>PPD-O-Matic</quote> online PPD generator
	program.</para></listitem>

	<listitem><para>Select the exact model and check either <guilabel moreinfo="none">Download</guilabel> or
	<guilabel moreinfo="none">Display PPD file</guilabel> and click <guilabel moreinfo="none">Generate PPD file</guilabel>.</para></listitem>

	<listitem><para>If you save the PPD file from the browser view, please
	do not use cut and paste (since it could possibly damage line endings
	and tabs, which makes the PPD likely to fail its duty), but use <guimenuitem moreinfo="none">Save
	as...</guimenuitem> in your browser's menu. (It is best to use the <guilabel moreinfo="none">Download</guilabel> option
	directly from the Web page.)</para></listitem>

	<listitem><para>Another interesting part on each driver page is
	the <guimenuitem moreinfo="none">Show execution details</guimenuitem> button. If you
	select your printer model and click on that button,
	a complete Ghostscript command line will be displayed, enumerating all options
	available for that combination of driver and printer model. This is a great way to
	<quote>learn Ghostscript by doing</quote>. It is also an excellent cheat sheet
	for all experienced users who need to reconstruct a good command line
	for that darned printing script, but can't remember the exact
	syntax. </para></listitem>

	<listitem><para>Sometime during your visit to Linuxprinting.org, save
	the PPD to a suitable place on your hard disk, say
	<filename moreinfo="none">/path/to/my-printer.ppd</filename> (if you prefer to install
	your printers with the help of the CUPS Web interface, save the PPD to
	the <filename moreinfo="none">/usr/share/cups/model/</filename> path and restart
	cupsd).</para></listitem>

	<listitem><para>Then install the printer with a suitable command line,
	like this: 
	</para>

	<para><screen format="linespecific">
	<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">lpadmin -p laserjet4plus -v parallel:/dev/lp0 -E \
		-P path/to/my-printer.ppd</userinput>
	</screen></para></listitem>

	<listitem><para>For all the new-style <quote>Foomatic-PPDs</quote>
	from Linuxprinting.org, you also need a special CUPS filter named
	foomatic-rip. 
	</para></listitem>

	<listitem><para>The foomatic-rip Perl script itself also makes some
	interesting <ulink url="http://www.linuxprinting.org/foomatic2.9/download.cgi?filename=foomatic-rip&amp;show=1">reading</ulink>
	because it is well documented by Kamppeter's in-line comments (even
	non-Perl hackers will learn quite a bit about printing by reading
	it).</para></listitem>

	<listitem><para>Save foomatic-rip either directly in
	<filename moreinfo="none">/usr/lib/cups/filter/foomatic-rip</filename> or somewhere in
	your $PATH (and remember to make it world-executable). Again,
	do not save by copy and paste but use the appropriate link or the
	<guimenuitem moreinfo="none">Save as...</guimenuitem>  menu item in your browser.</para></listitem>

	<listitem><para>If you save foomatic-rip in your $PATH, create a symlink:
	<screen format="linespecific">
	<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">cd /usr/lib/cups/filter/ ; ln -s `which foomatic-rip'</userinput>
	</screen>
	</para>

	<para>
	CUPS will discover this new available filter at startup after restarting
	cupsd.</para></listitem>
</itemizedlist>

<para>
Once you print to a print queue set up with the Foomatic PPD, CUPS will insert the appropriate commands and
comments into the resulting PostScript job file. foomatic-rip is able to read and act upon these and uses some
specially encoded Foomatic comments embedded in the job file. These in turn are used to construct
(transparently for you, the user) the complicated Ghostscript command line telling the printer driver exactly
how the resulting raster data should look and which printer commands to embed into the data stream. You need:
</para>

<itemizedlist>
	<listitem><para>A <quote>foomatic+something</quote> PPD <?latex --- ?> but this is not enough
	to print with CUPS (it is only <emphasis>one</emphasis> important
	component).</para></listitem>

	<listitem><para>The <parameter moreinfo="none">foomatic-rip</parameter> filter script (Perl) in
	<filename moreinfo="none">/usr/lib/cups/filters/</filename>.</para></listitem>

	<listitem><para>Perl to make foomatic-rip run.</para></listitem>

	<listitem><para>Ghostscript (because it is doing the main work,
	controlled by the PPD/foomatic-rip combo) to produce the raster data
	fit for your printer model's consumption.</para></listitem>

	<listitem><para>Ghostscript <emphasis>must</emphasis> (depending on
	the driver/model) contain support for a certain device representing
	the selected driver for your model (as shown by <literal>gs -h</literal>).</para></listitem>

	<listitem><para>foomatic-rip needs a new version of PPDs (PPD versions
	produced for cupsomatic do not work with foomatic-rip).</para></listitem>
</itemizedlist>
</sect2>
</sect1>

<sect1>
<title>Page Accounting with CUPS</title>


<para>
<indexterm significance="normal"><primary>CUPS</primary><secondary>Page Accounting</secondary></indexterm>
Often there are questions regarding print quotas where Samba users (that is, Windows clients) should not be
able to print beyond a certain number of pages or data volume per day, week, or month. This feature is
dependent on the real print subsystem you're using.  Samba's part is always to receive the job files from the
clients (filtered <emphasis>or</emphasis> unfiltered) and hand them over to this printing subsystem.
</para>

<para>
Of course one could hack things with one's own scripts. But then there is CUPS. CUPS supports quotas that can
be based on the size of jobs or on the number of pages or both, and can span any time period you want.
</para>

<sect2>
<title>Setting Up Quotas</title>

<para>
<indexterm significance="normal"><primary>CUPS</primary><secondary>quotas</secondary></indexterm>
This is an example command of how root would set a print quota in CUPS, assuming an existing printer named
<quote>quotaprinter</quote>:
<indexterm significance="normal"><primary>lpadmin</primary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">lpadmin -p quotaprinter -o job-quota-period=604800 \
	-o job-k-limit=1024 -o job-page-limit=100</userinput>
</screen></para>

<para>
This would limit every single user to print no more than 100 pages or 1024 KB of
data (whichever comes first) within the last 604,800 seconds ( = 1 week).
</para>
</sect2>

<sect2>
<title>Correct and Incorrect Accounting</title>

<para>
For CUPS to count correctly, the printfile needs to pass the CUPS pstops filter; otherwise it uses a dummy
count of <quote>one</quote>. Some print files do not pass it (e.g., image files), but then those are mostly
one-page jobs anyway. This also means that proprietary drivers for the target printer running on the client
computers and CUPS/Samba, which then spool these files as <quote>raw</quote> (i.e., leaving them untouched,
not filtering them), will be counted as one-pagers too!
</para>

<para>
You need to send PostScript from the clients (i.e., run a PostScript driver there) to have the chance to get
accounting done. If the printer is a non-PostScript model, you need to let CUPS do the job to convert the file
to a print-ready format for the target printer. This is currently working for about a thousand different
printer models.  Linuxprinting.org has a driver <ulink url="http://www.linuxprinting.org/printer_list.cgi">list</ulink>.
</para>
</sect2>

<sect2>
<title>Adobe and CUPS PostScript Drivers for Windows Clients</title>

<para>
<indexterm significance="normal"><primary>Adobe PostScript</primary></indexterm>
<indexterm significance="normal"><primary>pstops</primary></indexterm>
<indexterm significance="normal"><primary>PPD</primary></indexterm>
<indexterm significance="normal"><primary>pstoraster</primary></indexterm>
<indexterm significance="normal"><primary>PJL-header</primary></indexterm>
Before CUPS 1.1.16, your only option was to use the Adobe PostScript driver on the Windows clients. The output
of this driver was not always passed through the <literal>pstops</literal> filter on the CUPS/Samba side, and
therefore was not counted correctly (the reason is that it often, depending on the PPD being used, wrote a
PJL-header in front of the real PostScript, which caused CUPS to skip <literal>pstops</literal> and go
directly to the <literal>pstoraster</literal> stage).
</para>

<para>
From CUPS 1.1.16 and later releases, you can use the CUPS PostScript driver for Windows NT/200x/XP
clients (which is tagged in the download area of <filename moreinfo="none">http://www.cups.org/</filename> as the
<filename moreinfo="none">cups-samba-1.1.16.tar.gz</filename> package). It does <emphasis>not</emphasis> work for Windows
9x/Me clients, but it guarantees:
</para>

<itemizedlist>
	<listitem><para> <indexterm significance="normal"><primary>PJL</primary></indexterm> To not write a PJL-header.</para></listitem>

	<listitem><para>To still read and support all PJL-options named in the
	driver PPD with its own means.</para></listitem>

	<listitem><para>That the file will pass through the <literal>pstops</literal> filter
	on the CUPS/Samba server.</para></listitem>

	<listitem><para>To page-count correctly the print file.</para></listitem>
</itemizedlist>

<para>
You can read more about the setup of this combination in the man page for <literal>cupsaddsmb</literal> (which
is only present with CUPS installed, and only current from CUPS 1.1.16).
</para>
</sect2>

<sect2>
<title>The page_log File Syntax</title>

<para>
<indexterm significance="normal"><primary>page_log</primary></indexterm>
These are the items CUPS logs in the <filename moreinfo="none">page_log</filename> for every page of a job:
</para>

<itemizedlist>
	<listitem><para>Printer name</para></listitem>

	<listitem><para>User name</para></listitem>

	<listitem><para>Job ID</para></listitem>

	<listitem><para>Time of printing</para></listitem>

	<listitem><para>Page number</para></listitem>

	<listitem><para>Number of copies</para></listitem>

	<listitem><para>A billing information string (optional)</para></listitem>

	<listitem><para>The host that sent the job (included since version 1.1.19)</para></listitem>
</itemizedlist>

<para>
Here is an extract of my CUPS server's <filename moreinfo="none">page_log</filename> file to illustrate the
format and included items:
</para>

<para><screen format="linespecific">
tec_IS2027 kurt 401 [22/Apr/2003:10:28:43 +0100] 1 3 #marketing 10.160.50.13
tec_IS2027 kurt 401 [22/Apr/2003:10:28:43 +0100] 2 3 #marketing 10.160.50.13
tec_IS2027 kurt 401 [22/Apr/2003:10:28:43 +0100] 3 3 #marketing 10.160.50.13
tec_IS2027 kurt 401 [22/Apr/2003:10:28:43 +0100] 4 3 #marketing 10.160.50.13
Dig9110 boss 402 [22/Apr/2003:10:33:22 +0100] 1 440 finance-dep 10.160.51.33
</screen></para>

<para>
This was job ID <parameter moreinfo="none">401</parameter>, printed on <parameter moreinfo="none">tec_IS2027</parameter>
by user <parameter moreinfo="none">kurt</parameter>, a 64-page job printed in three copies, billed to
<parameter moreinfo="none">#marketing</parameter>, and sent from IP address <constant>10.160.50.13.</constant>
 The next job had ID <parameter moreinfo="none">402</parameter>, was sent by user <parameter moreinfo="none">boss</parameter>
from IP address <constant>10.160.51.33</constant>, printed from one page 440 copies, and
is set to be billed to <parameter moreinfo="none">finance-dep</parameter>.
</para>
</sect2>

<sect2>
<title>Possible Shortcomings</title>

<para>
What flaws or shortcomings are there with this quota system?
</para>

<itemizedlist>
	<listitem><para>The ones named above (wrongly logged job in case of
	printer hardware failure, and so on).</para></listitem>

	<listitem><para>In reality, CUPS counts the job pages that are being
	processed in <emphasis>software</emphasis> (that is, going through the
	RIP) rather than the physical sheets successfully leaving the
	printing device. Thus, if there is a jam while printing the fifth sheet out
	of 1,000 and the job is aborted by the printer, the page count will
	still show the figure of 1,000 for that job.</para></listitem>

	<listitem><para>All quotas are the same for all users (no flexibility
	to give the boss a higher quota than the clerk) and no support for
	groups.</para></listitem>

	<listitem><para>No means to read out the current balance or the
	<quote>used-up</quote> number of current quota.</para></listitem>

	<listitem><para>A user having used up 99 sheets of a 100 quota will
	still be able to send and print a 1,000 sheet job.</para></listitem>

	<listitem><para>A user being denied a job because of a filled-up quota
	does not get a meaningful error message from CUPS other than
	<quote>client-error-not-possible</quote>.</para></listitem>
</itemizedlist>
</sect2>

<sect2>
<title>Future Developments</title>

<para>
This is the best system currently available, and there are huge
improvements under development for CUPS 1.2:
</para>

<itemizedlist>
	<listitem><para>Page counting will go into the backends (these talk
	directly to the printer and will increase the count in sync with the
	actual printing process; thus, a jam at the fifth sheet will lead to a
	stop in the counting).</para></listitem>

	<listitem><para>Quotas will be handled more flexibly.</para></listitem>

	<listitem><para>Probably there will be support for users to inquire
	about their accounts in advance.</para></listitem>

	<listitem><para>Probably there will be support for some other tools
	around this topic.</para></listitem>
</itemizedlist>
</sect2>

<sect2>
<title>Other Accounting Tools</title>

<para>
Other accounting tools that can be used includes: PrintAnalyzer, pyKota, printbill, LogReport.
For more information regarding these tools you can try a Google search.
</para>

</sect2>
</sect1>

<sect1>
<title>Additional Material</title>

<para>
A printer queue with <emphasis>no</emphasis> PPD associated to it is a
<quote>raw</quote> printer, and all files will go directly there as received by the
spooler. The exceptions are file types <parameter moreinfo="none">application/octet-stream</parameter>
that need the pass-through feature enabled. <quote>Raw</quote> queues do not do any
filtering at all; they hand the file directly to the CUPS backend.
This backend is responsible for sending the data to the device
(as in the <quote>device URI</quote> notation: <filename moreinfo="none">lpd://, socket://,
smb://, ipp://, http://, parallel:/, serial:/, usb:/</filename>, and so on).
</para>

<para>
cupsomatic/Foomatic are <emphasis>not</emphasis> native CUPS drivers
and they do not ship with CUPS. They are a third-party add-on
developed at Linuxprinting.org. As such, they are a brilliant hack to
make all models (driven by Ghostscript drivers/filters in traditional
spoolers) also work via CUPS, with the same (good or bad!) quality as
in these other spoolers. <parameter moreinfo="none">cupsomatic</parameter> is only a vehicle to execute a
Ghostscript command line at that stage in the CUPS filtering chain
where normally the native CUPS <parameter moreinfo="none">pstoraster</parameter> filter would kick
in. <parameter moreinfo="none">cupsomatic</parameter> bypasses <parameter moreinfo="none">pstoraster</parameter>, kidnaps the print file from CUPS,
and redirects it to go through Ghostscript. CUPS accepts this
because the associated cupsomatic/foomatic-PPD specifies:

<programlisting format="linespecific">
*cupsFilter:  "application/vnd.cups-postscript 0 cupsomatic"
</programlisting>

This line persuades CUPS to hand the file to <parameter moreinfo="none">cupsomatic</parameter> once it has
successfully converted it to the MIME type
<parameter moreinfo="none">application/vnd.cups-postscript</parameter>. This conversion will not happen for
jobs arriving from Windows that are autotyped
<parameter moreinfo="none">application/octet-stream</parameter>, with the according changes in
<filename moreinfo="none">/etc/cups/mime.types</filename> in place.
</para>

<para>
CUPS is widely configurable and flexible, even regarding its filtering
mechanism. Another workaround in some situations would be to have in
<filename moreinfo="none">/etc/cups/mime.types</filename> entries as follows:

<programlisting format="linespecific">
application/postscript           application/vnd.cups-raw  0  -
application/vnd.cups-postscript  application/vnd.cups-raw  0  -
</programlisting>

This would prevent all PostScript files from being filtered (rather,
they will through the virtual <emphasis>nullfilter</emphasis>
denoted with <quote>-</quote>). This could only be useful for PostScript printers. If you
want to print PostScript code on non-PostScript printers (provided they support ASCII
text printing), an entry as follows could be useful:

<programlisting format="linespecific">
*/*           application/vnd.cups-raw  0  -
</programlisting>

and would effectively send <emphasis>all</emphasis> files to the
backend without further processing.
</para>

<para>
You could have the following entry:

<programlisting format="linespecific">
application/vnd.cups-postscript application/vnd.cups-raw 0 \
	my_PJL_stripping_filter
</programlisting>

You will need to write a <parameter moreinfo="none">my_PJL_stripping_filter</parameter>
(which could be a shell script) that parses the PostScript and removes the
unwanted PJL. This needs to conform to CUPS filter design
(mainly, receive and pass the parameters printername, job-id,
username, jobtitle, copies, print options, and possibly the
filename). It is installed as world executable into
<filename moreinfo="none">/usr/lib/cups/filters/</filename> and is called by CUPS
if it encounters a MIME type <parameter moreinfo="none">application/vnd.cups-postscript</parameter>.
</para>

<para>
CUPS can handle <parameter moreinfo="none">-o job-hold-until=indefinite</parameter>.
This keeps the job in the queue on hold. It will only be printed
upon manual release by the printer operator. This is a requirement in
many central reproduction departments, where a few operators manage
the jobs of hundreds of users on some big machine, where no user is
allowed to have direct access (such as when the operators often need
to load the proper paper type before running the 10,000 page job
requested by marketing for the mailing, and so on).
</para>
</sect1>

<sect1>
<title>Autodeletion or Preservation of CUPS Spool Files</title>

<para>
<indexterm significance="normal"><primary>/var/spool/samba</primary></indexterm>
<indexterm significance="normal"><primary>/var/spool/cups/</primary></indexterm>
<indexterm significance="normal"><primary>cupsd.conf</primary></indexterm>
Samba print files pass through two spool directories. One is the incoming directory managed by Samba (set in
the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PATH" xlink:href="smb.conf.5.html#PATH">path = /var/spool/samba</link> directive in the <parameter>[printers]</parameter> section of <filename moreinfo="none">smb.conf</filename>). The other is the spool directory of your UNIX print subsystem. For
CUPS it is normally <filename moreinfo="none">/var/spool/cups/</filename>, as set by the <filename moreinfo="none">cupsd.conf</filename>
directive <filename moreinfo="none">RequestRoot /var/spool/cups</filename>.
</para>

<sect2>
<title>CUPS Configuration Settings Explained</title>

<para>
Some important parameter settings in the CUPS configuration file
<filename moreinfo="none">cupsd.conf</filename> are:
</para>

<variablelist>

	<varlistentry><term>PreserveJobHistory Yes</term>
	<listitem><para>
	This keeps some details of jobs in cupsd's mind (well, it keeps the
	c12345, c12346, and so on, files in the CUPS spool directory, which does a
	similar job as the old-fashioned BSD-LPD control files). This is set
	to <quote>Yes</quote> as a default.
	</para></listitem></varlistentry>

	<varlistentry><term>PreserveJobFiles Yes</term>
	<listitem><para>
	This keeps the job files themselves in cupsd's mind
	(it keeps the d12345, d12346, etc., files in the CUPS spool
	directory). This is set to <quote>No</quote> as the CUPS
	default.
	</para></listitem></varlistentry>

	<varlistentry><term><quote>MaxJobs 500</quote></term>
	<listitem><para>
	This directive controls the maximum number of jobs
	that are kept in memory. Once the number of jobs reaches the limit,
	the oldest completed job is automatically purged from the system to
	make room for the new one. If all of the known jobs are still
	pending or active, then the new job will be rejected. Setting the
	maximum to 0 disables this functionality. The default setting is
	0.
	</para></listitem></varlistentry>
</variablelist>

<para>
(There are also additional settings for <parameter moreinfo="none">MaxJobsPerUser</parameter> and
<parameter moreinfo="none">MaxJobsPerPrinter</parameter>.)
</para>
</sect2>

<sect2>
<title>Preconditions</title>

<para>
For everything to work as it should, you need to have three things:
</para>

<itemizedlist>
	<listitem><para>A Samba smbd that is compiled against <filename moreinfo="none">libcups</filename> (check
	on Linux by running <userinput moreinfo="none">ldd `which smbd'</userinput>).</para></listitem>

	<listitem><para>A Samba-<filename moreinfo="none">smb.conf</filename> setting of
			<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTING" xlink:href="smb.conf.5.html#PRINTING">printing = cups</link>.</para></listitem>

	<listitem><para>Another Samba <filename moreinfo="none">smb.conf</filename> setting of
			<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTCAP" xlink:href="smb.conf.5.html#PRINTCAP">printcap = cups</link>.</para></listitem>
</itemizedlist>

<note><para>
In this case, all other manually set printing-related commands (like
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTCOMMAND" xlink:href="smb.conf.5.html#PRINTCOMMAND">print command</link>, 
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LPQCOMMAND" xlink:href="smb.conf.5.html#LPQCOMMAND">lpq command</link>, 
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LPRMCOMMAND" xlink:href="smb.conf.5.html#LPRMCOMMAND">lprm command</link>, 
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LPPAUSECOMMAND" xlink:href="smb.conf.5.html#LPPAUSECOMMAND">lppause command</link>, and
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LPRESUMECOMMAND" xlink:href="smb.conf.5.html#LPRESUMECOMMAND">lpresume command</link>) are ignored, and they should normally have no
influence whatsoever on your printing.
</para></note>
</sect2>

<sect2>
<title>Manual Configuration</title>

<para>
If you want to do things manually, replace the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTING" xlink:href="smb.conf.5.html#PRINTING">printing = cups</link>
by <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTING" xlink:href="smb.conf.5.html#PRINTING">printing = bsd</link>. Then your manually set commands may work
(I haven't tested this), and a <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTCOMMAND" xlink:href="smb.conf.5.html#PRINTCOMMAND">print command = lp -d %P %s; rm %s</link>
may do what you need.
</para>
</sect2>
</sect1>

<sect1>
<title>Printing from CUPS to Windows-Attached Printers</title>

<para>
<indexterm significance="normal"><primary>smbspool</primary></indexterm>
<indexterm significance="normal"><primary>backends</primary></indexterm>
From time to time the question arises, how can you print <emphasis>to</emphasis> a Windows-attached printer
<emphasis>from</emphasis> Samba? Normally the local connection from Windows host to printer would be done by
USB or parallel cable, but this does not matter to Samba. From here only an SMB connection needs to be opened
to the Windows host. Of course, this printer must be shared first. As you have learned by now, CUPS uses
<emphasis>backends</emphasis> to talk to printers and other servers. To talk to Windows shared printers, you
need to use the <filename moreinfo="none">smb</filename> (surprise, surprise!) backend. Check if this is in the CUPS backend
directory. This usually resides in <filename moreinfo="none">/usr/lib/cups/backend/</filename>. You need to find an
<filename moreinfo="none">smb</filename> file there. It should be a symlink to <filename moreinfo="none">smbspool</filename>, and the file
must exist and be executable:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">ls -l /usr/lib/cups/backend/</userinput>
total 253
drwxr-xr-x    3 root   root     720 Apr 30 19:04 .
drwxr-xr-x    6 root   root     125 Dec 19 17:13 ..
-rwxr-xr-x    1 root   root   10692 Feb 16 21:29 canon
-rwxr-xr-x    1 root   root   10692 Feb 16 21:29 epson
lrwxrwxrwx    1 root   root       3 Apr 17 22:50 http -&gt; ipp
-rwxr-xr-x    1 root   root   17316 Apr 17 22:50 ipp
-rwxr-xr-x    1 root   root   15420 Apr 20 17:01 lpd
-rwxr-xr-x    1 root   root    8656 Apr 20 17:01 parallel
-rwxr-xr-x    1 root   root    2162 Mar 31 23:15 pdfdistiller
lrwxrwxrwx    1 root   root      25 Apr 30 19:04 ptal -&gt; /usr/sbin/ptal-cups
-rwxr-xr-x    1 root   root    6284 Apr 20 17:01 scsi
lrwxrwxrwx    1 root   root      17 Apr  2 03:11 smb -&gt; /usr/bin/smbspool
-rwxr-xr-x    1 root   root    7912 Apr 20 17:01 socket
-rwxr-xr-x    1 root   root    9012 Apr 20 17:01 usb

<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">ls -l `which smbspool`</userinput>
-rwxr-xr-x    1 root   root  563245 Dec 28 14:49 /usr/bin/smbspool
</screen></para>

<para>
If this symlink does not exist, create it:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">ln -s `which smbspool` /usr/lib/cups/backend/smb</userinput>
</screen></para>

<para>
<indexterm significance="normal"><primary>smbspool</primary></indexterm>
<indexterm significance="normal"><primary>troubleshooting</primary></indexterm>
<literal>smbspool</literal> was written by Mike Sweet from the CUPS folks. It is included and ships with
Samba. It may also be used with print subsystems other than CUPS, to spool jobs to Windows printer shares. To
set up printer <replaceable>winprinter</replaceable> on CUPS, you need to have a driver for it. Essentially
this means to convert the print data on the CUPS/Samba host to a format that the printer can digest (the
Windows host is unable to convert any files you may send). This also means you should be able to print to the
printer if it were hooked directly at your Samba/CUPS host. For troubleshooting purposes, this is what you
should do to determine if that part of the process chain is in order. Then proceed to fix the network
connection/authentication to the Windows host, and so on.
</para>

<para>
To install a printer with the <parameter moreinfo="none">smb</parameter> backend on CUPS, use this command:
</para>

<para><screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">lpadmin -p winprinter -v smb://WINDOWSNETBIOSNAME/printersharename \
  -P /path/to/PPD</userinput>
</screen></para>

<para>
<indexterm significance="normal"><primary>PostScript printers</primary></indexterm>
<indexterm significance="normal"><primary>PPD</primary></indexterm>
<indexterm significance="normal"><primary>Windows NT PostScript driver</primary></indexterm>
The PPD must be able to direct CUPS to generate the print data for the target model. For PostScript printers,
just use the PPD that would be used with the Windows NT PostScript driver. But what can you do if the printer
is only accessible with a password? Or if the printer's host is part of another workgroup? This is provided
for: You can include the required parameters as part of the <filename moreinfo="none">smb://</filename> device-URI like this:
</para>

<itemizedlist>
	<listitem><para><filename moreinfo="none">smb://WORKGROUP/WINDOWSNETBIOSNAME/printersharename</filename></para></listitem>
	<listitem><para><filename moreinfo="none">smb://username:password@WORKGROUP/WINDOWSNETBIOSNAME/printersharename</filename></para></listitem>
	<listitem><para><filename moreinfo="none">smb://username:password@WINDOWSNETBIOSNAME/printersharename</filename></para></listitem>
</itemizedlist>

<para>
Note that the device URI will be visible in the process list of the Samba server (e.g., when someone uses the
<literal>ps -aux</literal> command on Linux), even if the username and passwords are sanitized before they get
written into the log files. This is an inherently insecure option; however, it is the only one. Don't use it
if you want to protect your passwords. Better share the printer in a way that does not require a password!
Printing will only work if you have a working NetBIOS name resolution up and running. Note that this is a
feature of CUPS and you do not necessarily need to have smbd running.

</para>
</sect1>

<sect1>
<title>More CUPS Filtering Chains</title>

<para>
The diagrams in <link linkend="cups1">Filtering Chain 1</link> and <link linkend="cups2">Filtering Chain with
cupsomatic</link> show how CUPS handles print jobs.
</para>

<figure id="cups1" float="0">
	<title>Filtering Chain 1.</title>
	<mediaobject><imageobject role="html"><imagedata fileref="images/cups1.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/cups1.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/cups1"/></imageobject></mediaobject>
</figure>

<!-- JJJ -->
<figure id="cups2" float="0">
	<title>Filtering Chain with cupsomatic</title>
	<mediaobject><imageobject role="html"><imagedata fileref="images/cups2.png" scale="45" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/cups2.png" scale="45" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/cups2"/></imageobject></mediaobject>
</figure>

</sect1>

<sect1>
<title>Common Errors</title>

	<sect2>
	<title>Windows 9x/Me Client Can't Install Driver</title>

	<para>For Windows 9x/Me, clients require the printer names to be eight
	characters (or <quote>8 plus 3 chars suffix</quote>) max; otherwise, the driver files
	will not get transferred when you want to download them from Samba.</para>

	</sect2>

	<sect2 id="root-ask-loop">
	<title><quote>cupsaddsmb</quote> Keeps Asking for Root Password in Never-ending Loop</title>

	<para>Have you set <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SECURITY" xlink:href="smb.conf.5.html#SECURITY">security = user</link>? Have
	you used <literal>smbpasswd</literal> to give root a Samba account?
	You can do two things: open another terminal and execute
	<literal>smbpasswd -a root</literal> to create the account and
	continue entering the password into the first terminal. Or, break
	out of the loop by pressing Enter twice (without trying to type a
	password).</para>

	<para>
	If the error is <quote>Tree connect failed: NT_STATUS_BAD_NETWORK_NAME</quote>, 
	you may have forgotten to create the <filename moreinfo="none">/etc/samba/drivers</filename> directory.
	</para>
	</sect2>

	<sect2>
	<title><quote>cupsaddsmb</quote> or <quote>rpcclient addriver</quote> Emit Error</title>

	<para>
	If <literal>cupsaddsmb</literal>, or <literal>rpcclient addriver</literal> emit the error message
	WERR_BAD_PASSWORD, refer to <link linkend="root-ask-loop">the previous common error</link>.
	</para>
	
	</sect2>

	<sect2>
	<title><quote>cupsaddsmb</quote> Errors</title>

	<para>
	The use of <quote>cupsaddsmb</quote> gives <quote>No PPD file for printer...</quote> 
	message while PPD file is present.  What might the problem be?
	</para>

	<para>
	Have you enabled printer sharing on CUPS? This means, do you have a <literal moreinfo="none">&lt;Location
	/printers&gt;....&lt;/Location&gt;</literal> section in CUPS server's <filename moreinfo="none">cupsd.conf</filename> that
	does not deny access to the host you run <quote>cupsaddsmb</quote> from?  It <emphasis>could</emphasis> be an
	issue if you use cupsaddsmb remotely, or if you use it with a <option>-h</option> parameter:
	<userinput moreinfo="none">cupsaddsmb -H sambaserver -h cupsserver -v printername</userinput>.
	</para>

	<para>Is your <parameter moreinfo="none">TempDir</parameter> directive in
	<filename moreinfo="none">cupsd.conf</filename> set to a valid value, and is it writable?
	</para>

	</sect2>

	<sect2>
		<title>Client Can't Connect to Samba Printer</title>

	<para>Use <literal>smbstatus</literal> to check which user
	you are from Samba's point of view. Do you have the privileges to
	write into the <parameter>[print$]</parameter>
	share?</para>

	</sect2>

	<sect2>
	<title>New Account Reconnection from Windows 200x/XP Troubles</title>

<para>
Once you are connected as the wrong user (for example, as <constant>nobody</constant>, which often occurs if
you have <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="MAPTOGUEST" xlink:href="smb.conf.5.html#MAPTOGUEST">map to guest = bad user</link>), Windows Explorer will not accept an
attempt to connect again as a different user. There will not be any bytes transferred on the wire to Samba,
but still you'll see a stupid error message that makes you think Samba has denied access. Use
<literal>smbstatus</literal> to check for active connections. Kill the PIDs. You still can't re-connect, and
you get the dreaded <computeroutput moreinfo="none">You can't connect with a second account from the same
machine</computeroutput> message as soon as you try. And you do not see a single byte arriving at Samba (see
logs; use <quote>ethereal</quote>) indicating a renewed connection attempt. Shut all Explorer Windows.  This
makes Windows forget what it has cached in its memory as established connections. Then reconnect as the right
user. The best method is to use a DOS terminal window and <emphasis>first</emphasis> do <userinput moreinfo="none">net use z:
\\GANDALF\print$ /user:root</userinput>. Check with <literal>smbstatus</literal> that you are
connected under a different account. Now open the <guilabel moreinfo="none">Printers</guilabel> folder (on the Samba server in
the <guilabel moreinfo="none">Network Neighborhood</guilabel>), right-click on the printer in question, and select
<guibutton moreinfo="none">Connect....</guibutton>.
</para>
</sect2>

<sect2>
<title>Avoid Being Connected to the Samba Server as the Wrong User</title>
	
<para>
<indexterm significance="normal"><primary>smbstatus</primary></indexterm>
You see per <literal>smbstatus</literal> that you are connected as user nobody, but you want to be root or
printer admin. This is probably due to <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="MAPTOGUEST" xlink:href="smb.conf.5.html#MAPTOGUEST">map to guest = bad user</link>, which
silently connected you under the guest account when you gave (maybe by accident) an incorrect username. Remove
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="MAPTOGUEST" xlink:href="smb.conf.5.html#MAPTOGUEST">map to guest</link> if you want to prevent this.
</para>
</sect2>

<sect2>
<title>Upgrading to CUPS Drivers from Adobe Drivers</title>

<para>
This information came from a mailing list posting regarding problems experienced when
upgrading from Adobe drivers to CUPS drivers on Microsoft Windows NT/200x/XP clients.
</para>

<para>First delete all old Adobe-using printers. Then delete all old Adobe drivers. (On Windows 200x/XP, right-click in
the background of <guilabel moreinfo="none">Printers</guilabel> folder, select <guimenuitem moreinfo="none">Server Properties...</guimenuitem>, select
tab <guilabel moreinfo="none">Drivers</guilabel>, and delete here).</para>
</sect2>

<sect2>
<title>Can't Use <quote>cupsaddsmb</quote> on Samba Server, Which Is a PDC</title>

<para>Do you use the <quote>naked</quote> root user name? Try to do it
this way: <userinput moreinfo="none">cupsaddsmb -U <replaceable>DOMAINNAME</replaceable>\\root -v
<replaceable>printername</replaceable></userinput>&gt; (note the two backslashes: the first one is
required to <quote>escape</quote> the second one).</para>

</sect2>

<sect2>
<title>Deleted Windows 200x Printer Driver Is Still Shown</title>

<para>Deleting a printer on the client will not delete the
driver too (to verify, right-click on the white background of the
<guilabel moreinfo="none">Printers</guilabel> folder, select <guimenuitem moreinfo="none">Server Properties</guimenuitem> and click on the
<guilabel moreinfo="none">Drivers</guilabel> tab). These same old drivers will be re-used when you try to
install a printer with the same name. If you want to update to a new
driver, delete the old ones first. Deletion is only possible if no
other printer uses the same driver.</para>
</sect2>

<sect2>
<title>Windows 200x/XP Local Security Policies</title>

<indexterm significance="normal"><primary>Local security policies</primary></indexterm>
<indexterm significance="normal"><primary>unsigned drivers</primary></indexterm>
<para>Local security policies may not allow the installation of unsigned drivers <?latex --- ?> <quote>local
security policies</quote> may not allow the installation of printer drivers at all.</para>

</sect2>

<sect2>
<title>Administrator Cannot Install Printers for All Local Users</title>

<para>
<indexterm significance="normal"><primary>SMB printers</primary></indexterm>
<indexterm significance="normal"><primary>IPP client</primary></indexterm>
Windows XP handles SMB printers on a <quote>per-user</quote> basis.
This means every user needs to install the printer himself or herself. To have a printer available for
everybody, you might want to use the built-in IPP client capabilities of Win XP. Add a printer with the print
path of <parameter moreinfo="none">http://cupsserver:631/printers/printername</parameter>.  We're still looking into this one.
Maybe a logon script could automatically install printers for all users.
</para>

</sect2>

<sect2>
<title>Print Change, Notify Functions on NT Clients</title>

<para>For print change, notify functions on NT++ clients.  These need to run the <literal>Server</literal>
service first (renamed to <literal>File &amp; Print Sharing for MS Networks</literal> in XP).</para>

</sect2>

<sect2>
<title>Windows XP SP1</title>

<para>Windows XP SP1 introduced a Point and Print Restriction Policy (this restriction does not apply to
<quote>Administrator</quote> or <quote>Power User</quote> groups of users). In Group Policy Object Editor, go
to <guimenu moreinfo="none">User Configuration -&gt; Administrative Templates -&gt; Control Panel -&gt; Printers</guimenu>. The policy
is automatically set to <constant>Enabled</constant> and the <constant>Users can only Point and Print to
machines in their Forest</constant> . You probably need to change it to <constant>Disabled</constant> or
<constant>Users can only Point and Print to these servers</constant> to make driver downloads from Samba
possible.
</para>
</sect2>

<sect2>
<title>Print Options for All Users Can't Be Set on Windows 200x/XP</title>

<para>How are you doing it? I bet the wrong way (it is not easy to find out, though). There are three
different ways to bring you to a dialog that <emphasis>seems</emphasis> to set everything. All three dialogs
<emphasis>look</emphasis> the same, yet only one of them does what you intend. You need to be Administrator or
Print Administrator to do this for all users. Here is how I do it on XP:
</para>

<orderedlist numeration="upperalpha" inheritnum="ignore" continuation="restarts">

	<listitem><para>The first wrong way:

		<orderedlist inheritnum="ignore" continuation="restarts">
		<listitem><para>Open the <guilabel moreinfo="none">Printers</guilabel>
		folder.</para></listitem>

		<listitem><para>Right-click on the printer
		(<guilabel moreinfo="none">remoteprinter on cupshost</guilabel>) and
		select in context menu <guimenuitem moreinfo="none">Printing
		Preferences...</guimenuitem></para></listitem>.

		<listitem><para>Look at this dialog closely and remember what it looks like.</para></listitem>
		</orderedlist>
	</para></listitem>

	<listitem><para>The second wrong way:
	<orderedlist inheritnum="ignore" continuation="restarts">
		<listitem><para>Open the <guilabel moreinfo="none">Printers</guilabel> folder.</para></listitem>

		<listitem><para>Right-click on the printer (<guilabel moreinfo="none">remoteprinter on
		cupshost</guilabel>) and select the context menu
		<guimenuitem moreinfo="none">Properties</guimenuitem>.</para></listitem>

		<listitem><para>Click on the <guilabel moreinfo="none">General</guilabel> tab.</para></listitem>

		<listitem><para>Click on the button <guibutton moreinfo="none">Printing
		Preferences...</guibutton></para></listitem>.

		<listitem><para>A new dialog opens. Keep this dialog open and go back
		to the parent dialog.</para></listitem>
	</orderedlist>
	</para></listitem>

	<listitem><para>The third and correct way: 
	<orderedlist inheritnum="ignore" continuation="restarts">
		<listitem><para>Open the <guilabel moreinfo="none">Printers</guilabel> folder.</para></listitem>

		<listitem><para>Right-click on the printer (<guilabel moreinfo="none">remoteprinter on
		cupshost</guilabel>) and select the context menu
		<guimenuitem moreinfo="none">Properties</guimenuitem>.</para></listitem>

		<listitem><para>Click on the <guilabel moreinfo="none">Advanced</guilabel>
		tab. (If everything is <quote>grayed out,</quote> then you are not logged
		in as a user with enough privileges).</para></listitem>

		<listitem><para>Click on the <guibutton moreinfo="none">Printing
		Defaults...</guibutton> button.</para></listitem>

		<listitem><para>On any of the two new tabs, click on the
		<guibutton moreinfo="none">Advanced...</guibutton> button.</para></listitem>

		<listitem><para>A new dialog opens. Compare this one to the other
		identical-looking one from step <quote>B.5</quote> or A.3".</para></listitem>
	</orderedlist>
	</para></listitem>
</orderedlist>

<para>
Do you see any difference? I don't either. However, only the last one, which you arrived at with steps
<quote>C.1. to C.6.</quote>, will save any settings permanently and be the defaults for new users. If you want
all clients to get the same defaults, you need to conduct these steps <emphasis>as Administrator</emphasis>
(<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRINTERADMIN" xlink:href="smb.conf.5.html#PRINTERADMIN">printer admin</link> in <filename moreinfo="none">smb.conf</filename>) <emphasis>before</emphasis> a client downloads the
driver (the clients can later set their own <emphasis>per-user defaults</emphasis> by following the procedures
<emphasis>A</emphasis> or <emphasis>B</emphasis>).
</para>

</sect2>

<sect2>
<title>Most Common Blunders in Driver Settings on Windows Clients</title>

<para>
Don't use <parameter moreinfo="none">Optimize for Speed</parameter>, but use <parameter moreinfo="none">Optimize for Portability</parameter>
instead (Adobe PS Driver). Don't use <parameter moreinfo="none">Page Independence: No</parameter>. Always settle with
<parameter moreinfo="none">Page Independence: Yes</parameter> (Microsoft PS Driver and CUPS PS Driver for Windows NT/200x/XP).
If there are problems with fonts, use <parameter moreinfo="none">Download as Softfont into printer</parameter> (Adobe PS
Driver). For <guilabel moreinfo="none">TrueType Download Options</guilabel> choose <constant>Outline</constant>. Use
PostScript Level 2 if you are having trouble with a non-PS printer and if there is a choice.
</para>

</sect2>

<sect2>
<title><literal>cupsaddsmb</literal> Does Not Work with Newly Installed Printer</title>

<para>
Symptom: The last command of <literal>cupsaddsmb</literal> does not complete successfully. If the <literal>cmd
= setdriver printername printername</literal> result was NT_STATUS_UNSUCCESSFUL, then possibly the printer was
not yet recognized by Samba. Did it show up in Network Neighborhood? Did it show up in <literal>rpcclient
hostname -c `enumprinters'</literal>? Restart smbd (or send a <literal>kill -HUP</literal> to all processes
listed by <literal>smbstatus</literal>, and try again.
</para></sect2>

<sect2>
<title>Permissions on <filename moreinfo="none">/var/spool/samba/</filename> Get Reset After Each Reboot</title>

<para>
Have you ever by accident set the CUPS spool directory to the same location (<parameter moreinfo="none">RequestRoot
/var/spool/samba/</parameter> in <filename moreinfo="none">cupsd.conf</filename> or the other way round:
<filename moreinfo="none">/var/spool/cups/</filename> is set as <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PATH" xlink:href="smb.conf.5.html#PATH">path</link>&gt; in the <parameter>[printers]</parameter> section)? These <parameter moreinfo="none">must</parameter> be different. Set <parameter moreinfo="none">RequestRoot
/var/spool/cups/</parameter> in <filename moreinfo="none">cupsd.conf</filename> and <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PATH" xlink:href="smb.conf.5.html#PATH">path = 
/var/spool/samba</link> in the <parameter>[printers]</parameter> section of <filename moreinfo="none">smb.conf</filename>. Otherwise,
cupsd will sanitize permissions to its spool directory with each restart and printing will not work reliably.
</para>

</sect2>

<sect2>
<title>Print Queue Called <quote>lp</quote> Mishandles Print Jobs</title>

<para>
In this case a print queue called <quote>lp</quote> intermittently swallows jobs and
spits out completely different ones from what was sent.
</para>

<para>
<indexterm significance="normal"><primary>lp</primary></indexterm>
<indexterm significance="normal"><primary>Implicit Classes</primary></indexterm>
<indexterm significance="normal"><primary>BrowseShortNames</primary></indexterm>
It is a bad idea to name any printer <quote>lp</quote>. This is the traditional UNIX name for the default
printer. CUPS may be set up to do an automatic creation of Implicit Classes. This means, to group all printers
with the same name to a pool of devices and load-balance the jobs across them in a round-robin fashion.
Chances are high that someone else has a printer named <quote>lp</quote> too. You may receive that person's
jobs and send your own to his or her device unwittingly. To have tight control over the printer names, set
<parameter moreinfo="none">BrowseShortNames No</parameter>. It will present any printer as
<replaceable>printername@cupshost</replaceable>, which gives you better control over what may happen in a
large networked environment.
</para>

</sect2>

<sect2>
<title>Location of Adobe PostScript Driver Files for <quote>cupsaddsmb</quote></title>

<para>
Use <literal>smbclient</literal> to connect to any Windows box with a shared PostScript printer:
<literal>smbclient //windowsbox/print\$ -U guest</literal>. You can navigate to the
<filename moreinfo="none">W32X86/2</filename> subdir to <literal>mget ADOBE*</literal> and other files or to
<filename moreinfo="none">WIN40/0</filename> to do the same.  Another option is to download the <filename moreinfo="none">*.exe</filename>
packaged files from the Adobe Web site.
</para>

</sect2>

</sect1>

<sect1>
<title>Overview of the CUPS Printing Processes</title>

<para>
A complete overview of the CUPS printing processes can be found in <link linkend="a_small">the CUPS
Printing Overview diagram</link>.
</para>

<figure id="a_small" float="0">
	<title>CUPS Printing Overview.</title>
	<mediaobject><imageobject role="html"><imagedata fileref="images/a_small.png" scale="45" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/a_small.png" scale="45" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/a_small"/></imageobject></mediaobject>
</figure>
</sect1>

</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="VFS">
<chapterinfo>
	<author>
<firstname>Jelmer</firstname><othername>R.</othername><surname>Vernooij</surname><othername>R.</othername>
<affiliation>
	<orgname>The Samba Team</orgname>
	<address format="linespecific"><email>jelmer@samba.org</email></address>
</affiliation></author>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
	<author>
<firstname>Tim</firstname><surname>Potter</surname>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>tpot@samba.org</email></address>
</affiliation>
</author>
	<author><firstname>Simo</firstname><surname>Sorce</surname><contrib>original vfs_skel README</contrib></author>
	<author><firstname>Alexander</firstname><surname>Bokovoy</surname><contrib>original vfs_netatalk docs</contrib></author>
	<author><firstname>Stefan</firstname><surname>Metzmacher</surname><contrib>Update for multiple modules</contrib></author>
	<author><firstname>Ed</firstname><surname>Riddle</surname><contrib>original shadow_copy docs</contrib></author>
</chapterinfo>
<title>Stackable VFS modules</title>

<sect1>
<title>Features and Benefits</title>

<para>
<indexterm significance="normal"><primary>Virtual File System</primary><see>VFS</see></indexterm>
<indexterm significance="normal"><primary>modules</primary></indexterm>
<indexterm significance="normal"><primary>loaded modules</primary></indexterm>
Stackable VFS (Virtual File System) modules support was new to Samba-3 and has proven quite popular. Samba
passes each request to access the UNIX file system through the loaded VFS modules. This chapter covers the
modules that come with the Samba source and provides references to some external modules.
</para>


</sect1>

<sect1>
<title>Discussion</title>

<para>
<indexterm significance="normal"><primary>IRIX</primary></indexterm>
<indexterm significance="normal"><primary>GNU/Linux</primary></indexterm>
If not supplied with your platform distribution binary Samba package, you may have problems compiling these
modules, as shared libraries are compiled and linked in different ways on different systems. They currently
have been tested against GNU/Linux and IRIX.
</para>

<para>
<indexterm significance="normal"><primary>VFS modules</primary></indexterm>
<indexterm significance="normal"><primary>modules</primary></indexterm>
<indexterm significance="normal"><primary>recycle bin</primary></indexterm>
To use the VFS modules, create a share similar to the one below. The important parameter is the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="VFSOBJECTS" xlink:href="smb.conf.5.html#VFSOBJECTS">vfs objects</link> parameter where you can list one or more VFS modules by name. For example, to log all
access to files and put deleted files in a recycle bin, see <link linkend="vfsrecyc">the smb.conf with VFS
modules example</link>:
</para>

<example id="vfsrecyc">
<title>smb.conf with VFS modules</title>
<simplelist>
<member> </member><member><parameter>[audit]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = Audited /data directory</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /data</parameter></member>
<member><indexterm>vfs objects</indexterm><parameter>vfs objects = audit recycle</parameter></member>
<member><indexterm>writeable</indexterm><parameter>writeable = yes</parameter></member>
<member><indexterm>browseable</indexterm><parameter>browseable = yes</parameter></member>
</simplelist>
</example>

<para>
<indexterm significance="normal"><primary>virus scanner</primary></indexterm>
<indexterm significance="normal"><primary>scanner module</primary></indexterm>
<indexterm significance="normal"><primary>recycle bin</primary></indexterm>
The modules are used in the order in which they are specified.  Let's say that you want to both have a virus
scanner module and a recycle bin module. It is wise to put the virus scanner module as the first one so that
it is the first to get run and may detect a virus immediately, before any action is performed on that file.
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="VFSOBJECTS" xlink:href="smb.conf.5.html#VFSOBJECTS">vfs objects = vscan-clamav recycle</link>
</para>

<para>
<indexterm significance="normal"><primary>/usr/local/samba/lib/vfs</primary></indexterm>
<indexterm significance="normal"><primary>/usr/lib/samba/vfs</primary></indexterm>
Samba will attempt to load modules from the <filename moreinfo="none">/lib</filename> directory in the root directory of the
Samba installation (usually <filename moreinfo="none">/usr/lib/samba/vfs</filename> or
<filename moreinfo="none">/usr/local/samba/lib/vfs</filename>).
</para>

<para>
<indexterm significance="normal"><primary>modules</primary></indexterm>
<indexterm significance="normal"><primary>VFS</primary></indexterm>
<indexterm significance="normal"><primary>multiple modules</primary></indexterm>
<indexterm significance="normal"><primary>multiple VFS</primary></indexterm>
Some modules can be used twice for the same share.  This can be done using a configuration similar to the one
shown in <link linkend="multimodule">the smb.conf with multiple VFS modules</link>.

<example id="multimodule">
<title>smb.conf with multiple VFS modules</title>
<simplelist>
<member> </member><member><parameter>[test]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = VFS TEST</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /data</parameter></member>
<member><indexterm>writeable</indexterm><parameter>writeable = yes</parameter></member>
<member><indexterm>browseable</indexterm><parameter>browseable = yes</parameter></member>
<member><indexterm>vfs objects</indexterm><parameter>vfs objects = example:example1 example example:test</parameter></member>
<member><indexterm>example1: parameter</indexterm><parameter>example1: parameter = 1</parameter></member>
<member><indexterm>example:  parameter</indexterm><parameter>example:  parameter = 5</parameter></member>
<member><indexterm>test:     parameter</indexterm><parameter>test:     parameter = 7</parameter></member>
</simplelist>
</example>
</para>

</sect1>

<sect1>
<title>Included Modules</title>

	<sect2>
	<title>audit</title>

		<para>
<indexterm significance="normal"><primary>audit file access</primary></indexterm>
		A simple module to audit file access to the syslog facility. The following operations are logged:
		<itemizedlist>
			<listitem><para>share</para></listitem>
			<listitem><para>connect/disconnect</para></listitem>
			<listitem><para>directory opens/create/remove</para></listitem>
			<listitem><para>file open/close/rename/unlink/chmod</para></listitem>
		</itemizedlist>
		</para>

	</sect2>

	<sect2>
	<title>default_quota</title>

	<para>
	This module allows the default quota values, in the windows explorer GUI, to be stored on a Samba-3 server.
	The challenge is that linux filesystems only store quotas for users and groups, but no default quotas.
	</para>

	<para>
	Samba returns NO_LIMIT as the default quotas by default and refuses to update them. With this module you 
	can store the default quotas that are reported to a windows client, in the quota record of a user. By
	default the root user is taken because quota limits for root are typically not enforced.
	</para>

	<para>
	This module takes 2 parametric entries in the <filename moreinfo="none">smb.conf</filename> file.  The default prefix for each is the
	<quote>default_quota</quote>. This can be overwrittem when you load the module in the <emphasis>vfs
	modules</emphasis> parameter like this:
<screen format="linespecific">
vfs objects = default_quota:myprefix
</screen>
	</para>

	<para>
	The parametric entries that may be specified for the default_quotas module are:
	</para>

	<variablelist>
		<varlistentry>
            <term>myprefix:uid</term>
			<listitem><para>
			This parameter takes a integer argument that specifies the uid of the quota record that will be 
			used for storing the default user quotas.
			</para>

			<para>
			The default value is 0 (for root user). An example of use is:
<screen format="linespecific">
vfs objects = default_quota
default_quota:	uid = 65534
</screen>
			The above demonstrates the case where the <constant>myprefix</constant> was omitted, thus the
			default prefix is the name of the module. When a <constant>myprefix</constant> parameter is
			specified the above can be re-written like this:
<screen format="linespecific">
vfs objects = default_quota:myprefix
myprefix:	uid = 65534
</screen>
			</para></listitem>
		</varlistentry>

		<varlistentry>
            <term>myprefix:uid nolimit</term>
			<listitem><para>
			This parameter takes a boolean argument that specifies if the stored default quota values also be
			reported for the user record, or if the value <constant>NO_LIMIT</constant> should be reported to 
			the windows client for the user specified by the <parameter moreinfo="none">prefix:uid</parameter> parameter.
			</para>

			<para>
			The default value is <constant>yes</constant> (which means to report NO_LIMIT). An example of use
			is shown here:
<screen format="linespecific">
vfs objects = default_quota:myprefix
myprefix:	uid nolimit = no
</screen>
			</para></listitem>
		</varlistentry>

		<varlistentry>
			<term>myprefix:gid</term>
			<listitem><para>
			This parameter takes an integer argument, it's just like the <parameter moreinfo="none">prefix&gt;:uid</parameter> but 
			for group quotas.  NOTE: group quotas are not supported from the windows explorer.
			</para>

			<para>
			The default value is 0 (for root group). An example of use is shown here:
<screen format="linespecific">
vfs objects = default_quota
default_quota:	gid = 65534
</screen>
			</para></listitem>
		</varlistentry>

		<varlistentry>
			<term>myprefix:gid nolimit</term>
			<listitem><para>
			This parameter takes a boolean argument, just like the <parameter moreinfo="none">prefix&gt;:uid nolimit</parameter> 
			but for group quotas.  NOTE: group quotas are not supported from the windows explorer.
			</para>

			<para>
			The default value is <constant>yes</constant> (which means to report NO_LIMIT). An example of use
			is shown here:
<screen format="linespecific">
vfs objects = default_quota
default_quota:	uid nolimit = no
</screen>
			</para></listitem>
		</varlistentry>
	</variablelist>

	<para>
	An example of use of multiple parametric specifications is shown here:
<screen format="linespecific">
...
vfs objects = default_quota:quotasettings
quotasettings:	uid nolimit = no
quotasettings:	gid = 65534
quotasettings:	gid nolimit = no
...
</screen>
	</para>

	</sect2>

	<sect2>
	<title>extd_audit</title>

		<para>
<indexterm significance="normal"><primary>audit module</primary></indexterm>
<indexterm significance="normal"><primary>extd_audit module</primary></indexterm>
<indexterm significance="normal"><primary>smbd</primary></indexterm>
		This module is identical with the <literal>audit</literal> module above except
		that it sends audit logs to both syslog as well as the <literal>smbd</literal> log files. The 
		<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOGLEVEL" xlink:href="smb.conf.5.html#LOGLEVEL">log level</link> for this module is set in the <filename moreinfo="none">smb.conf</filename> file. 
		</para>

		<para>
		Valid settings and the information that will be recorded are shown in <link linkend="xtdaudit">the next table</link>.
		</para>

		<table frame="all" id="xtdaudit">
			<title>Extended Auditing Log Information</title>
		<tgroup cols="2" align="center">
			<thead>
			<row><entry align="center">Log Level</entry><entry>Log Details - File and Directory Operations</entry></row>
			</thead>
			<tbody>
			<row><entry align="center">0</entry><entry align="left">Make Directory, Remove Directory, Unlink</entry></row>
			<row><entry align="center">1</entry><entry align="left">Open Directory, Rename File, Change Permissions/ACLs</entry></row>
			<row><entry align="center">2</entry><entry align="left">Open &amp; Close File</entry></row>
			<row><entry align="center">10</entry><entry align="left">Maximum Debug Level</entry></row>
			</tbody>
		</tgroup>
		</table>

		<sect3>
		<title>Configuration of Auditing</title>

		<para>
<indexterm significance="normal"><primary>logging</primary></indexterm>
		This auditing tool is more felxible than most people readily will recognize. There are a number of ways
		by which useful logging information can be recorded.
		</para>

		<itemizedlist>
			<listitem><para>Syslog can be used to record all transaction. This can be disabled by setting
					in the <filename moreinfo="none">smb.conf</filename> file <parameter moreinfo="none">syslog = 0</parameter>.</para></listitem>
			<listitem><para>Logging can take place to the default log file (<filename moreinfo="none">log.smbd</filename>)
					for all loaded VFS modules just by setting in the <filename moreinfo="none">smb.conf</filename> file
					<parameter moreinfo="none">log level = 0 vfs:x</parameter>, where x is the log level.
					This will disable general logging while activating all logging of VFS
					module activity at the log level specified.</para></listitem>
			<listitem><para>Detailed logging can be obtained per user, per client machine, etc.
					This requires the above together with the creative use of the
					<parameter moreinfo="none">log file</parameter> settings.</para>
					<para>An example of detailed per-user and per-machine logging can
					be obtained by setting 
					<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOGFILE" xlink:href="smb.conf.5.html#LOGFILE">log file = /var/log/samba/%U.%m.log</link>.
					</para></listitem>
		</itemizedlist>

		<para>
		Auditing information often must be preserved for a long time. So that the log files do not get rotated
		it is essential that the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="MAXLOGSIZE" xlink:href="smb.conf.5.html#MAXLOGSIZE">max log size = 0</link> be set
		in the <filename moreinfo="none">smb.conf</filename> file.
		</para>

		</sect3>

	</sect2>

	<sect2 id="fakeperms">
	<title>fake_perms</title>

		<para>
<indexterm significance="normal"><primary>fake_perms</primary></indexterm>
<indexterm significance="normal"><primary>Roaming Profile</primary></indexterm>
<indexterm significance="normal"><primary>writeable</primary></indexterm>
<indexterm significance="normal"><primary>read only</primary></indexterm>
		This module was created to allow Roaming Profile files and directories to be set (on the Samba server
		under UNIX) as read only. This module will, if installed on the Profiles share, report to the client
		that the Profile files and directories are writeable. This satisfies the client even though the files
		will never be overwritten as the client logs out or shuts down.
		</para>

	</sect2>

	<sect2>
	<title>recycle</title>

		<para>
<indexterm significance="normal"><primary>recycle</primary></indexterm>
<indexterm significance="normal"><primary>unlink calls</primary></indexterm>
<indexterm significance="normal"><primary>recycle directory</primary></indexterm>
		A Recycle Bin-like module. Where used, unlink calls will be intercepted and files moved
		to the recycle directory instead of being deleted. This gives the same effect as the
		<guiicon moreinfo="none">Recycle Bin</guiicon> on Windows computers.
		</para>

		<para>
<indexterm significance="normal"><primary>recycle</primary></indexterm>
<indexterm significance="normal"><primary>.recycle</primary></indexterm>
<indexterm significance="normal"><primary>recycle:keeptree</primary></indexterm>
<indexterm significance="normal"><primary>deleted files</primary></indexterm>
		The <guiicon moreinfo="none">Recycle Bin</guiicon> will not appear in
		<application moreinfo="none">Windows Explorer</application> views of the network
		file system (share) nor on any mapped drive. Instead, a directory
		called <filename moreinfo="none">.recycle</filename> will be automatically created
		when the first file is deleted and <parameter moreinfo="none">recycle:repository</parameter>
		is not configured.
		If <parameter moreinfo="none">recycle:repository</parameter> is configured, the name
		of the created directory depends on <parameter moreinfo="none">recycle:repository</parameter>.
		Users can recover files from the recycle bin. If the
		<parameter moreinfo="none">recycle:keeptree</parameter> has been specified,	deleted
		files will be found in a path identical with that from which the
		file was deleted.
		</para>
		
		<para>Supported options for the <literal>recycle</literal> module are as follow:
		<variablelist>
			<varlistentry>
			<term>recycle:repository</term>
				<listitem><para>
<indexterm significance="normal"><primary>recycle:repository</primary></indexterm>
				Path of the directory where deleted files should be moved.
				</para></listitem>
			</varlistentry>

			<varlistentry>
			<term>recycle:directory_mode</term>
				<listitem><para>
<indexterm significance="normal"><primary>directory_mode</primary></indexterm>
				Set it to the octal mode you want for the recycle directory. With
				this mode the recycle directory	will be created if it not
				exists and the first file is deleted.
				If <parameter moreinfo="none">recycle:subdir_mode</parameter> is not set, these
				mode also apply to sub directories.
				If <parameter moreinfo="none">directory_mode</parameter> not exists, the default
				mode 0700 is used.
				</para></listitem>
			</varlistentry>

			<varlistentry>
			<term>recycle:subdir_mode</term>
				<listitem><para>
<indexterm significance="normal"><primary>recycle:subdir_mode</primary></indexterm>
				Set it to the octal mode you want for the sub directories of
				the recycle directory. With this mode	the sub directories will
				be created.
				If <parameter moreinfo="none">recycle:subdir_mode</parameter> is not set, the
				sub directories will be created with the mode from
				<parameter moreinfo="none">directory_mode</parameter>.
				</para></listitem>
			</varlistentry>

			<varlistentry>
			<term>recycle:keeptree</term>
				<listitem><para>
<indexterm significance="normal"><primary>recycle:keeptree</primary></indexterm>
				Specifies whether the directory structure should be kept or if the files in the directory that is being 
				deleted should be kept separately in the recycle bin.
				</para></listitem>
			</varlistentry>
			
			<varlistentry>
			<term>recycle:versions</term>
				<listitem><para>
<indexterm significance="normal"><primary>recycle:versions</primary></indexterm>
				If this option is set, two files 
				with the same name that are deleted will both 
				be kept in the recycle bin. Newer deleted versions 
				of a file will be called <quote>Copy #x of <replaceable>filename</replaceable></quote>.
				</para></listitem>
			</varlistentry>

			<varlistentry>
			<term>recycle:touch</term>
				<listitem><para>
<indexterm significance="normal"><primary>recycle:touch</primary></indexterm>
				Specifies whether a file's access date should be touched when the file is moved to the recycle bin.
				</para></listitem>
			</varlistentry>

			<varlistentry>
			<term>recycle:touch_mtime</term>
				<listitem><para>
<indexterm significance="normal"><primary>recycle:touch</primary></indexterm>
				Specifies whether a file's last modify date date should be touched when the file is moved to the recycle bin.
				</para></listitem>
			</varlistentry>

			<varlistentry>
			<term>recycle:maxsize</term>
				<listitem><para>
<indexterm significance="normal"><primary>recycle:maxsize</primary></indexterm>
				Files that are larger than the number of bytes specified by this parameter will not be put into the recycle bin.
				</para></listitem>
			</varlistentry>

			<varlistentry>
			<term>recycle:exclude</term>
				<listitem><para>
<indexterm significance="normal"><primary>recycle:exclude</primary></indexterm>
				List of files that should not be put into the recycle bin when deleted, but deleted in the regular way.
				</para></listitem>
			</varlistentry>

			<varlistentry>
			<term>recycle:exclude_dir</term>
				<listitem><para>
<indexterm significance="normal"><primary>recycle:exclude_dir</primary></indexterm>
				Contains a list of directories. When files from these directories are
				deleted, they are not put into the
				recycle bin but are deleted in the
				regular way.
				</para></listitem>
			</varlistentry>

			<varlistentry>
			<term>recycle:noversions</term>
				<listitem><para>
<indexterm significance="normal"><primary>recycle:noversions</primary></indexterm>
				Specifies a list of paths (wildcards such as * and ? are supported) for which no versioning
				should be used. Only useful when <emphasis>recycle:versions</emphasis> is enabled.
				</para></listitem>
			</varlistentry>
		</variablelist>
		</para>

	</sect2>

	<sect2>
	<title>netatalk</title>

		<para>
<indexterm significance="normal"><primary>netatalk</primary></indexterm>
		A netatalk module will ease co-existence of Samba and netatalk file sharing services.
		</para>

		<para>Advantages compared to the old netatalk module:
		<itemizedlist>
<indexterm significance="normal"><primary>.AppleDouble</primary></indexterm>
			<listitem><para>Does not care about creating .AppleDouble forks, just keeps them in sync.</para></listitem>
			<listitem><para>If a share in <filename moreinfo="none">smb.conf</filename> does not contain .AppleDouble item in hide or veto list, it will be added automatically.</para></listitem>
		</itemizedlist>
		</para>

	</sect2>

    <sect2>
      <title>shadow_copy</title>

	<warning><para>
<indexterm significance="normal"><primary>shadow_copy</primary></indexterm>
	<emphasis>THIS IS NOT A BACKUP, ARCHIVAL, OR VERSION CONTROL SOLUTION!</emphasis>
	</para>

	<para>
<indexterm significance="normal"><primary>version control</primary></indexterm>
	With Samba or Windows servers, shadow_copy is designed to be an end-user tool only.  It does not replace or
	enhance your backup and archival solutions and should in no way be considered as such.  Additionally, if you
	need version control, implement a version control system.  You have been warned.
	</para></warning>


	<para>
	The shadow_copy module allows you to setup functionality that is similar to MS shadow copy services.  When
	setup properly, this module allows Microsoft shadow copy clients to browse "shadow copies" on Samba shares.
	You will need to install the shadow copy client.  You can get the MS shadow copy client <ulink noescape="1" url="http://www.microsoft.com/windowsserver2003/downloads/shadowcopyclient.mspx">here.</ulink>.  Note the
	additional requirements for pre-Windows XP clients.  I did not test this functionality with any pre-Windows XP
	clients.  You should be able to get more information about MS Shadow Copy <ulink noescape="1" url="http://www.microsoft.com/windowsserver2003/techinfo/overview/scr.mspx">from the Microsoft's site</ulink>.
	</para>

	<para>
<indexterm significance="normal"><primary>shadow_copy</primary></indexterm>
<indexterm significance="normal"><primary>VFS module</primary></indexterm>
<indexterm significance="normal"><primary>shadow_copy module</primary></indexterm>
<indexterm significance="normal"><primary>LVM</primary></indexterm>
<indexterm significance="normal"><primary>EVMS</primary></indexterm>
<indexterm significance="normal"><primary>Logical Volume Manager</primary><see>LVM</see></indexterm>
	The shadow_copy VFS module requires some underlying file system setup with some sort of Logical Volume Manager
	(LVM) such as LVM1, LVM2, or EVMS.  Setting up LVM is beyond the scope of this document; however, we will
	outline the steps we took to test this functionality for <emphasis>example purposes only.</emphasis> You need
	to make sure the LVM implementation you choose to deploy is ready for production.  Make sure you do plenty of
	tests.
	</para>

	<para>
	Here are some common resources for LVM and EVMS:
	</para>

	<itemizedlist>
	  <listitem>
	    <para><ulink noescape="1" url="http://www.sistina.com/products_lvm_download.htm">Sistina's
	    LVM1 and LVM2</ulink></para>
	  </listitem>
	  <listitem>
	    <para><ulink url="http://evms.sourceforge.net/">Enterprise Volume Management System (EVMS)</ulink></para>
	  </listitem>
	  <listitem>
	    <para><ulink url="http://tldp.org/HOWTO/LVM-HOWTO/">The LVM HOWTO</ulink></para>
	  </listitem>
	  <listitem>
	    <para>
	      See <ulink url="http://www-106.ibm.com/developerworks/linux/library/l-lvm/">Learning
	      Linux LVM, Part 1</ulink> and <ulink url="http://www-106.ibm.com/developerworks/library/l-lvm2.html">Learning
	      Linux LWM, Part 2</ulink> for Daniel Robbins' well-written, two part tutorial on Linux and LVM using LVM
	      source code and reiserfs.</para>
	  </listitem>
	</itemizedlist>

	<sect3>
	<title>Shadow Copy Setup</title>
	<para>
<indexterm significance="normal"><primary>XFS file system</primary></indexterm>
<indexterm significance="normal"><primary>Debian Sarge</primary></indexterm>
	At the time of this writing, not much testing has been done.  I tested the shadow copy VFS module with a
	specific scenario which was not deployed in a production environment, but more as a proof of concept.  The
	scenario involved a Samba-3 file server on Debian Sarge with an XFS file system and LVM1.  I do NOT recommend
	you use this as a solution without doing your own due diligence with regard to all the components presented
	here.  That said, following is an basic outline of how I got things going.
	</para>

	<orderedlist inheritnum="ignore" continuation="restarts">
	  <listitem>
	    <formalpara><title>Installed Operating System </title> 
	    <para>
		In my tests, I used <ulink url="http://www.debian.org/devel/debian-installer/">Debian
		Sarge</ulink> (i.e., testing) on an XFS file system.  Setting up the OS is a bit beyond the scope of this
		document.  It is assumed that you have a working OS capable of running Samba.
		</para></formalpara>
	  </listitem>

	  <listitem>
	    <formalpara><title>Install &amp; Configure Samba</title>
		<para>
		See the <link linkend="introduction">installation section</link> of this HOWTO for more detail on this.
		It doesn't matter if it is a Domain Controller or Member File Server, but it is assumed that you have a
		working Samba 3.0.3 or later server running.
		</para></formalpara>
	  </listitem>

	  <listitem>
	    <formalpara><title>Install &amp; Configure LVM</title>
		<para>
<indexterm significance="normal"><primary>shadow copies</primary></indexterm>
<indexterm significance="normal"><primary>Snapshots</primary></indexterm>
		Before you can make shadow copies available to the client, you have to create the shadow copies.  This is
		done by taking some sort of file system snapshot.  Snapshots are a typical feature of Logical Volume
		Managers such as LVM, so we first need to have that setup.
		</para></formalpara>

	    <itemizedlist>
		<para>
		The following is provided as an example and will be most helpful for Debian users.  Again, this was tested
		using the "testing" or "Sarge" distribution.
		</para>

			<listitem>
			<para>
<indexterm significance="normal"><primary>lvm10 package</primary></indexterm>
<indexterm significance="normal"><primary>devfsd package</primary></indexterm>
<indexterm significance="normal"><primary>Debian</primary></indexterm>
<indexterm significance="normal"><primary>xfsprogs</primary></indexterm>
<indexterm significance="normal"><primary>apt-get</primary></indexterm>
			Install lvm10 and devfsd packages if you have not done so already.  On Debian systems, you are warned of the
			interaction of devfs and lvm1 which requires the use of devfs filenames.  Running <literal>apt-get update
			&amp;&amp; apt-get install lvm10 devfsd xfsprogs</literal> should do the trick for this example.
			</para></listitem>

			<listitem><para>
<indexterm significance="normal"><primary>create volume</primary></indexterm>
<indexterm significance="normal"><primary>create partition</primary></indexterm>
<indexterm significance="normal"><primary>fdisk</primary></indexterm>
<indexterm significance="normal"><primary>cfdisk</primary></indexterm>
<indexterm significance="normal"><primary>Linux LVM</primary></indexterm>
			Now you need to create a volume.  You will need to create a partition (or partitions) to add to your volume.
			Use your favorite partitioning tool (e.g., Linux fdisk, cfdisk, etc.).  The partition type should be set to
			0x8e for "Linux LVM."  In this example, we will use /dev/hdb1.
			</para>

			<para>
<indexterm significance="normal"><primary>Linux LVM partition</primary></indexterm>
<indexterm significance="normal"><primary>LVM volume</primary></indexterm>
<indexterm significance="normal"><primary>modprobe</primary></indexterm>
			Once you have the Linux LVM partition (type 0x8e), you can run a series of commands to create the LVM volume.
			You can use several disks and/or partitions, but we will use only one in this example.  You may also need to
			load the kernel module with something like <literal>modprobe lvm-mod</literal> and set your system up to load
			it on reboot by adding it to (<filename moreinfo="none">/etc/modules</filename>).
			</para></listitem>

			<listitem><para>
<indexterm significance="normal"><primary>pvcreate</primary></indexterm>
			Create the physical volume with <literal>pvcreate /dev/hdb1</literal>
			</para></listitem>

			<listitem><para>
<indexterm significance="normal"><primary>vgcreate</primary></indexterm>
<indexterm significance="normal"><primary>volume group</primary></indexterm>
			Create the volume group and add /dev/hda1 to it with <literal>vgcreate shadowvol /dev/hdb1</literal>
			</para>

			<para>
<indexterm significance="normal"><primary>vgdisplay</primary></indexterm>
			You can use <literal>vgdisplay</literal> to review information about the volume group.
			</para></listitem>

			<listitem><para>
<indexterm significance="normal"><primary>lvcreate</primary></indexterm>
			Now you can create the logical volume with something like <literal>lvcreate -L400M -nsh_test shadowvol</literal>
			</para>

			<para>
<indexterm significance="normal"><primary>/dev/shadowvol</primary></indexterm>
			This creates the logical volume of 400 MBs named "sh_test" in the volume group we created called shadowvol.
			If everything is working so far, you should see them in <filename moreinfo="none">/dev/shadowvol</filename>.
			</para></listitem>

			<listitem><para>
<indexterm significance="normal"><primary>mkfs.xfs</primary></indexterm>
			Now we should be ready to format the logical volume we named sh_test with <literal>mkfs.xfs
			/dev/shadowvol/sh_test</literal>
			</para>

			<para>
<indexterm significance="normal"><primary>logical volume</primary></indexterm>
<indexterm significance="normal"><primary>LVM</primary></indexterm>
<indexterm significance="normal"><primary>freezing</primary></indexterm>
<indexterm significance="normal"><primary>resizing</primary></indexterm>
<indexterm significance="normal"><primary>growing</primary></indexterm>
			You can format the logical volume with any file system you choose, but make sure to use one that allows you to
			take advantage of the additional features of LVM such as freezing, resizing, and growing your file systems.
			</para>

			<para>
<indexterm significance="normal"><primary>LVM volume</primary></indexterm>
<indexterm significance="normal"><primary>shadow_copy</primary></indexterm>
<indexterm significance="normal"><primary>module</primary></indexterm>
			Now we have an LVM volume where we can play with the shadow_copy VFS module.
			</para></listitem>

			<listitem><para>
<indexterm significance="normal"><primary>mkdir</primary></indexterm>
<indexterm significance="normal"><primary>permissions</primary></indexterm>
<indexterm significance="normal"><primary>chmod</primary></indexterm>
			Now we need to prepare the directory with something like
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> mkdir -p /data/shadow_share
</screen>
			or whatever you want to name your shadow copy-enabled Samba share.  Make sure you set the permissions so that
			you can use it.  If in doubt, use <literal>chmod 777 /data/shadow_share</literal> and tighten the permissions
			once you get things working.
			</para></listitem>

			<listitem><para>
<indexterm significance="normal"><primary>mount</primary></indexterm>
			Mount the LVM volume using something like <literal>mount /dev/shadowvol/sh_test /data/shadow_share</literal>
			</para>

			<para>
<indexterm significance="normal"><primary>/etc/fstab</primary></indexterm>
			You may also want to edit your <filename moreinfo="none">/etc/fstab</filename> so that this partition mounts during the system boot.
			</para></listitem>
		</itemizedlist>

		</listitem>

	  <listitem>
	    <formalpara><title>Install &amp; Configure the shadow_copy VFS Module</title>
		<para>
		Finally we get to the actual shadow_copy VFS module.  The shadow_copy VFS module should be available in Samba
		3.0.3 and higher.  The smb.conf configuration is pretty standard.  Here is our example of a share configured
		with the shadow_copy VFS module:
		</para></formalpara>

		<example id="vfsshadow">
		<title>Share With shadow_copy VFS</title>
		<simplelist>
		<member> </member><member><parameter>[shadow_share]</parameter></member>
		<member><indexterm>comment</indexterm><parameter>comment = Shadow Copy Enabled Share</parameter></member>
		<member><indexterm>path</indexterm><parameter>path = /data/shadow_share</parameter></member>
		<member><indexterm>vfs objects</indexterm><parameter>vfs objects = shadow_copy</parameter></member>
		<member><indexterm>writeable</indexterm><parameter>writeable = yes</parameter></member>
		<member><indexterm>browseable</indexterm><parameter>browseable = yes</parameter></member>
		</simplelist>
		</example>

		</listitem>

		<listitem>
	    <formalpara><title>Create Snapshots and Make Them Available to shadow_copy.so</title> 
		<para>
<indexterm significance="normal"><primary>shadow_copy</primary></indexterm>
<indexterm significance="normal"><primary>LVM snapshots</primary></indexterm>
<indexterm significance="normal"><primary>module</primary></indexterm>
		Before you can browse the shadow copies, you must create them and mount them.  This will most likely be done
		with a script that runs as a cron job.  With this particular solution, the shadow_copy VFS module is used to
		browse LVM snapshots.  Those snapshots are not created by the module.  They are not made available by the
		module either.  This module allows the shadow copy-enabled client to browse the snapshots you take and make
		available.
		</para></formalpara>

	    <para>
		Here is a simple script used to create and mount the snapshots:
<screen format="linespecific">
#!/bin/bash
# This is a test, this is only a test
SNAPNAME=`date +%Y.%m.%d-%H.%M.%S`
xfs_freeze -f /data/shadow_share/
lvcreate -L10M -s -n $SNAPNAME /dev/shadowvol/sh_test
xfs_freeze -u /data/shadow_share/
mkdir /data/shadow_share/@GMT-$SNAPNAME
mount /dev/shadowvol/$SNAPNAME \
       /data/shadow_share/@GMT-$SNAPNAME -onouuid,ro
</screen>
		Note that the script does not handle other things like remounting snapshots on reboot.
	    </para></listitem>

		<listitem>
	    <formalpara><title>Test From Client</title>
		<para>
		To test, you will need to install the shadow copy client which you can obtain from the <ulink url="http://www.microsoft.com/windowsserver2003/downloads/shadowcopyclient.mspx">Microsoft web site.</ulink> I
		only tested this with an XP client so your results may vary with other pre-XP clients.  Once installed, with
		your XP client you can right-click on specific files or in the empty space of the shadow_share and view the
		"properties."  If anything has changed, then you will see it on the "Previous Versions" tab of the properties
		window.
		</para></formalpara>
	  </listitem>
	</orderedlist>

	</sect3>
</sect2>

</sect1>

<sect1>
<title>VFS Modules Available Elsewhere</title>

<para>
<indexterm significance="normal"><primary>VFS modules</primary></indexterm>
This section contains a listing of various other VFS modules that have been posted but do not currently reside
in the Samba CVS tree for one reason or another (e.g., it is easy for the maintainer to have his or her own
CVS tree).
</para>

<para>
No statements about the stability or functionality of any module should be implied due to its presence here.
</para>

<sect2>
<title>DatabaseFS</title>

<para>
<indexterm significance="normal"><primary>DatabaseFS</primary></indexterm>
URL: <ulink noescape="1" url="http://www.css.tayloru.edu/~elorimer/databasefs/index.php">
Taylors University DatabaeFS</ulink>
</para>

<para>By <ulink url="mailto:elorimer@css.tayloru.edu">Eric Lorimer.</ulink></para>

<para>
I have created a VFS module that implements a fairly complete read-only filesystem. It presents information
from a database as a filesystem in a modular and generic way to allow different databases to be used.
(Originally designed for organizing MP3s under directories such as <quote>Artists,</quote> <quote>Song
Keywords,</quote> and so on. I have since easily applied it to a student roster database.) The directory
structure is stored in the database itself and the module makes no assumptions about the database structure
beyond the table it requires to run.
</para>

<para>
Any feedback would be appreciated: comments, suggestions, patches, and so on. If nothing else, it
might prove useful for someone else who wishes to create a virtual filesystem.
</para>

</sect2>

<sect2>
<title>vscan</title>

<indexterm significance="normal"><primary>vscan</primary></indexterm>
<para>URL: <ulink noescape="1" url="http://www.openantivirus.org/projects.php#samba-vscan">
Open Anti-Virus vscan</ulink>
</para>

<para>
<indexterm significance="normal"><primary>samba-vscan</primary></indexterm>
samba-vscan is a proof-of-concept module for Samba, which provides on-access anti-virus support for files
shared using Samba.  samba-vscan supports various virus scanners and is maintained by Rainer Link.
</para>

</sect2>

<sect2>
<title>vscan-clamav</title>
<para>
Samba users have been using the RPMS from SerNet without a problem.
OpenSUSE Linux users have also used the vscan scanner for quite some time
with excellent results. It does impact overall write performance though.
</para>

<para>
The following share stanza is a good guide for those wanting to configure vscan-clamav:
</para>

<screen format="linespecific">
[share]
vfs objects = vscan-clamav
vscan-clamav: config-file = /etc/samba/vscan-clamav.conf
</screen>

<para>
The following example of the <filename moreinfo="none">vscan-clamav.conf</filename> file may help to get this
fully operational:
</para>

<screen format="linespecific">
<title>VFS: Vscan ClamAV Control File</title>
#
# /etc/samba/vscan-clamav.conf
#

[samba-vscan]
; run-time configuration for vscan-samba using
; clamd
; all options are set to default values

; do not scan files larger than X bytes. If set to 0 (default),
; this feature is disable (i.e. all files are scanned)
max file size = 10485760

; log all file access (yes/no). If set to yes, every access will
; be logged. If set to no (default), only access to infected files
; will be logged
verbose file logging = no

; if set to yes (default), a file will be scanned while opening
scan on open = yes
; if set to yes, a file will be scanned while closing (default is yes)
scan on close = yes

; if communication to clamd fails, should access to file denied?
; (default: yes)
deny access on error = no

; if daemon failes with a minor error (corruption, etc.),
; should access to file denied?
; (default: yes)
deny access on minor error = no

; send a warning message via Windows Messenger service
; when virus is found?
; (default: yes)
send warning message = yes

; what to do with an infected file
; quarantine: try to move to quantine directory
; delete:     delete infected file
; nothing:    do nothing (default)
infected file action = quarantine

; where to put infected files - you really want to change this!
quarantine directory  = /opt/clamav/quarantine
; prefix for files in quarantine
quarantine prefix = vir-

; as Windows tries to open a file multiple time in a (very) short time
; of period, samba-vscan use a last recently used file mechanism to avoid
; multiple scans of a file. This setting specified the maximum number of
; elements of the last recently used file list. (default: 100)
max lru files entries = 100

; an entry is invalidad after lru file entry lifetime (in seconds).
; (Default: 5)
lru file entry lifetime = 5

; exclude files from being scanned based on the MIME-type! Semi-colon
; seperated list (default: empty list). Use this with care!
exclude file types =

; socket name of clamd (default: /var/run/clamd). Setting will be ignored if
; libclamav is used
clamd socket name = /tmp/clamd

; limits, if vscan-clamav was build for using the clamav library (libclamav)
; instead of clamd

; maximum number of files in archive (default: 1000)
libclamav max files in archive = 1000

; maximum archived file size, in bytes (default: 10 MB)
libclamav max archived file size = 5242880

; maximum recursion level (default: 5)
libclamav max recursion level = 5
</screen>

<para>
Obviously, a running clam daemon is necessary for this to work. This is a working example for me using ClamAV.
The ClamAV documentation should provide additional configuration examples. On your system these may be located
under the <filename moreinfo="none">/usr/share/doc/</filename> directory. Some examples may also target other virus scanners,
any of which can be used.
</para>

</sect2>
</sect1>

</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="winbind">

<chapterinfo>
	<author>
		<firstname>Tim</firstname><surname>Potter</surname>
		<affiliation>
			<orgname>Samba Team</orgname>
			<address format="linespecific"><email>tpot@linuxcare.com.au</email></address>
		</affiliation>
	</author>
	
<author>
	<firstname>Andrew</firstname><surname>Tridgell</surname>
	<affiliation>
		<orgname>Samba Team</orgname>
		<address format="linespecific"><email>tridge@samba.org</email></address>
	</affiliation>
</author>
	<author>
		<firstname>Naag</firstname><surname>Mummaneni</surname>
		<affiliation>
			<address format="linespecific"><email>getnag@rediffmail.com</email></address>
		</affiliation>
		<contrib>Notes for Solaris</contrib>
	</author>
	<author>
		<firstname>John</firstname><surname>Trostel</surname>
		<affiliation>
			<orgname>SNAP</orgname>
			<address format="linespecific"><email>jtrostel@snapserver.com</email></address>
		</affiliation>
	</author>
	<author>
<firstname>Jelmer</firstname><othername>R.</othername><surname>Vernooij</surname><othername>R.</othername>
<affiliation>
	<orgname>The Samba Team</orgname>
	<address format="linespecific"><email>jelmer@samba.org</email></address>
</affiliation></author>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
	<pubdate>June 15, 2005</pubdate>
</chapterinfo>

<title>Winbind: Use of Domain Accounts</title>

<sect1>
	<title>Features and Benefits</title>

	<para>
<indexterm significance="normal"><primary>holy grail</primary></indexterm>
<indexterm significance="normal"><primary>heterogeneous computing</primary></indexterm>
	Integration of UNIX and Microsoft Windows NT through a unified logon has
	been considered a <quote>holy grail</quote> in heterogeneous computing environments for
	a long time.
	</para>

	<para>
<indexterm significance="normal"><primary>interoperability</primary></indexterm>
<indexterm significance="normal"><primary>domain user</primary></indexterm>
<indexterm significance="normal"><primary>domain group</primary></indexterm>
<indexterm significance="normal"><primary>group ownership</primary></indexterm>
	There is one other facility without which UNIX and Microsoft Windows network
	interoperability would suffer greatly. It is imperative that there be a
	mechanism for sharing files across UNIX systems and to be able to assign
	domain user and group ownerships with integrity.
	</para>

	<para>
<indexterm significance="normal"><primary>Pluggable Authentication Modules</primary><see>PAM</see></indexterm>
<indexterm significance="normal"><primary>winbind</primary></indexterm>
<indexterm significance="normal"><primary>NSS</primary></indexterm>
<indexterm significance="normal"><primary>RPC</primary></indexterm>
	<emphasis>winbind</emphasis> is a component of the Samba suite of programs that
	solves the unified logon problem. Winbind uses a UNIX implementation of Microsoft
	RPC calls, Pluggable Authentication Modules (PAMs), and the name service switch (NSS) to
	allow Windows NT domain users to appear and operate as UNIX users on a UNIX
	machine. This chapter describes the Winbind system, the functionality
	it provides, how it is configured, and how it works internally.
	</para>

	<para>
	Winbind provides three separate functions:
	</para>

	<itemizedlist>
		<listitem><para>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
<indexterm significance="normal"><primary>NT4 domain</primary></indexterm>
		Authentication of user credentials (via PAM). This makes it possible to
		log onto a UNIX/Linux system using user and group accounts from a Windows
		NT4 (including a Samba domain) or an Active Directory domain.
		</para></listitem>

		<listitem><para>
<indexterm significance="normal"><primary>identity resolution</primary></indexterm>
<indexterm significance="normal"><primary>NSS</primary></indexterm>
		Identity resolution (via NSS). This is the default when winbind is not used.
		</para></listitem>

		<listitem><para>
<indexterm significance="normal"><primary>UID</primary></indexterm>
<indexterm significance="normal"><primary>GID</primary></indexterm>
<indexterm significance="normal"><primary>SID</primary></indexterm>
<indexterm significance="normal"><primary>idmap uid</primary></indexterm>
<indexterm significance="normal"><primary>idmap gid</primary></indexterm>
<indexterm significance="normal"><primary>idmap backend</primary></indexterm>
<indexterm significance="normal"><primary/>LDAP</indexterm>
		Winbind maintains a database called winbind_idmap.tdb in which it stores
		mappings between UNIX UIDs, GIDs, and NT SIDs. This mapping is used only
		for users and groups that do not have a local UID/GID. It stores the UID/GID
		allocated from the idmap uid/gid range that it has mapped to the NT SID.
		If <parameter moreinfo="none">idmap backend</parameter> has been specified as <constant>ldap:ldap://hostname[:389]</constant>,
		then instead of using a local mapping, Winbind will obtain this information
		from the LDAP database.
		</para></listitem>
	</itemizedlist>

	<note><para>
	<indexterm significance="normal"><primary>winbindd</primary></indexterm>
	<indexterm significance="normal"><primary>starting samba</primary><secondary>winbindd</secondary></indexterm>
<indexterm significance="normal"><primary>/etc/passwd</primary></indexterm>
<indexterm significance="normal"><primary>/etc/group</primary></indexterm>
<indexterm significance="normal"><primary>smbd</primary></indexterm>
<indexterm significance="normal"><primary>NSS</primary></indexterm>
	If <literal>winbindd</literal> is not running, smbd (which calls <literal>winbindd</literal>) will fall back to
	using purely local information from <filename moreinfo="none">/etc/passwd</filename> and <filename moreinfo="none">/etc/group</filename> and no dynamic
	mapping will be used. On an operating system that has been enabled with the NSS,
	the resolution of user and group information will be accomplished via NSS.
	</para></note>


	<figure id="winbind_idmap" float="0">
	<title>Winbind Idmap</title>
	<mediaobject><imageobject role="html"><imagedata fileref="images/idmap_winbind_no_loop.png" scale="45" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/idmap_winbind_no_loop.png" scale="45" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/idmap_winbind_no_loop"/></imageobject></mediaobject>
	</figure>

</sect1>


<sect1>
	<title>Introduction</title>

	<para>It is well known that UNIX and Microsoft Windows NT have
	different models for representing user and group information and
	use different technologies for implementing them. This fact has
	made it difficult to integrate the two systems in a satisfactory
	manner.</para>

	<para>
<indexterm significance="normal"><primary>synchronization problems</primary></indexterm>
<indexterm significance="normal"><primary>passwords</primary></indexterm>
	One common solution in use today has been to create
	identically named user accounts on both the UNIX and Windows systems
	and use the Samba suite of programs to provide file and print services
	between the two. This solution is far from perfect, however, because
	adding and deleting users on both sets of machines becomes a chore,
	and two sets of passwords are required <?latex --- ?> both of which
	can lead to synchronization problems between the UNIX and Windows
	systems and confusion for users.</para>

	<para>We divide the unified logon problem for UNIX machines into
	three smaller problems:</para>

	<itemizedlist>
		<listitem><para>Obtaining Windows NT user and group information.
		</para></listitem>

		<listitem><para>Authenticating Windows NT users.
		</para></listitem>

		<listitem><para>Password changing for Windows NT users.
		</para></listitem>
	</itemizedlist>


	<para>
<indexterm significance="normal"><primary>unified logon</primary></indexterm>
<indexterm significance="normal"><primary>duplication of information</primary></indexterm>
	Ideally, a prospective solution to the unified logon problem
	would satisfy all the above components without duplication of
	information on the UNIX machines and without creating additional
	tasks for the system administrator when maintaining users and
	groups on either system. The Winbind system provides a simple
	and elegant solution to all three components of the unified logon
	problem.</para>
</sect1>


<sect1>
	<title>What Winbind Provides</title>

	<para>
<indexterm significance="normal"><primary>Windows account management</primary></indexterm>
<indexterm significance="normal"><primary>UNIX users</primary></indexterm>
<indexterm significance="normal"><primary>UNIX groups</primary></indexterm>
<indexterm significance="normal"><primary>NT domain</primary></indexterm>
	Winbind unifies UNIX and Windows NT account management by
	allowing a UNIX box to become a full member of an NT domain. Once
	this is done, the UNIX box will see NT users and groups as if
	they were <quote>native</quote> UNIX users and groups, allowing the NT domain
	to be used in much the same manner that NIS+ is used within
	UNIX-only environments.</para>

	<para>
<indexterm significance="normal"><primary>Winbind hooks</primary></indexterm>
<indexterm significance="normal"><primary>domain controller</primary></indexterm>
<indexterm significance="normal"><primary>NSS</primary></indexterm>
<indexterm significance="normal"><primary>redirection</primary></indexterm>
	The end result is that whenever a
	program on the UNIX machine asks the operating system to look up
	a user or group name, the query will be resolved by asking the
	NT domain controller for the specified domain to do the lookup.
	Because Winbind hooks into the operating system at a low level
	(via the NSS name resolution modules in the C library), this
	redirection to the NT domain controller is completely
	transparent.</para>

	<para>
<indexterm significance="normal"><primary>user and group</primary></indexterm>
<indexterm significance="normal"><primary>domain user</primary></indexterm>
	Users on the UNIX machine can then use NT user and group
	names as they would <quote>native</quote> UNIX names. They can chown files
	so they are owned by NT domain users or even login to the
	UNIX machine and run a UNIX X-Window session as a domain user.</para>

	<para>
<indexterm significance="normal"><primary>domain controller</primary></indexterm>
	The only obvious indication that Winbind is being used is
	that user and group names take the form <constant>DOMAIN\user</constant> and
	<constant>DOMAIN\group</constant>. This is necessary because it allows Winbind to determine
	that redirection to a domain controller is wanted for a particular
	lookup and which trusted domain is being referenced.</para>

	<para>
<indexterm significance="normal"><primary>PAM-enabled</primary></indexterm>
<indexterm significance="normal"><primary>domain controller</primary></indexterm>
	Additionally, Winbind provides an authentication service that hooks into the PAM system
	to provide authentication via an NT domain to any PAM-enabled
	applications. This capability solves the problem of synchronizing
	passwords between systems, since all passwords are stored in a single
	location (on the domain controller).</para>

	<sect2>
		<title>Target Uses</title>

		<para>
<indexterm significance="normal"><primary>infrastructure</primary></indexterm>
		Winbind is targeted at organizations that have an
		existing NT-based domain infrastructure into which they wish
		to put UNIX workstations or servers. Winbind will allow these
		organizations to deploy UNIX workstations without having to
		maintain a separate account infrastructure. This greatly
		simplifies the administrative overhead of deploying UNIX
		workstations into an NT-based organization.</para>

		<para>
<indexterm significance="normal"><primary>Appliances</primary></indexterm>
<indexterm significance="normal"><primary>Winbind</primary></indexterm>
		Another interesting way in which we expect Winbind to
		be used is as a central part of UNIX-based appliances. Appliances
		that provide file and print services to Microsoft-based networks
		will be able to use Winbind to provide seamless integration of
		the appliance into the domain.</para>
	</sect2>

	<sect2>
	<title>Handling of Foreign SIDs</title>

	<para>
<indexterm significance="normal"><primary>foreign SID</primary></indexterm>
	The term <emphasis>foreign SID</emphasis> is often met with the reaction that it
	is not relevant to a particular environment. The following documents an interchange
	that took place on the Samba mailing list. It is a good example of the confusion
	often expressed regarding the use of winbind.
	</para>

	<para>
<indexterm significance="normal"><primary>local domain</primary></indexterm>
	Fact: Winbind is needed to handle users who use workstations that are NOT part
	of the local domain.
	</para>

	<para>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
	Response: <quote>Why? I've used Samba with workstations that are not part of my domains
	lots of times without using winbind. I thought winbind was for using Samba as a member server
	in a domain controlled by another Samba/Windows PDC.</quote>
	</para>

	<para>
<indexterm significance="normal"><primary>UID</primary></indexterm>
<indexterm significance="normal"><primary>GID</primary></indexterm>
<indexterm significance="normal"><primary>foreign user</primary></indexterm>
	If the Samba server will be accessed from a domain other than the local Samba domain, or
	if there will be access from machines that are not local domain members, winbind will
	permit the allocation of UIDs and GIDs from the assigned pool that will keep the identity
	of the foreign user separate from users that are members of the Samba domain.
	</para>

	<para>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>domain member</primary></indexterm>
<indexterm significance="normal"><primary>domain non-member</primary></indexterm>
<indexterm significance="normal"><primary>SID</primary></indexterm>
	This means that winbind is eminently useful in cases where a single
	Samba PDC on a local network is combined with both domain member and domain non-member workstations.
	If winbind is not used, the user george on a Windows workstation that is not a domain
	member will be able to access the files of a user called george in the account database
	of the Samba server that is acting as a PDC. When winbind is used, the default condition
	is that the local user george will be treated as the account DOMAIN\george and the
	foreign (non-member of the domain) account will be treated as MACHINE\george because
	each has a different SID.
	</para>

	</sect2>
</sect1>



<sect1>
	<title>How Winbind Works</title>

	<para>
<indexterm significance="normal"><primary>winbindd</primary></indexterm>
<indexterm significance="normal"><primary>UNIX domain socket</primary></indexterm>
<indexterm significance="normal"><primary>NSS</primary></indexterm>
<indexterm significance="normal"><primary>PAM</primary></indexterm>
	The Winbind system is designed around a client/server
	architecture. A long-running <literal>winbindd</literal> daemon
	listens on a UNIX domain socket waiting for requests
	to arrive. These requests are generated by the NSS and PAM
	clients and are processed sequentially.</para>

	<para>The technologies used to implement Winbind are described
	in detail below.</para>

	<sect2>
		<title>Microsoft Remote Procedure Calls</title>

		<para>
<indexterm significance="normal"><primary>Microsoft Remote Procedure Call</primary><see>MSRPC</see></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>remote management</primary></indexterm>
<indexterm significance="normal"><primary>user authentication</primary></indexterm>
<indexterm significance="normal"><primary>print spooling</primary></indexterm>
		Over the last few years, efforts have been underway by various Samba Team members to implement various aspects of
		the Microsoft Remote Procedure Call (MSRPC) system. This system is used for most network-related operations
		between Windows NT machines, including remote management, user authentication, and print spooling. Although
		initially this work was done to aid the implementation of Primary Domain Controller (PDC) functionality in
		Samba, it has also yielded a body of code that can be used for other purposes.
		</para>

		<para>
<indexterm significance="normal"><primary>MSRPC</primary></indexterm>
<indexterm significance="normal"><primary>enumerate domain users</primary></indexterm>
<indexterm significance="normal"><primary>enumerate domain groups</primary></indexterm>
		Winbind uses various MSRPC calls to enumerate domain users and groups and to obtain detailed information about
		individual users or groups. Other MSRPC calls can be used to authenticate NT domain users and to change user
		passwords. By directly querying a Windows PDC for user and group information, Winbind maps the NT account
		information onto UNIX user and group names.
		</para>
	</sect2>

	<sect2>
		<title>Microsoft Active Directory Services</title>

		<para>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
<indexterm significance="normal"><primary>Kerberos</primary></indexterm>
<indexterm significance="normal"><primary>Winbind</primary></indexterm>
<indexterm significance="normal"><primary>native mode</primary></indexterm>
		Since late 2001, Samba has gained the ability to interact with Microsoft Windows 2000 using its <quote>native
		mode</quote> protocols rather than the NT4 RPC services.  Using LDAP and Kerberos, a domain member running
		Winbind can enumerate users and groups in exactly the same way as a Windows 200x client would, and in so doing
		provide a much more efficient and effective Winbind implementation.
		</para>
	</sect2>

	<sect2>
		<title>Name Service Switch</title>

		<para>
<indexterm significance="normal"><primary>NSS</primary></indexterm>
<indexterm significance="normal"><primary>networked workstation</primary></indexterm>
<indexterm significance="normal"><primary>NIS</primary></indexterm>
<indexterm significance="normal"><primary>DNS</primary></indexterm>
		The NSS is a feature that is present in many UNIX operating systems. It allows system
		information such as hostnames, mail aliases, and user information
		to be resolved from different sources. For example, a standalone
		UNIX workstation may resolve system information from a series of
		flat files stored on the local file system. A networked workstation
		may first attempt to resolve system information from local files,
		and then consult an NIS database for user information or a DNS server
		for hostname information.</para>

		<para>
<indexterm significance="normal"><primary>NSS</primary></indexterm>
<indexterm significance="normal"><primary>MSRPC</primary></indexterm>
<indexterm significance="normal"><primary>trusted domain</primary></indexterm>
<indexterm significance="normal"><primary>local users</primary></indexterm>
<indexterm significance="normal"><primary>local groups</primary></indexterm>
		The NSS application programming interface allows Winbind to present itself as a source of system
		information when resolving UNIX usernames and groups. Winbind uses this interface and information obtained
		from a Windows NT server using MSRPC calls to provide a new source of account enumeration. Using standard UNIX
		library calls, you can enumerate the users and groups on a UNIX machine running Winbind and see all users and
		groups in an NT domain plus any trusted domain as though they were local users and groups.
		</para>

		<para>
<indexterm significance="normal"><primary>NSS</primary></indexterm>
<indexterm significance="normal"><primary>/etc/nsswitch.conf</primary></indexterm>
<indexterm significance="normal"><primary>passwd</primary></indexterm>
		The primary control file for NSS is <filename moreinfo="none">/etc/nsswitch.conf</filename>.  When a UNIX application
		makes a request to do a lookup, the C library looks in <filename moreinfo="none">/etc/nsswitch.conf</filename> for a line that
		matches the service type being requested; for example, the <quote>passwd</quote> service type is used when
		user or group names are looked up. This config line specifies which implementations of that service should be
		tried and in what order. If the passwd config line is:
<screen format="linespecific">
passwd: files example
</screen>
<indexterm significance="normal"><primary>/lib/libnss_files.so</primary></indexterm>
<indexterm significance="normal"><primary>/lib/libnss_example.so</primary></indexterm>
<indexterm significance="normal"><primary>resolver functions</primary></indexterm>
		then the C library will first load a module called <filename moreinfo="none">/lib/libnss_files.so</filename> followed
		by the module <filename moreinfo="none">/lib/libnss_example.so</filename>. The C library will dynamically load each of these
		modules in turn and call resolver functions within the modules to try to resolve the request. Once the request
		is resolved, the C library returns the result to the application.
		</para>

		<para>
<indexterm significance="normal"><primary>NSS</primary></indexterm>
<indexterm significance="normal"><primary>libnss_winbind.so</primary></indexterm>
<indexterm significance="normal"><primary>/etc/nsswitch.conf</primary></indexterm>
		This NSS interface provides an easy way for Winbind to hook into the operating system. All that needs
		to be done is to put <filename moreinfo="none">libnss_winbind.so</filename> in <filename moreinfo="none">/lib/</filename> then add
		<quote>winbind</quote> into <filename moreinfo="none">/etc/nsswitch.conf</filename> at the appropriate place. The C library
		will then call Winbind to resolve user and group names.
		</para>
	</sect2>

	<sect2>
		<title>Pluggable Authentication Modules</title>

		<para>
<indexterm significance="normal"><primary>PAM</primary></indexterm>
<indexterm significance="normal"><primary>authentication methods</primary></indexterm>
<indexterm significance="normal"><primary>authorization</primary></indexterm>
<indexterm significance="normal"><primary>NIS database</primary></indexterm>
		PAMs provide a system for abstracting authentication and authorization technologies. With a PAM
		module, it is possible to specify different authentication methods for different system applications without
		having to recompile these applications. PAM is also useful for implementing a particular policy for
		authorization. For example, a system administrator may only allow console logins from users stored in the
		local password file but only allow users resolved from an NIS database to log in over the network.
		</para>

		<para>
<indexterm significance="normal"><primary>PAM</primary></indexterm>
<indexterm significance="normal"><primary>Winbind</primary></indexterm>
<indexterm significance="normal"><primary>authentication management</primary></indexterm>
<indexterm significance="normal"><primary>password management</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
		Winbind uses the authentication management and password management PAM interface to integrate Windows
		NT users into a UNIX system. This allows Windows NT users to log in to a UNIX machine and be authenticated
		against a suitable PDC.  These users can also change their passwords and have this change take effect directly
		on the PDC.
		</para>

		<para>
<indexterm significance="normal"><primary>PAM</primary></indexterm>
<indexterm significance="normal"><primary>/etc/pam.d/</primary></indexterm>
<indexterm significance="normal"><primary>pam_winbind.so</primary></indexterm>
<indexterm significance="normal"><primary>/lib/security/</primary></indexterm>
		PAM is configured by providing control files in the directory <filename moreinfo="none">/etc/pam.d/</filename> for
		each of the services that require authentication. When an authentication request is made by an application,
		the PAM code in the C library looks up this control file to determine what modules to load to do the
		authentication check and in what order. This interface makes adding a new authentication service for Winbind
		very easy: simply copy the <filename moreinfo="none">pam_winbind.so</filename> module to <filename moreinfo="none">/lib/security/</filename>,
		and the PAM control files for relevant services are updated to allow authentication via Winbind. See the PAM
		documentation in <link linkend="pam">PAM-Based Distributed Authentication</link>, for more information.
		</para>
	</sect2>

	<sect2>
		<title>User and Group ID Allocation</title>

		<para>
<indexterm significance="normal"><primary>RID</primary></indexterm>
<indexterm significance="normal"><primary>Winbind</primary></indexterm>
<indexterm significance="normal"><primary>UNIX ID</primary></indexterm>
		When a user or group is created under Windows NT/200x, it is allocated a numerical relative identifier
		(RID). This is slightly different from UNIX, which has a range of numbers that are used to identify users and
		the same range used to identify groups. It is Winbind's job to convert RIDs to UNIX ID numbers and vice versa.
		When Winbind is configured, it is given part of the UNIX user ID space and a part of the UNIX group ID space
		in which to store Windows NT users and groups. If a Windows NT user is resolved for the first time, it is
		allocated the next UNIX ID from the range. The same process applies for Windows NT groups. Over time, Winbind
		will have mapped all Windows NT users and groups to UNIX user IDs and group IDs.
		</para>

		<para>
<indexterm significance="normal"><primary>ID mapping database</primary></indexterm>
<indexterm significance="normal"><primary>tdb</primary></indexterm>
<indexterm significance="normal"><primary>UNIX ID</primary></indexterm>
<indexterm significance="normal"><primary>RID</primary></indexterm>
		The results of this mapping are stored persistently in an ID mapping database held in a tdb database.
		This ensures that RIDs are mapped to UNIX IDs in a consistent way.
		</para>
	</sect2>

	<sect2>
		<title>Result Caching</title>

		<para>
<indexterm significance="normal"><primary>SAM</primary></indexterm>
<indexterm significance="normal"><primary>caching scheme</primary></indexterm>
<indexterm significance="normal"><primary>Winbind</primary></indexterm>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
		An active directory system can generate a lot of user and group name lookups. To reduce the network
		cost of these lookups, Winbind uses a caching scheme based on the SAM sequence number supplied by NT domain
		controllers. User or group information returned by a PDC is cached by Winbind along with a sequence number
		also returned by the PDC. This sequence number is incremented by Windows NT whenever any user or group
		information is modified. If a cached entry has expired, the sequence number is requested from the PDC and
		compared against the sequence number of the cached entry.  If the sequence numbers do not match, then the
		cached information is discarded and up-to-date information is requested directly from the PDC.
		</para>
	</sect2>
</sect1>


<sect1>
	<title>Installation and Configuration</title>

<sect2>
<title>Introduction</title>

<para>
<indexterm significance="normal"><primary>Winbind</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>authentication control</primary></indexterm>
This section describes the procedures used to get Winbind up and running. Winbind is capable of providing
access and authentication control for Windows Domain users through an NT or Windows 200x PDC for regular
services, such as telnet and ftp, as well for Samba services.
</para>

<itemizedlist>
<listitem>
	<para>
	<emphasis>Why should I do this?</emphasis>
	</para>

	<para>
<indexterm significance="normal"><primary>Samba administrator</primary></indexterm>
<indexterm significance="normal"><primary>authentication mechanisms</primary></indexterm>
<indexterm significance="normal"><primary>domain members</primary></indexterm>
<indexterm significance="normal"><primary>accounts</primary></indexterm>
This allows the Samba administrator to rely on the authentication mechanisms on the Windows NT/200x PDC
for the authentication of domain members. Windows NT/200x users no longer need to have separate accounts on
the Samba server.
	</para>
</listitem>

<listitem>
	<para>
	<emphasis>Who should be reading this document?</emphasis>
	</para>

	<para>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>Windows NT/200x</primary></indexterm>
This document is designed for system administrators. If you are implementing Samba on a file server and wish
to (fairly easily) integrate existing Windows NT/200x users from your PDC onto the Samba server, this document
is for you.
	</para>
</listitem>
</itemizedlist>
</sect2>


<sect2>
<title>Requirements</title>

<para>
<indexterm significance="normal"><primary>PAM</primary></indexterm>
<indexterm significance="normal"><primary>back up</primary></indexterm>
<indexterm significance="normal"><primary>boot disk`</primary></indexterm>
If you have a Samba configuration file that you are currently using, <emphasis>BACK IT UP!</emphasis>
If your system already uses PAM, <emphasis>back up the <filename moreinfo="none">/etc/pam.d</filename> directory
contents!</emphasis> If you haven't already made a boot disk, <emphasis>MAKE ONE NOW!</emphasis>
</para>

<para>
<indexterm significance="normal"><primary>PAM configuration</primary></indexterm>
<indexterm significance="normal"><primary>/etc/pam.d</primary></indexterm>
<indexterm significance="normal"><primary>single-user mode</primary></indexterm>
Messing with the PAM configuration files can make it nearly impossible to log in to your machine. That's
why you want to be able to boot back into your machine in single-user mode and restore your
<filename moreinfo="none">/etc/pam.d</filename> to the original state it was in if you get frustrated with the
way things are going.
</para>

<para>
<indexterm significance="normal"><primary>winbindd</primary></indexterm>
<indexterm significance="normal"><primary>daemon</primary></indexterm>
The latest version of Samba-3 includes a functioning winbindd daemon. Please refer to the <ulink url="http://samba.org/">main Samba Web page</ulink>, or better yet, your closest Samba mirror site for
instructions on downloading the source code.
</para>

<para>
<indexterm significance="normal"><primary>domain users</primary></indexterm>
<indexterm significance="normal"><primary>shares and files</primary></indexterm>
<indexterm significance="normal"><primary>PAM</primary></indexterm>
<indexterm significance="normal"><primary>development libraries</primary></indexterm>
To allow domain users the ability to access Samba shares and files, as well as potentially other services
provided by your Samba machine, PAM must be set up properly on your
machine. In order to compile the Winbind modules, you should have at least the PAM development libraries installed
on your system. Please refer to the PAM Web site <ulink url="http://www.kernel.org/pub/linux/libs/pam/"/>.
</para>
</sect2>

<sect2>
<title>Testing Things Out</title>

<para>
<indexterm significance="normal"><primary>smbd</primary></indexterm>
<indexterm significance="normal"><primary>nmbd</primary></indexterm>
<indexterm significance="normal"><primary>winbindd</primary></indexterm>
<indexterm significance="normal"><primary>/etc/pam.d</primary></indexterm>
<indexterm significance="normal"><primary>PAM</primary></indexterm>
Before starting, it is probably best to kill off all the Samba-related daemons running on your server.
Kill off all <application moreinfo="none">smbd</application>, <application moreinfo="none">nmbd</application>, and <application moreinfo="none">winbindd</application> processes that may be running. To use PAM,
make sure that you have the standard PAM package that supplies the <filename moreinfo="none">/etc/pam.d</filename>
directory structure, including the PAM modules that are used by PAM-aware services, several PAM libraries,
and the <filename moreinfo="none">/usr/doc</filename> and <filename moreinfo="none">/usr/man</filename> entries for PAM. Winbind is built
better in Samba if the pam-devel package is also installed. This package includes the header files
needed to compile PAM-aware applications.
</para>

<sect3>
<title>Configure <filename moreinfo="none">nsswitch.conf</filename> and the Winbind Libraries on Linux and Solaris</title>

<para>
<indexterm significance="normal"><primary>PAM</primary></indexterm>
<indexterm significance="normal"><primary>pam-devel</primary></indexterm>
<indexterm significance="normal"><primary>Winbind</primary></indexterm>
<indexterm significance="normal"><primary>/etc/nsswitch.conf</primary></indexterm>
PAM is a standard component of most current generation UNIX/Linux systems. Unfortunately, few systems install
the <filename moreinfo="none">pam-devel</filename> libraries that are needed to build PAM-enabled Samba. Additionally, Samba-3
may auto-install the Winbind files into their correct locations on your system, so before you get too far down
the track, be sure to check if the following configuration is really
necessary. You may only need to configure
<filename moreinfo="none">/etc/nsswitch.conf</filename>.
</para>

<para>
The libraries needed to run the <application moreinfo="none">winbindd</application> daemon through nsswitch need to be copied to their proper locations:
</para>

<para>
<indexterm significance="normal"><primary>libnss_winbind.so</primary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">cp ../samba/source/nsswitch/libnss_winbind.so /lib</userinput>
</screen>
</para>

<para>
I also found it necessary to make the following symbolic link:
</para>

<para>
<prompt moreinfo="none">root# </prompt> <userinput moreinfo="none">ln -s /lib/libnss_winbind.so /lib/libnss_winbind.so.2</userinput>
</para>

<para>And, in the case of Sun Solaris:
<indexterm significance="normal"><primary>nss_winbind.so.1</primary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">ln -s /usr/lib/libnss_winbind.so /usr/lib/libnss_winbind.so.1</userinput>
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">ln -s /usr/lib/libnss_winbind.so /usr/lib/nss_winbind.so.1</userinput>
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">ln -s /usr/lib/libnss_winbind.so /usr/lib/nss_winbind.so.2</userinput>
</screen>
</para>

<para>
<indexterm significance="normal"><primary>/etc/nsswitch.conf</primary></indexterm>
As root, edit <filename moreinfo="none">/etc/nsswitch.conf</filename> to allow user and group entries to be visible from the
<application moreinfo="none">winbindd</application> daemon. My <filename moreinfo="none">/etc/nsswitch.conf</filename> file looked like this after editing:
<programlisting format="linespecific">
passwd:     files winbind
shadow:     files
group:      files winbind
</programlisting></para>

<para>
<indexterm significance="normal"><primary>winbindd</primary></indexterm>
<indexterm significance="normal"><primary>ldconfig</primary></indexterm>
<indexterm significance="normal"><primary>libnss_winbind</primary></indexterm>
<indexterm significance="normal"><primary>grep</primary></indexterm>
<indexterm significance="normal"><primary>dynamic link loader</primary></indexterm>
The libraries needed by the <literal>winbindd</literal> daemon will be automatically
entered into the <literal>ldconfig</literal> cache the next time
your system reboots, but it is faster (and you do not need to reboot) if you do it manually:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">/sbin/ldconfig -v | grep winbind</userinput>
</screen>
This makes <filename moreinfo="none">libnss_winbind</filename> available to winbindd and reports the current
search path that is used by the dynamic link loader. The use of the <literal>grep</literal>
filters the output of the <literal>ldconfig</literal> command so that we may see proof that
this library is indeed recognized by the dynamic link loader.
</para>

<para>
<indexterm significance="normal"><primary>dynamic link loader</primary></indexterm>
<indexterm significance="normal"><primary>crle</primary></indexterm>
<indexterm significance="normal"><primary>/usr/local/lib</primary></indexterm>
<indexterm significance="normal"><primary>link loader configuration</primary></indexterm>
<indexterm significance="normal"><primary>object module dependencies</primary></indexterm>
The Sun Solaris dynamic link loader management tool is called <literal>crle</literal>. The
use of this tool is necessary to instruct the dynamic link loader to search directories that
contain library files that were not supplied as part of the original operating system platform.
The following example shows how to use this tool to add the directory <filename moreinfo="none">/usr/local/lib</filename>
to the dynamic link loader's search path:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> crle -u -l /usr/lib:/usr/local/lib
</screen>
When executed without arguments, <literal>crle</literal> reports the current dynamic
link loader configuration. This is demonstrated here:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> crle

Configuration file [version 4]: /var/ld/ld.config
  Default Library Path (ELF):   /lib:/usr/lib:/usr/local/lib
  Trusted Directories (ELF):    /lib/secure:/usr/lib/secure  (system default)

Command line:
  crle -c /var/ld/ld.config -l /lib:/usr/lib:/usr/local/lib
</screen>
From this it is apparent that the <filename moreinfo="none">/usr/local/lib</filename> directory is included
in the search dynamic link libraries in order to satisfy object module dependencies.
</para>

</sect3>

<sect3>
<title>NSS Winbind on AIX</title>

<para>(This section is only for those running AIX.)</para>

<para>
<indexterm significance="normal"><primary>AIX</primary></indexterm>
<indexterm significance="normal"><primary>Winbind</primary></indexterm>
<indexterm significance="normal"><primary>/usr/lib/security</primary></indexterm>
<indexterm significance="normal"><primary>authentication module API</primary></indexterm>
<indexterm significance="normal"><primary>/usr/lib/security/methods.cfg</primary></indexterm>
<indexterm significance="normal"><primary>PAM module</primary></indexterm>
The Winbind AIX identification module gets built as <filename moreinfo="none">libnss_winbind.so</filename> in the
nsswitch directory of the Samba source. This file can be copied to <filename moreinfo="none">/usr/lib/security</filename>,
and the AIX naming convention would indicate that it should be named WINBIND. A stanza like the following:
<programlisting format="linespecific">
WINBIND:
        program = /usr/lib/security/WINBIND
        options = authonly
</programlisting>
can then be added to <filename moreinfo="none">/usr/lib/security/methods.cfg</filename>. This module only supports
identification, but there have been reports of success using the standard Winbind PAM module for
authentication. Use caution configuring loadable authentication modules, since misconfiguration can make
it impossible to log on to the system.  Information regarding the AIX authentication module API can
be found in the <quote>Kernel Extensions and Device Support Programming Concepts for AIX</quote> document that
describes the <ulink url="http://publibn.boulder.ibm.com/doc_link/en_US/a_doc_lib/aixprggd/kernextc/sec_load_mod.htm">
Loadable Authentication Module Programming Interface</ulink> for AIX. Further information on administering the modules
can be found in the <ulink url="http://publibn.boulder.ibm.com/doc_link/en_US/a_doc_lib/aixbman/baseadmn/iandaadmin.htm">System
Management Guide: Operating System and Devices.</ulink>
</para>
</sect3>

<sect3>
<title>Configure smb.conf</title>

<para>
<indexterm significance="normal"><primary>winbind</primary></indexterm>
<indexterm significance="normal"><primary>man page</primary></indexterm>
<indexterm significance="normal"><primary>winbindd</primary></indexterm>
Several parameters are needed in the <filename moreinfo="none">smb.conf</filename> file to control the behavior of <application moreinfo="none">winbindd</application>. These
are described in more detail in the <citerefentry><refentrytitle>winbindd</refentrytitle>
<manvolnum>8</manvolnum></citerefentry> man page. My <filename moreinfo="none">smb.conf</filename> file, as shown in <link linkend="winbindcfg">the smb.conf for Winbind Setup</link>, was modified to include the necessary entries in the [global] section.
</para>

<example id="winbindcfg">
<title>smb.conf for Winbind Setup</title>
<simplelist>
<member> </member><member><parameter>[global]</parameter></member>
<member>#  separate domain and username with '\', like DOMAIN\username</member>
<member><indexterm>winbind separator</indexterm><parameter>winbind separator = \</parameter></member>
<member>#  use uids from 10000 to 20000 for domain users</member>
<member><indexterm>idmap uid</indexterm><parameter>idmap uid = 10000-20000</parameter></member>
<member>#  use gids from 10000 to 20000 for domain groups</member>
<member><indexterm>idmap gid</indexterm><parameter>idmap gid = 10000-20000</parameter></member>
<member>#  allow enumeration of winbind users and groups</member>
<member><indexterm>winbind enum users</indexterm><parameter>winbind enum users = yes</parameter></member>
<member><indexterm>winbind enum groups</indexterm><parameter>winbind enum groups = yes</parameter></member>
<member>#  give winbind users a real shell (only needed if they have telnet access)</member>
<member><indexterm>template homedir</indexterm><parameter>template homedir = /home/winnt/%D/%U</parameter></member>
<member><indexterm>template shell</indexterm><parameter>template shell = /bin/bash</parameter></member>
</simplelist>
</example>

</sect3>


<sect3>
<title>Join the Samba Server to the PDC Domain</title>

<para>
<indexterm significance="normal"><primary>domain security</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>BDC</primary></indexterm>
All machines that will participate in domain security should be members of
the domain. This applies also to the PDC and all BDCs.
</para>

<para>
<indexterm significance="normal"><primary>joining domain</primary></indexterm>
<indexterm significance="normal"><primary>domain join</primary></indexterm>
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>join</tertiary></indexterm>
<indexterm significance="normal"><primary>smbd</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>domain controller</primary></indexterm>
<indexterm significance="normal"><primary>MS DCE RPC</primary></indexterm>
<indexterm significance="normal"><primary>DCE RPC</primary></indexterm>
<indexterm significance="normal"><primary>RPC</primary></indexterm>
The process of joining a domain requires the use of the <literal>net rpc join</literal>
command. This process communicates with the domain controller it will register with
(usually the PDC) via MS DCE RPC. This means, of course, that the <literal>smbd</literal>
process must be running on the target domain controller. It is therefore necessary to temporarily
start Samba on a PDC so that it can join its own domain.
</para>

<para>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>administrative privileges</primary></indexterm>
<indexterm significance="normal"><primary>Administrator</primary></indexterm>
Enter the following command to make the Samba server join the domain, where <replaceable>PDC</replaceable> is
the name of your PDC and <replaceable>Administrator</replaceable> is a domain user who has administrative
privileges in the domain.
</para>

<note><para>
<indexterm significance="normal"><primary>domain controller</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>tcp ports</primary></indexterm>
<indexterm significance="normal"><primary>udp ports</primary></indexterm>
Before attempting to join a machine to the domain, verify that Samba is running
on the target domain controller (usually PDC) and that it is capable of being reached via ports
137/udp, 135/tcp, 139/tcp, and 445/tcp (if Samba or Windows Server 2Kx).
</para></note>

<para>
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>join</tertiary></indexterm>
The use of the <literal>net rpc join</literal> facility is shown here:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">/usr/local/samba/bin/net rpc join -S PDC -U Administrator</userinput>
</screen>
The proper response to the command should be <quote>Joined the domain
<replaceable>DOMAIN</replaceable></quote> where <replaceable>DOMAIN</replaceable>
is your domain name.
</para>

</sect3>

<sect3>
<title>Starting and Testing the <literal>winbindd</literal> Daemon</title>

<para>
<indexterm significance="normal"><primary>startup script</primary></indexterm>
<indexterm significance="normal"><primary>winbindd</primary></indexterm>
<indexterm significance="normal"><primary>Winbind services</primary></indexterm>
Eventually, you will want to modify your Samba startup script to automatically invoke the winbindd daemon when
the other parts of Samba start, but it is possible to test out just the Winbind portion first. To start up
Winbind services, enter the following command as root:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">/usr/local/samba/sbin/winbindd</userinput>
</screen>
Use the appropriate path to the location of the <literal>winbindd</literal> executable file.
</para>

<note><para>
<indexterm significance="normal"><primary>Winbind</primary></indexterm>
<indexterm significance="normal"><primary>/usr/local/samba</primary></indexterm>
The command to start up Winbind services assumes that Samba has been installed in the
<filename moreinfo="none">/usr/local/samba</filename> directory tree. You may need to search for the location of Samba files
if this is not the location of <literal>winbindd</literal> on your system.
</para></note>

<para>
<indexterm significance="normal"><primary>paranoid</primary></indexterm>
<indexterm significance="normal"><primary>daemon running</primary></indexterm>
I'm always paranoid and like to make sure the daemon is really running.
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">ps -ae | grep winbindd</userinput>
</screen>
</para>

<para>
<indexterm significance="normal"><primary>winbindd</primary></indexterm>
This command should produce output like the following if the daemon is running.
<screen format="linespecific">
3025 ?        00:00:00 winbindd
</screen>
</para>

<para>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>wbinfo</primary></indexterm>
Now, for the real test, try to get some information about the users on your PDC:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">/usr/local/samba/bin/wbinfo -u</userinput>
</screen>
This should echo back a list of users on your Windows users on your PDC. For example, I get the following
response:
<screen format="linespecific">
CEO\Administrator
CEO\burdell
CEO\Guest
CEO\jt-ad
CEO\krbtgt
CEO\TsInternetUser
</screen>
Obviously, I have named my domain <quote>CEO</quote> and my <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="WINBINDSEPARATOR" xlink:href="smb.conf.5.html#WINBINDSEPARATOR">winbind separator</link> is
<quote>\</quote>.
</para>

<para>
<indexterm significance="normal"><primary>wbinfo</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
You can do the same sort of thing to get group information from the PDC:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">/usr/local/samba/bin/wbinfo -g</userinput>
CEO\Domain Admins
CEO\Domain Users
CEO\Domain Guests
CEO\Domain Computers
CEO\Domain Controllers
CEO\Cert Publishers
CEO\Schema Admins
CEO\Enterprise Admins
CEO\Group Policy Creator Owners
</screen></para>

<para>
<indexterm significance="normal"><primary>getent</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>/etc/passwd</primary></indexterm>
<indexterm significance="normal"><primary>UID</primary></indexterm>
<indexterm significance="normal"><primary>GID</primary></indexterm>
<indexterm significance="normal"><primary>home directories</primary></indexterm>
<indexterm significance="normal"><primary>default shells</primary></indexterm>
The function <literal>getent</literal> can now be used to get unified lists of both local and PDC users and
groups. Try the following command:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">getent passwd</userinput>
</screen>
You should get a list that looks like your <filename moreinfo="none">/etc/passwd</filename>
list followed by the domain users with their new UIDs, GIDs, home
directories, and default shells.
</para>

<para>
The same thing can be done for groups with the command:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">getent group</userinput>
</screen>
</para>

</sect3>


<sect3>
<title>Fix the init.d Startup Scripts</title>

<sect4>
<title>Linux</title>

<para>
<indexterm significance="normal"><primary>winbindd daemon</primary></indexterm>
<indexterm significance="normal"><primary>smbd</primary></indexterm>
<indexterm significance="normal"><primary>nmbd</primary></indexterm>
<indexterm significance="normal"><primary>/etc/init.d/smb</primary></indexterm>
<indexterm significance="normal"><primary>/etc/init.d/samba</primary></indexterm>
<indexterm significance="normal"><primary>/usr/local/samba/bin</primary></indexterm>
<indexterm significance="normal"><primary/></indexterm>
<indexterm significance="normal"><primary/></indexterm>
<indexterm significance="normal"><primary/></indexterm>
The <application moreinfo="none">winbindd</application> daemon needs to start up after the <application moreinfo="none">smbd</application> and <application moreinfo="none">nmbd</application> daemons are running.  To accomplish this
task, you need to modify the startup scripts of your system.  They are located at
<filename moreinfo="none">/etc/init.d/smb</filename> in Red Hat Linux and in <filename moreinfo="none">/etc/init.d/samba</filename> in Debian
Linux. Edit your script to add commands to invoke this daemon in the proper sequence. My startup script starts
up <application moreinfo="none">smbd</application>, <application moreinfo="none">nmbd</application>, and <application moreinfo="none">winbindd</application> from the <filename moreinfo="none">/usr/local/samba/bin</filename> directory directly. The
<literal>start</literal> function in the script looks like this:
<programlisting format="linespecific">
start() {
        KIND="SMB"
        echo -n $"Starting $KIND services: "
        daemon /usr/local/samba/bin/smbd $SMBDOPTIONS
        RETVAL=$?
        echo
        KIND="NMB"
        echo -n $"Starting $KIND services: "
        daemon /usr/local/samba/bin/nmbd $NMBDOPTIONS
        RETVAL2=$?
        echo
        KIND="Winbind"
        echo -n $"Starting $KIND services: "
        daemon /usr/local/samba/sbin/winbindd
        RETVAL3=$?
        echo
        [ $RETVAL -eq 0 -a $RETVAL2 -eq 0 -a $RETVAL3 -eq 0 ] &amp;&amp; \
		touch /var/lock/subsys/smb || RETVAL=1
        return $RETVAL
}
</programlisting></para>

<para>If you would like to run winbindd in dual daemon mode, replace the line:
<programlisting format="linespecific">
        daemon /usr/local/samba/sbin/winbindd
</programlisting>

in the example above with:

<programlisting format="linespecific">
        daemon /usr/local/samba/sbin/winbindd -B
</programlisting>.
</para>

<para>
The <literal>stop</literal> function has a corresponding entry to shut down the services and looks like this:
</para>

<para><programlisting format="linespecific">
stop() {
        KIND="SMB"
        echo -n $"Shutting down $KIND services: "
        killproc smbd
        RETVAL=$?
        echo
        KIND="NMB"
        echo -n $"Shutting down $KIND services: "
        killproc nmbd
        RETVAL2=$?
        echo
        KIND="Winbind"
        echo -n $"Shutting down $KIND services: "
        killproc winbindd
        RETVAL3=$?
        [ $RETVAL -eq 0 -a $RETVAL2 -eq 0 -a $RETVAL3 -eq 0 ] &amp;&amp; \
		 rm -f /var/lock/subsys/smb
        echo ""
        return $RETVAL
}
</programlisting></para>
</sect4>

<sect4>
<title>Solaris</title>

<para>
Winbind does not work on Solaris 9; see <link linkend="winbind-solaris9">Winbind on Solaris 9 section</link>
for details.
</para>

<para>
<indexterm significance="normal"><primary>Solaris 9</primary></indexterm>
<indexterm significance="normal"><primary>/etc/init.d/samba.server</primary></indexterm>
<indexterm significance="normal"><primary>/usr/local/samba/bin</primary></indexterm>
<indexterm significance="normal"><primary>smbd</primary></indexterm>
<indexterm significance="normal"><primary>nmbd</primary></indexterm>
<indexterm significance="normal"><primary>winbindd</primary></indexterm>
On Solaris, you need to modify the <filename moreinfo="none">/etc/init.d/samba.server</filename> startup script. It
usually only starts smbd and nmbd but should now start winbindd, too. If you have Samba installed in
<filename moreinfo="none">/usr/local/samba/bin</filename>, the file could contains something like this:
</para>

<para>
	<programlisting format="linespecific">
	##
	## samba.server
	##

	if [ ! -d /usr/bin ]
	then                    # /usr not mounted
		exit
	fi

	killproc() {            # kill the named process(es)
		pid=`/usr/bin/ps -e |
		     /usr/bin/grep -w $1 |
		     /usr/bin/sed -e 's/^  *//' -e 's/ .*//'`
		[ "$pid" != "" ] &amp;&amp; kill $pid
	}

	# Start/stop processes required for Samba server

	case "$1" in

	'start')
	#
	# Edit these lines to suit your installation (paths, workgroup, host)
	#
	echo Starting SMBD
	   /usr/local/samba/bin/smbd -D -s \
		/usr/local/samba/smb.conf

	echo Starting NMBD
	   /usr/local/samba/bin/nmbd -D -l \
		/usr/local/samba/var/log -s /usr/local/samba/smb.conf

	echo Starting Winbind Daemon
	   /usr/local/samba/sbin/winbindd
	   ;;

	'stop')
	   killproc nmbd
	   killproc smbd
	   killproc winbindd
	   ;;

	*)
	   echo "Usage: /etc/init.d/samba.server { start | stop }"
	   ;;
	esac
</programlisting></para>

<para>
Again, if you would like to run Samba in dual daemon mode, replace:
<programlisting format="linespecific">
/usr/local/samba/sbin/winbindd
</programlisting>
in the script above with:
<programlisting format="linespecific">
/usr/local/samba/sbin/winbindd -B
</programlisting>
</para>

</sect4>

<sect4>
<title>Restarting</title>
<para>
<indexterm significance="normal"><primary>daemons</primary></indexterm>
<indexterm significance="normal"><primary>local user</primary></indexterm>
If you restart the <application moreinfo="none">smbd</application>, <application moreinfo="none">nmbd</application>, and <application moreinfo="none">winbindd</application> daemons at this point, you
should be able to connect to the Samba server as a domain member just as
if you were a local user.
</para>
</sect4>
</sect3>

<sect3>
<title>Configure Winbind and PAM</title>

<para>
<indexterm significance="normal"><primary>winbindd</primary></indexterm>
<indexterm significance="normal"><primary>authentication</primary></indexterm>
<indexterm significance="normal"><primary>PAM</primary></indexterm>
<indexterm significance="normal"><primary>/etc/pam.d</primary></indexterm>
If you have made it this far, you know that <literal>winbindd</literal> and Samba are working together. If you
want to use Winbind to provide authentication for other services, keep reading. The PAM configuration files
need to be altered in this step. (Did you remember to make backups of your original
<filename moreinfo="none">/etc/pam.d</filename> files? If not, do it now.)
</para>

<para>
<indexterm significance="normal"><primary>NSS</primary></indexterm>
<indexterm significance="normal"><primary>../source/nsswitch</primary></indexterm>
<indexterm significance="normal"><primary>pam_winbind.so</primary></indexterm>
<indexterm significance="normal"><primary>/lib/security</primary></indexterm>
<indexterm significance="normal"><primary>Solaris</primary></indexterm>
<indexterm significance="normal"><primary>/usr/lib/security</primary></indexterm>
You will need a PAM module to use winbindd with these other services. This module will be compiled in the
<filename moreinfo="none">../source/nsswitch</filename> directory by invoking the command:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">make nsswitch/pam_winbind.so</userinput>
</screen>
from the <filename moreinfo="none">../source</filename> directory. The <filename moreinfo="none">pam_winbind.so</filename> file should be
copied to the location of your other PAM security modules. On my Red Hat system, this was the
<filename moreinfo="none">/lib/security</filename> directory. On Solaris, the PAM security modules reside in
<filename moreinfo="none">/usr/lib/security</filename>.
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">cp ../samba/source/nsswitch/pam_winbind.so /lib/security</userinput>
</screen>
</para>

<sect4>
<title>Linux/FreeBSD-Specific PAM Configuration</title>

<para>
<indexterm significance="normal"><primary>/etc/pam.d/samba</primary></indexterm>
The <filename moreinfo="none">/etc/pam.d/samba</filename> file does not need to be changed. I just left this file as it was:
<programlisting format="linespecific">
auth    required  /lib/security/pam_stack.so service=system-auth
account required  /lib/security/pam_stack.so service=system-auth
</programlisting></para>

<para>
<indexterm significance="normal"><primary>Winbind</primary></indexterm>
<indexterm significance="normal"><primary>authentication service</primary></indexterm>
<indexterm significance="normal"><primary>login</primary></indexterm>
<indexterm significance="normal"><primary>console</primary></indexterm>
<indexterm significance="normal"><primary>telnet logins</primary></indexterm>
<indexterm significance="normal"><primary>ftp service</primary></indexterm>
<indexterm significance="normal"><primary>/etc/xinetd.d</primary></indexterm>
<indexterm significance="normal"><primary>/etc/inetd.conf</primary></indexterm>
<indexterm significance="normal"><primary>/etc/xinetd.d/telnet</primary></indexterm>
The other services that I modified to allow the use of Winbind as an authentication service were the normal
login on the console (or a terminal session), telnet logins, and ftp service. In order to enable these
services, you may first need to change the entries in <filename moreinfo="none">/etc/xinetd.d</filename> (or
<filename moreinfo="none">/etc/inetd.conf</filename>).  Red Hat Linux 7.1 and later uses the new xinetd.d structure, in this
case you need to change the lines in <filename moreinfo="none">/etc/xinetd.d/telnet</filename> and
<filename moreinfo="none">/etc/xinetd.d/wu-ftp</filename> from:
<programlisting format="linespecific">
	enable = no
</programlisting>
to
<programlisting format="linespecific">
	enable = yes
</programlisting></para>

<para>
<indexterm significance="normal"><primary>ftp services</primary></indexterm>
<indexterm significance="normal"><primary>home directory template</primary></indexterm>
<indexterm significance="normal"><primary>domain users</primary></indexterm>
For ftp services to work properly, you will also need to either have individual directories for the domain
users already present on the server or change the home directory template to a general directory for all
domain users. These can be easily set using the <filename moreinfo="none">smb.conf</filename> global entry <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="TEMPLATEHOMEDIR" xlink:href="smb.conf.5.html#TEMPLATEHOMEDIR">template homedir</link>.
</para>

<note><para>
<indexterm significance="normal"><primary>pam_mkhomedir</primary></indexterm>
The directory in <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="TEMPLATEHOMEDIR" xlink:href="smb.conf.5.html#TEMPLATEHOMEDIR">template homedir</link> is not created automatically! Use pam_mkhomedir or
pre-create the directories of users to make sure users can log in on UNIX with their own home directory.
</para></note>

<para>
<indexterm significance="normal"><primary>/etc/pam.d/ftp</primary></indexterm>
<indexterm significance="normal"><primary>Winbind</primary></indexterm>
<indexterm significance="normal"><primary>ftp access</primary></indexterm>
The <filename moreinfo="none">/etc/pam.d/ftp</filename> file can be changed to allow Winbind ftp access in a manner similar to
the samba file. My <filename moreinfo="none">/etc/pam.d/ftp</filename> file was changed to look like this:
<programlisting format="linespecific">
auth       required     /lib/security/pam_listfile.so item=user sense=deny \
	 file=/etc/ftpusers onerr=succeed
auth       sufficient   /lib/security/pam_winbind.so
auth       required     /lib/security/pam_stack.so service=system-auth
auth       required     /lib/security/pam_shells.so
account    sufficient   /lib/security/pam_winbind.so
account    required     /lib/security/pam_stack.so service=system-auth
session    required     /lib/security/pam_stack.so service=system-auth
</programlisting></para>

<para>
<indexterm significance="normal"><primary>/etc/pam.d/login</primary></indexterm>
The <filename moreinfo="none">/etc/pam.d/login</filename> file can be changed in nearly the same way. It now looks like this:
<programlisting format="linespecific">
auth       required     /lib/security/pam_securetty.so
auth       sufficient   /lib/security/pam_winbind.so
auth       sufficient   /lib/security/pam_unix.so use_first_pass
auth       required     /lib/security/pam_stack.so service=system-auth
auth       required     /lib/security/pam_nologin.so
account    sufficient   /lib/security/pam_winbind.so
account    required     /lib/security/pam_stack.so service=system-auth
password   required     /lib/security/pam_stack.so service=system-auth
session    required     /lib/security/pam_stack.so service=system-auth
session    optional     /lib/security/pam_console.so
</programlisting>
<indexterm significance="normal"><primary>pam_winbind.so</primary></indexterm>
<indexterm significance="normal"><primary>pam_securetty.so</primary></indexterm>
<indexterm significance="normal"><primary>pam_unix.so</primary></indexterm>
In this case, I added the <programlisting format="linespecific">auth sufficient /lib/security/pam_winbind.so</programlisting> lines
as before, but also added the <programlisting format="linespecific">required pam_securetty.so</programlisting> above it to disallow
root logins over the network. I also added a <programlisting format="linespecific">sufficient /lib/security/pam_unix.so
use_first_pass</programlisting> line after the <literal>winbind.so</literal> line to get rid of annoying
double prompts for passwords.
</para>

</sect4>

<sect4>
<title>Solaris-Specific Configuration</title>

<para>
<indexterm significance="normal"><primary>/etc/pam.conf</primary></indexterm>
<indexterm significance="normal"><primary/></indexterm>
The <filename moreinfo="none">/etc/pam.conf</filename> needs to be changed. I changed this file so my Domain
users can log on both locally as well as with telnet. The following are the changes
that I made. You can customize the <filename moreinfo="none">pam.conf</filename> file as per your requirements, but
be sure of those changes because in the worst case it will leave your system
nearly impossible to boot.
<programlisting format="linespecific">
#
#ident "@(#)pam.conf 1.14 99/09/16 SMI"
#
# Copyright (c) 1996-1999, Sun Microsystems, Inc.
# All Rights Reserved.
#
# PAM configuration
#
# Authentication management
#
login   auth required   /usr/lib/security/pam_winbind.so
login auth required  /usr/lib/security/$ISA/pam_unix.so.1 try_first_pass
login auth required  /usr/lib/security/$ISA/pam_dial_auth.so.1 try_first_pass
#
rlogin  auth sufficient /usr/lib/security/pam_winbind.so
rlogin  auth sufficient /usr/lib/security/$ISA/pam_rhosts_auth.so.1
rlogin auth required  /usr/lib/security/$ISA/pam_unix.so.1 try_first_pass
#
dtlogin auth sufficient /usr/lib/security/pam_winbind.so
dtlogin auth required  /usr/lib/security/$ISA/pam_unix.so.1 try_first_pass
#
rsh auth required /usr/lib/security/$ISA/pam_rhosts_auth.so.1
other   auth sufficient /usr/lib/security/pam_winbind.so
other auth required /usr/lib/security/$ISA/pam_unix.so.1 try_first_pass
#
# Account management
#
login   account sufficient      /usr/lib/security/pam_winbind.so
login account requisite /usr/lib/security/$ISA/pam_roles.so.1
login account required /usr/lib/security/$ISA/pam_unix.so.1
#
dtlogin account sufficient      /usr/lib/security/pam_winbind.so
dtlogin account requisite /usr/lib/security/$ISA/pam_roles.so.1
dtlogin account required /usr/lib/security/$ISA/pam_unix.so.1
#
other   account sufficient      /usr/lib/security/pam_winbind.so
other account requisite /usr/lib/security/$ISA/pam_roles.so.1
other account required /usr/lib/security/$ISA/pam_unix.so.1
#
# Session management
#
other session required /usr/lib/security/$ISA/pam_unix.so.1
#
# Password management
#
#other   password sufficient     /usr/lib/security/pam_winbind.so
other password required /usr/lib/security/$ISA/pam_unix.so.1
dtsession auth required /usr/lib/security/$ISA/pam_unix.so.1
#
# Support for Kerberos V5 authentication (uncomment to use Kerberos)
#
#rlogin auth optional /usr/lib/security/$ISA/pam_krb5.so.1 try_first_pass
#login auth optional /usr/lib/security/$ISA/pam_krb5.so.1 try_first_pass
#dtlogin auth optional /usr/lib/security/$ISA/pam_krb5.so.1 try_first_pass
#other auth optional /usr/lib/security/$ISA/pam_krb5.so.1 try_first_pass
#dtlogin account optional /usr/lib/security/$ISA/pam_krb5.so.1
#other account optional /usr/lib/security/$ISA/pam_krb5.so.1
#other session optional /usr/lib/security/$ISA/pam_krb5.so.1
#other password optional /usr/lib/security/$ISA/pam_krb5.so.1 try_first_pass
</programlisting></para>

<para>
<indexterm significance="normal"><primary>winbind.so</primary></indexterm>
I also added a <parameter moreinfo="none">try_first_pass</parameter> line after the <filename moreinfo="none">winbind.so</filename>
line to get rid of annoying double prompts for passwords.
</para>

<para>
Now restart your Samba and try connecting through your application that you
configured in the pam.conf.
</para>

</sect4>

</sect3>

</sect2>

</sect1>

<sect1>
<title>Conclusion</title>

<para>
<indexterm significance="normal"><primary>Winbind</primary></indexterm>
<indexterm significance="normal"><primary>NSS</primary></indexterm>
<indexterm significance="normal"><primary>PAM</primary></indexterm>
<indexterm significance="normal"><primary>RPC calls</primary></indexterm>
<indexterm significance="normal"><primary>domain users</primary></indexterm>
The Winbind system, through the use of the NSS, PAMs, and appropriate Microsoft RPC calls, have allowed us to
provide seamless integration of Microsoft Windows NT domain users on a UNIX system. The result is a great
reduction in the administrative cost of running a mixed UNIX and NT network.
</para>

</sect1>

<sect1>
<title>Common Errors</title>

	<para>
	Winbind has a number of limitations in its current released version that we hope to overcome in future releases:
	</para>

	<itemizedlist>
		<listitem><para>
		Winbind is currently only available for the Linux, Solaris, AIX, and IRIX operating systems, although
		ports to other operating systems are certainly possible. For such ports to be feasible, we require the C
		library of the target operating system to support the NSS and PAM systems. This is becoming more common as NSS
		and PAM gain support among UNIX vendors.
		</para></listitem>

		<listitem><para>
		The mappings of Windows NT RIDs to UNIX IDs is not made algorithmically and depends on the order in
		which unmapped users or groups are seen by Winbind. It may be difficult to recover the mappings of RID to UNIX
		ID if the file containing this information is corrupted or destroyed.
		</para></listitem>

		<listitem><para>
		Currently the Winbind PAM module does not take into account possible workstation and logon time
		restrictions that may be set for Windows NT users; this is instead up to the PDC to enforce.
		</para></listitem>
	</itemizedlist>

	<sect2>
	<title>NSCD Problem Warning</title>

	<warning><para>
	Do not under any circumstances run <literal>nscd</literal> on any system
	on which <literal>winbindd</literal> is running.
	</para></warning>

	<para>
	If <literal>nscd</literal> is running on the UNIX/Linux system, then
	even though NSSWITCH is correctly configured, it will not be possible to resolve
	domain users and groups for file and directory controls.
	</para>

	</sect2>

	<sect2>
	<title>Winbind Is Not Resolving Users and Groups</title>

	<para><quote>
	My <filename moreinfo="none">smb.conf</filename> file is correctly configured. I have specified <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="IDMAPUID" xlink:href="smb.conf.5.html#IDMAPUID">idmap uid = 12000</link>,
	and <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="IDMAPGID" xlink:href="smb.conf.5.html#IDMAPGID">idmap gid = 3000-3500</link> and <literal>winbind</literal> is running.
	When I do the following, it all works fine.
	</quote></para>

<para><screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">wbinfo -u</userinput>
MIDEARTH\maryo
MIDEARTH\jackb
MIDEARTH\ameds
...
MIDEARTH\root

<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">wbinfo -g</userinput>
MIDEARTH\Domain Users
MIDEARTH\Domain Admins
MIDEARTH\Domain Guests
...
MIDEARTH\Accounts

<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">getent passwd</userinput>
root:x:0:0:root:/root:/bin/bash
bin:x:1:1:bin:/bin:/bin/bash
...
maryo:x:15000:15003:Mary Orville:/home/MIDEARTH/maryo:/bin/false
</screen></para>

<para><quote>
But the following command just fails:
</quote>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">chown maryo a_file</userinput>
chown: `maryo': invalid user
</screen>
<quote>
This is driving me nuts! What can be wrong?
</quote></para>

<para>
Same problem as the one above.
Your system is likely running <literal>nscd</literal>, the name service
caching daemon. Shut it down, do not restart it! You will find your problem resolved.
Alternately, fix the operation of nscd to resolve the problem.
</para>

</sect2>
</sect1>

</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="AdvancedNetworkManagement">
<chapterinfo>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
    <pubdate>June 15 2005</pubdate>
</chapterinfo>

<title>Advanced Network Management</title>

<para>
<indexterm significance="normal"><primary>access control</primary></indexterm>
This section documents peripheral issues that are of great importance to network
administrators who want to improve network resource access control, to automate the user
environment, and to make their lives a little easier.
</para>

<sect1>
<title>Features and Benefits</title>

<para>
Often the difference between a working network environment and a well-appreciated one can
best be measured by the <emphasis>little things</emphasis> that make everything work more
harmoniously. A key part of every network environment solution is the ability to remotely
manage MS Windows workstations, remotely access the Samba server, provide customized
logon scripts, as well as other housekeeping activities that help to sustain more reliable
network operations.
</para>

<para>
This chapter presents information on each of these areas. They are placed here, and not in
other chapters, for ease of reference.
</para>

</sect1>

<sect1>
<title>Remote Server Administration</title>


<para><quote>How do I get User Manager and Server Manager?</quote></para>

<para>
<indexterm significance="normal"><primary>User Manager</primary></indexterm>
<indexterm significance="normal"><primary>Server Manager</primary></indexterm>
<indexterm significance="normal"><primary>Event Viewer</primary></indexterm>
Since I do not need to buy an <application moreinfo="none">NT4 server</application>, how do I get the User Manager for Domains
and the Server Manager?
</para>

<para>
<indexterm significance="normal"><primary>Nexus.exe</primary></indexterm>
<indexterm significance="normal"><primary>Windows 9x/Me</primary></indexterm>
Microsoft distributes a version of these tools called <filename moreinfo="none">Nexus.exe</filename> for installation 
on <application moreinfo="none">Windows 9x/Me</application> systems. The tools set includes:
</para>

<itemizedlist>
	<listitem><para>Server Manager</para></listitem>
	<listitem><para>User Manager for Domains</para></listitem>
	<listitem><para>Event Viewer</para></listitem>
</itemizedlist>

<para>
Download the archived file at the Microsoft <ulink noescape="1" url="ftp://ftp.microsoft.com/Softlib/MSLFILES/NEXUS.EXE">Nexus</ulink> link.
</para>

<para>
<indexterm significance="normal"><primary>SRVTOOLS.EXE</primary></indexterm>
<indexterm significance="normal"><primary>User Manager for Domains</primary></indexterm>
<indexterm significance="normal"><primary>Server Manager</primary></indexterm>
The <application moreinfo="none">Windows NT 4.0</application> version of the User Manager for 
Domains and Server Manager are available from Microsoft
<ulink url="ftp://ftp.microsoft.com/Softlib/MSLFILES/SRVTOOLS.EXE">via ftp</ulink>.
</para>

</sect1>

<sect1>
<title>Remote Desktop Management</title>

<para>
<indexterm significance="normal"><primary>remote desktop management</primary></indexterm>
<indexterm significance="normal"><primary>network environment</primary></indexterm>
There are a number of possible remote desktop management solutions that range from free
through costly. Do not let that put you off. Sometimes the most costly solution is the
most cost effective. In any case, you will need to draw your own conclusions as to which
is the best tool in your network environment.
</para>

	<sect2>
	<title>Remote Management from NoMachine.Com</title>

	<para>
	<indexterm significance="normal"><primary>NoMachine.Com</primary></indexterm>
	The following information was posted to the Samba mailing list at Apr 3 23:33:50 GMT 2003.
	It is presented in slightly edited form (with author details omitted for privacy reasons).
	The entire answer is reproduced below with some comments removed.
	</para>

		<para><quote>
<indexterm significance="normal"><primary>remote desktop capabilities</primary></indexterm>
		I have a wonderful Linux/Samba server running as PDC for a network. Now I would like to add remote
		desktop capabilities so users outside could login to the system and get their desktop up from home or
		another country.
		</quote></para>

		<para><quote>
<indexterm significance="normal"><primary>Windows Terminal server</primary></indexterm>
<indexterm significance="normal"><primary>BDC</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>remote login</primary></indexterm>
		Is there a way to accomplish this? Do I need a Windows Terminal server?  Do I need to configure it so
		it is a member of the domain or a BDC or PDC? Are there any hacks for MS Windows XP to enable remote login
		even if the computer is in a domain?
		</quote></para>

		<para>
		Answer provided: Check out the new offer of <quote>NX</quote> software from
		<ulink noescape="1" url="http://www.nomachine.com/">NoMachine</ulink>.
		</para>

	<para>
<indexterm significance="normal"><primary>Remote X protocol</primary></indexterm>
<indexterm significance="normal"><primary>VNC/RFB</primary></indexterm>
<indexterm significance="normal"><primary>rdesktop/RDP</primary></indexterm>
	It implements an easy-to-use interface to the Remote X protocol as
	well as incorporating VNC/RFB and rdesktop/RDP into it, but at a speed
	performance much better than anything you may have ever seen.
	</para>

	<para>
<indexterm significance="normal"><primary>modem/ISDN</primary></indexterm>
	Remote X is not new at all, but what they did achieve successfully is
	a new way of compression and caching technologies that makes the thing
	fast enough to run even over slow modem/ISDN connections.
	</para>

	<para>
<indexterm significance="normal"><primary>KDE konqueror</primary></indexterm>
<indexterm significance="normal"><primary>mouse-over</primary></indexterm>
<indexterm significance="normal"><primary>rdesktop</primary></indexterm>
<indexterm significance="normal"><primary/></indexterm>
	I test drove their (public) Red Hat machine in Italy, over a loaded
	Internet connection, with enabled thumbnail previews in KDE konqueror,
	which popped up immediately on <quote>mouse-over</quote>. From inside that (remote X)
	session I started a rdesktop session on another, a Windows XP machine.
	To test the performance, I played Pinball. I am proud to announce
	that my score was 631,750 points at first try.
	</para>

	<para>
<indexterm significance="normal"><primary>NX</primary></indexterm>
<indexterm significance="normal"><primary>TightVNC</primary></indexterm>
<indexterm significance="normal"><primary>rdesktop</primary></indexterm>
<indexterm significance="normal"><primary>Remote X</primary></indexterm>
	NX performs better on my local LAN than any of the other <quote>pure</quote>
	connection methods I use from time to time: TightVNC, rdesktop or
	Remote X. It is even faster than a direct crosslink connection between
	two nodes.
	</para>

	<para>
<indexterm significance="normal"><primary>Remote X</primary></indexterm>
<indexterm significance="normal"><primary>KDE session</primary></indexterm>
<indexterm significance="normal"><primary>copy'n'paste</primary></indexterm>
	I even got sound playing from the Remote X app to my local boxes, and
	had a working <quote>copy'n'paste</quote> from an NX  window (running a KDE session
	in Italy) to my Mozilla mailing agent. These guys are certainly doing
	something right!
	</para>

	<para>
	I recommend test driving NX to anybody with a only a passing interest in remote computing
	the <ulink noescape="1" url="http://www.nomachine.com/testdrive.php">NX</ulink> utility.
	</para>

	<para>
	Just download the free-of-charge client software (available for Red Hat,
	SuSE, Debian and Windows) and be up and running within 5 minutes (they
	need to send you your account data, though, because you are assigned
	a real UNIX account on their testdrive.nomachine.com box).
	</para>

	<para>
	They plan to get to the point were you can have NX application servers
	running as a cluster of nodes, and users simply start an NX session locally
	and can select applications to run transparently (apps may even run on
	another NX node, but pretend to be on the same as used for initial login,
	because it displays in the same window. You also can run it
	full-screen, and after a short time you forget that it is a remote session
	at all).
	</para>

	<para>
<indexterm significance="normal"><primary>GPL</primary></indexterm>
	Now the best thing for last: All the core compression and caching
	technologies are released under the GPL and available as source code
	to anybody who wants to build on it! These technologies are working,
	albeit started from the command line only (and very inconvenient to
	use in order to get a fully running remote X session up and running).
	</para>

	<para>
	To answer your questions:
	</para>

	<itemizedlist>
		<listitem><para>
		You do not need to install a terminal server; XP has RDP support built in.
		</para></listitem>

		<listitem><para>
		NX is much cheaper than Citrix <?latex --- ?> and comparable in performance, probably faster.
		</para></listitem>

		<listitem><para>
		You do not need to hack XP <?latex --- ?> it just works.
		</para></listitem>

		<listitem><para>
		You log into the XP box from remote transparently (and I think there is no
		need to change anything to get a connection, even if authentication is against a domain).
		</para></listitem>

		<listitem><para>
		The NX core technologies are all Open Source and released under the GPL <?latex --- ?>
		you can now use a (very inconvenient) command line at no cost,
		but you can buy a comfortable (proprietary) NX GUI front end for money.
		</para></listitem>

		<listitem><para>
<indexterm significance="normal"><primary>OSS/Free Software</primary></indexterm>
<indexterm significance="normal"><primary>LTSP</primary></indexterm>
<indexterm significance="normal"><primary>KDE</primary></indexterm>
<indexterm significance="normal"><primary>GNOME</primary></indexterm>
<indexterm significance="normal"><primary>NoMachine</primary></indexterm>
		NoMachine is encouraging and offering help to OSS/Free Software implementations
		for such a front-end too, even if it means competition to them (they have written
		to this effect even to the LTSP, KDE, and GNOME developer mailing lists).
		</para></listitem>
	</itemizedlist>

	</sect2>
	<sect2>
	<title>Remote Management with ThinLinc</title>
	<para>
	Another alternative for remote access is <emphasis>ThinLinc</emphasis> from Cendio.
	</para>

	<para>
<indexterm significance="normal"><primary>ThinLinc</primary></indexterm>
<indexterm significance="normal"><primary>terminal server</primary></indexterm>
<indexterm significance="normal"><primary>Linux</primary></indexterm>
<indexterm significance="normal"><primary>Solaris</primary></indexterm>
<indexterm significance="normal"><primary>TightVNC</primary></indexterm>
<indexterm significance="normal"><primary>SSH</primary></indexterm>
<indexterm significance="normal"><primary>NFS</primary></indexterm>
<indexterm significance="normal"><primary>PulseAudio</primary></indexterm>
	ThinLinc is a terminal server solution that is available for Linux and Solaris based on standard
	protocols such as SSH, TightVNC, NFS and PulseAudio.
	</para>

	<para>
<indexterm significance="normal"><primary>LAN</primary></indexterm>
<indexterm significance="normal"><primary>thin client</primary></indexterm>
	ThinLinc an be used both in the LAN environment to implement a Thin Client strategy for an organization, and as
	secure remote access solution for people working from remote locations, even over smallband connections.
	ThinLinc is free to use for a single concurrent user.
	</para>

	<para>
<indexterm significance="normal"><primary>Citrix</primary></indexterm>
<indexterm significance="normal"><primary>Windows Terminal Server</primary></indexterm>
<indexterm significance="normal"><primary>Java</primary></indexterm>
	The product can also be used as a frontend to access Windows Terminal Server or Citrix farms, or even Windows
	XP machines, securing the connection via the ssh protocol. The client is available both for Linux (supporting
	all Linux distributions as well as numerous thin terminals) and for Windows. A Java-based Web client is also
	available.
	</para>

	<para>
	ThinLinc may be evaluated by connecting to Cendio's demo system, see
	<ulink noescape="1" url="http://www.cendio.com">Cendio's</ulink> web site
	<ulink noescape="1" url="http://www.cendio.com/testdrive">testdrive</ulink> center.
	</para>

	<para>
	Cendio is a major contributor to several open source projects including
	<ulink noescape="1" url="http://www.tightvnc.com">TightVNC</ulink>,
	<ulink noescape="1" url="http://pulseaudio.org">PulseAudio</ulink> , unfsd,
	<ulink noescape="1" url="http://www.python.org">Python</ulink> and
	<ulink noescape="1" url="http://www.rdesktop.org">rdesktop</ulink>.
	</para>

	</sect2>
</sect1>

<sect1>
<title>Network Logon Script Magic</title>

<para>
There are several opportunities for creating a custom network startup configuration environment.
</para>

<itemizedlist>
	<listitem><para>No Logon Script.</para></listitem>
	<listitem><para>Simple universal Logon Script that applies to all users.</para></listitem>
	<listitem><para>Use of a conditional Logon Script that applies per-user or per-group attributes.</para></listitem>
	<listitem><para>Use of Samba's preexec and postexec functions on access to the NETLOGON share to create
		a custom logon script and then execute it.</para></listitem>
	<listitem><para>User of a tool such as KixStart.</para></listitem>
</itemizedlist>

<para>
The Samba source code tree includes two logon script generation/execution tools.
See <filename moreinfo="none">examples</filename> directory <filename moreinfo="none">genlogon</filename> and
<filename moreinfo="none">ntlogon</filename> subdirectories.
</para>

<para>
The following listings are from the genlogon directory.
</para>


<para>
<indexterm significance="normal"><primary>genlogon.pl</primary></indexterm>
This is the <filename moreinfo="none">genlogon.pl</filename> file:

<programlisting format="linespecific">
	#!/usr/bin/perl
	#
	# genlogon.pl
	#
	# Perl script to generate user logon scripts on the fly, when users
	# connect from a Windows client. This script should be called from 
	# smb.conf with the %U, %G and %L parameters. I.e:
	#
	#       root preexec = genlogon.pl %U %G %L
	#
	# The script generated will perform
	# the following:
	#
	# 1. Log the user connection to /var/log/samba/netlogon.log
	# 2. Set the PC's time to the Linux server time (which is maintained
	#    daily to the National Institute of Standards Atomic clock on the
	#    internet.
	# 3. Connect the user's home drive to H: (H for Home).
	# 4. Connect common drives that everyone uses.
	# 5. Connect group-specific drives for certain user groups.
	# 6. Connect user-specific drives for certain users.
	# 7. Connect network printers.

	# Log client connection
	#($sec,$min,$hour,$mday,$mon,$year,$wday,$yday,$isdst) = localtime(time);
	($sec,$min,$hour,$mday,$mon,$year,$wday,$yday,$isdst) = localtime(time);
	open LOG, "&gt;&gt;/var/log/samba/netlogon.log";
	print LOG "$mon/$mday/$year $hour:$min:$sec";
	print LOG " - User $ARGV[0] logged into $ARGV[1]\n";
	close LOG;

	# Start generating logon script
	open LOGON, "&gt;/shared/netlogon/$ARGV[0].bat";
	print LOGON "\@ECHO OFF\r\n";

	# Connect shares just use by Software Development group
	if ($ARGV[1] eq "SOFTDEV" || $ARGV[0] eq "softdev")
	{
		print LOGON "NET USE M: \\\\$ARGV[2]\\SOURCE\r\n";
	}

	# Connect shares just use by Technical Support staff
	if ($ARGV[1] eq "SUPPORT" || $ARGV[0] eq "support")
	{
		print LOGON "NET USE S: \\\\$ARGV[2]\\SUPPORT\r\n";
	}

	# Connect shares just used by Administration staff
	If ($ARGV[1] eq "ADMIN" || $ARGV[0] eq "admin")
	{
		print LOGON "NET USE L: \\\\$ARGV[2]\\ADMIN\r\n";
		print LOGON "NET USE K: \\\\$ARGV[2]\\MKTING\r\n";
	}

	# Now connect Printers. We handle just two or three users a little
	# differently, because they are the exceptions that have desktop
	# printers on LPT1: - all other user's go to the LaserJet on the
	# server.
	if ($ARGV[0] eq 'jim'
	    || $ARGV[0] eq 'yvonne')
	{
		print LOGON "NET USE LPT2: \\\\$ARGV[2]\\LJET3\r\n";
		print LOGON "NET USE LPT3: \\\\$ARGV[2]\\FAXQ\r\n";
	}
	else
	{
		print LOGON "NET USE LPT1: \\\\$ARGV[2]\\LJET3\r\n";
		print LOGON "NET USE LPT3: \\\\$ARGV[2]\\FAXQ\r\n";
	}

	# All done! Close the output file.
	close LOGON;
</programlisting>
</para>

<para>
Those wishing to use a more elaborate or capable logon processing system should check out these sites:
</para>

<itemizedlist>
	<listitem><para><ulink noescape="1" url="http://www.craigelachie.org/rhacer/ntlogon">http://www.craigelachie.org/rhacer/ntlogon</ulink></para></listitem>
	<listitem><para><ulink noescape="1" url="http://www.kixtart.org">http://www.kixtart.org</ulink></para></listitem>
</itemizedlist>

<sect2>
<title>Adding Printers without User Intervention</title>


<para>
<indexterm significance="normal"><primary>rundll32</primary></indexterm>
Printers may be added automatically during logon script processing through the use of:
<screen format="linespecific">
<prompt moreinfo="none">C:\&gt; </prompt><userinput moreinfo="none">rundll32 printui.dll,PrintUIEntry /?</userinput>
</screen>

See the documentation in the <ulink url="http://support.microsoft.com/default.asp?scid=kb;en-us;189105">Microsoft Knowledge Base article 189105</ulink>.
</para>
</sect2>

<sect2>
	<title>Limiting Logon Connections</title>

	<para>
		Sometimes it is necessary to limit the number of concurrent connections to a
		Samba shared resource. For example, a site may wish to permit only one network
		logon per user.
	</para>

	<para>
		The Samba <parameter moreinfo="none">preexec script</parameter> parameter can be used to permit only one
		connection per user. Though this method is not foolproof and may have side effects,
		the following contributed method may inspire someone to provide a better solution.
	</para>

	<para>
		This is not a perfect solution because Windows clients can drop idle connections
		with an auto-reconnect capability that could result in the appearance that a share
		is no longer in use, while actually it is. Even so, it demonstrates the principle
		of use of the <parameter moreinfo="none">preexec script</parameter> parameter.
	</para>

	<para>
		The following share configuration demonstrates use of the script shown in <link linkend="Tpees"/>.
<programlisting format="linespecific">
[myshare]
	...
	preexec script = /sbin/PermitSingleLogon.sh
	preexec close = Yes
	...
</programlisting>
	</para>

<example id="Tpees">
<title>Script to Enforce Single Resource Logon</title>
<screen format="linespecific">
#!/bin/bash

IFS="-"
RESULT=$(smbstatus -S -u $1 2&gt; /dev/null | awk 'NF \
        &gt; 6 {print $1}' | sort | uniq -d)

if [ "X${RESULT}" == X  ]; then
  exit 0
else
  exit 1
fi
</screen>
</example>

</sect2>

</sect1>

</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="PolicyMgmt">
<chapterinfo>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
    <pubdate>April 3 2003</pubdate>
</chapterinfo>

<title>System and Account Policies</title>

<para>
<indexterm significance="normal"><primary>validation</primary></indexterm>
This chapter summarizes the current state of knowledge derived from personal
practice and knowledge from Samba mailing list subscribers. Before reproduction
of posted information, every effort has been made to validate the information given.
Where additional information was uncovered through this validation, it is provided
also.
</para>

<sect1>
<title>Features and Benefits</title>

<para>
<indexterm significance="normal"><primary>Group Policies</primary></indexterm>
<indexterm significance="normal"><primary>users</primary></indexterm>
<indexterm significance="normal"><primary>groups</primary></indexterm>
When MS Windows NT 3.5 was introduced, the hot new topic was the ability to implement
Group Policies for users and groups. Then along came MS Windows NT4 and a few sites
started to adopt this capability. How do we know that? By the number of <quote>boo-boos</quote>
(or mistakes) administrators made and then requested help to resolve.
</para>

<para>
<indexterm significance="normal"><primary>group policies</primary></indexterm>
<indexterm significance="normal"><primary>Group Policy Objects</primary><see>GPO</see></indexterm>
<indexterm significance="normal"><primary>GPOs</primary></indexterm>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
<indexterm significance="normal"><primary>group policy objects</primary><see>GPOs</see></indexterm>
By the time that MS Windows 2000 and Active Directory was released, administrators
got the message: Group Policies are a good thing! They can help reduce administrative
costs and actually make happier users. But adoption of the true
potential of MS Windows 200x Active Directory and Group Policy Objects (GPOs) for users
and machines were picked up on rather slowly. This was obvious from the Samba
mailing list back in 2000 and 2001 when there were few postings regarding GPOs and
how to replicate them in a Samba environment.
</para>

<para>
<indexterm significance="normal"><primary>exploit opportunities</primary></indexterm>
Judging by the traffic volume since mid 2002, GPOs have become a standard part of
the deployment in many sites. This chapter reviews techniques and methods that can
be used to exploit opportunities for automation of control over user desktops and
network client workstations.
</para>

</sect1>

<sect1>
<title>Creating and Managing System Policies</title>

<para>
<indexterm significance="normal"><primary>NETLOGON</primary></indexterm>
<indexterm significance="normal"><primary>domain controller</primary></indexterm>
<indexterm significance="normal"><primary>registry</primary></indexterm>
<indexterm significance="normal"><primary>affect users</primary></indexterm>
Under MS Windows platforms, particularly those following the release of MS Windows
NT4 and MS Windows 95, it is possible to create a type of file that would be placed
in the NETLOGON share of a domain controller. As the client logs onto the network,
this file is read and the contents initiate changes to the registry of the client
machine. This file allows changes to be made to those parts of the registry that
affect users, groups of users, or machines.
</para>

<para>
<indexterm significance="normal"><primary>Config.POL</primary></indexterm>
<indexterm significance="normal"><primary>poledit.exe</primary></indexterm>
<indexterm significance="normal"><primary>policy editor</primary></indexterm>
For MS Windows 9x/Me, this file must be called <filename moreinfo="none">Config.POL</filename> and may
be generated using a tool called <filename moreinfo="none">poledit.exe</filename>, better known as the
Policy Editor. The policy editor was provided on the Windows 98 installation CD-ROM, but
disappeared again with the introduction of MS Windows Me. From
comments of MS Windows network administrators, it would appear that this tool became
a part of the MS Windows Me Resource Kit.
</para>

<para>
<indexterm significance="normal"><primary>System Policy Editor</primary></indexterm>
MS Windows NT4 server products include the <emphasis>System Policy Editor</emphasis>
under <guimenu moreinfo="none">Start -&gt; Programs -&gt; Administrative Tools</guimenu>.
For MS Windows NT4 and later clients, this file must be called <filename moreinfo="none">NTConfig.POL</filename>.
</para>

<para>
<indexterm significance="normal"><primary>MMC</primary></indexterm>
New with the introduction of MS Windows 2000 was the Microsoft Management Console
or MMC. This tool is the new wave in the ever-changing landscape of Microsoft
methods for management of network access and security. Every new Microsoft product
or technology seems to make the old rules obsolete and introduces newer and more
complex tools and methods. To Microsoft's credit, the MMC does appear to
be a step forward, but improved functionality comes at a great price.
</para>

<para>
<indexterm significance="normal"><primary>network policies</primary></indexterm>
<indexterm significance="normal"><primary>system policies</primary></indexterm>
<indexterm significance="normal"><primary>Profiles</primary></indexterm>
<indexterm significance="normal"><primary>Policies</primary></indexterm>
Before embarking on the configuration of network and system policies, it is highly
advisable to read the documentation available from Microsoft's Web site regarding
<ulink url="http://www.microsoft.com/ntserver/techresources/management/prof_policies.asp">
Implementing Profiles and Policies in Windows NT 4.0</ulink>.
There are a large number of documents in addition to this old one that should also
be read and understood. Try searching on the Microsoft Web site for <quote>Group Policies</quote>.
</para>

<para>
What follows is a brief discussion with some helpful notes. The information provided
here is incomplete <?latex --- ?> you are warned.
</para>

	<sect2>
	<title>Windows 9x/ME Policies</title>

	<para>
<indexterm significance="normal"><primary>Group Policy Editor</primary></indexterm>
<indexterm significance="normal"><primary>tools\reskit\netadmin\poledit</primary></indexterm>
	You need the Windows 98 Group Policy Editor to set up Group Profiles under Windows 9x/Me.
	It can be found on the original full-product Windows 98 installation CD-ROM under
	<filename moreinfo="none">tools\reskit\netadmin\poledit</filename>. Install this using the
	Add/Remove Programs facility, and then click on <guiicon moreinfo="none">Have Disk</guiicon>.
	</para>


	<para>
<indexterm significance="normal"><primary>NTConfig.POL</primary></indexterm>
<indexterm significance="normal"><primary>Config.POL</primary></indexterm>
	Use the Group Policy Editor to create a policy file that specifies the location of
	user profiles and/or <filename moreinfo="none">My Documents</filename>, and so on. Then save these
	settings in a file called <filename moreinfo="none">Config.POL</filename> that needs to be placed in the
	root of the <parameter>[NETLOGON]</parameter> share. If Windows 98 is configured to log onto
	the Samba domain, it will automatically read this file and update the Windows 9x/Me registry
	of the machine as it logs on.
	</para>

	<para>
	Further details are covered in the Windows 98 Resource Kit documentation.
	</para>

	<para>
<indexterm significance="normal"><primary>registry</primary></indexterm>
	If you do not take the correct steps, then every so often Windows 9x/Me will check the
	integrity of the registry and restore its settings from the backup
	copy of the registry it stores on each Windows 9x/Me machine. So, you will
	occasionally notice things changing back to the original settings.
	</para>

	<para>
<indexterm significance="normal"><primary>grouppol.inf</primary></indexterm>
<indexterm significance="normal"><primary>Group Policy</primary></indexterm>
	Install the Group Policy handler for Windows 9x/Me to pick up Group Policies. Look on the
	Windows 98 CD-ROM in <filename moreinfo="none">\tools\reskit\netadmin\poledit</filename>.
	Install Group Policies on a Windows 9x/Me client by double-clicking on
	<filename moreinfo="none">grouppol.inf</filename>. Log off and on again a couple of times and see
	if Windows 98 picks up Group Policies. Unfortunately, this needs to be done on every
	Windows 9x/Me machine that uses Group Policies.
	</para>

	</sect2>
	<sect2>
	<title>Windows NT4-Style Policy Files</title>

	<para>
<indexterm significance="normal"><primary>ntconfig.pol</primary></indexterm>
<indexterm significance="normal"><primary>poledit.exe</primary></indexterm>
<indexterm significance="normal"><primary>Policy Editor</primary></indexterm>
<indexterm significance="normal"><primary>domain policies</primary></indexterm>
	To create or edit <filename moreinfo="none">ntconfig.pol</filename>, you must use the NT Server
	Policy Editor, <literal>poledit.exe</literal>, which is included with NT4 Server
	but not with NT workstation. There is a Policy Editor on an NT4
	Workstation but it is not suitable for creating domain policies.
	Furthermore, although the Windows 95 Policy Editor can be installed on an NT4
	workstation/server, it will not work with NT clients. However, the files from
	the NT Server will run happily enough on an NT4 workstation.
	</para>

	<para>
<indexterm significance="normal"><primary>poledit.exe</primary></indexterm>
<indexterm significance="normal"><primary>common.adm</primary></indexterm>
<indexterm significance="normal"><primary>winnt.adm</primary></indexterm>
<indexterm significance="normal"><primary>c:\winnt\inf</primary></indexterm>
	You need <filename moreinfo="none">poledit.exe</filename>, <filename moreinfo="none">common.adm</filename>, and <filename moreinfo="none">winnt.adm</filename>.
	It is convenient to put the two <filename moreinfo="none">*.adm</filename> files in  the <filename moreinfo="none">c:\winnt\inf</filename> 
	directory, which is where the binary will look for them unless told otherwise. This
	directory is normally <quote>hidden.</quote>
	</para>

	<para>
<indexterm significance="normal"><primary>Policy Editor</primary></indexterm>
<indexterm significance="normal"><primary>Nt4sp6ai.exe</primary></indexterm>
<indexterm significance="normal"><primary>poledit.exe</primary></indexterm>
<indexterm significance="normal"><primary>Zero Administration Kit</primary></indexterm>
	The Windows NT Policy Editor is also included with the Service Pack 3 (and
	later) for Windows NT 4.0. Extract the files using <literal>servicepackname /x</literal>
	<?latex --- ?> that's <literal>Nt4sp6ai.exe /x</literal> for Service Pack 6a. The Policy Editor,
	<literal>poledit.exe</literal>, and the associated template files (*.adm) should
	be extracted as well. It is also possible to download the policy template
	files for Office97 and get a copy of the Policy Editor. Another possible
	location is with the Zero Administration Kit available for download from Microsoft.
	</para>

		<sect3>
		<title>Registry Spoiling</title>

			<para>
<indexterm significance="normal"><primary>NTConfig.POL</primary></indexterm>
<indexterm significance="normal"><primary>HKEY_LOCAL_MACHINE</primary></indexterm>
			With NT4-style registry-based policy changes, a large number of settings are not
			automatically reversed as the user logs off. The settings that were in the
			<filename moreinfo="none">NTConfig.POL</filename> file were applied to the client machine registry and apply to the
			hive key HKEY_LOCAL_MACHINE are permanent until explicitly reversed. This is known
			as tattooing. It can have serious consequences downstream, and the administrator must
			be extremely careful not to lock out the ability to manage the machine at a later date.
			</para>

		</sect3>
	</sect2>
	<sect2>
	<title>MS Windows 200x/XP Professional Policies</title>

	<para>
<indexterm significance="normal"><primary>registry</primary></indexterm>
	Windows NT4 system policies allow the setting of registry parameters specific to
	users, groups, and computers (client workstations) that are members of the NT4-style
	domain. Such policy files will work with MS Windows 200x/XP clients also.
	</para>

	<para>
	New to MS Windows 2000, Microsoft recently introduced a style of Group Policy that confers
	a superset of capabilities compared with NT4-style policies. Obviously, the tool used
	to create them is different, and the mechanism for implementing them is much improved.
	</para>

	<para>
	<indexterm significance="normal"><primary>GPOs</primary></indexterm>
<indexterm significance="normal"><primary>Administrative Templates</primary></indexterm>
	The older NT4-style registry-based policies are known as <emphasis>Administrative Templates</emphasis>
	in MS Windows 2000/XP GPOs. The latter includes the ability to set various security
	configurations, enforce Internet Explorer browser settings, change and redirect aspects of the
	users desktop (including the location of <filename moreinfo="none">My Documents</filename> files, as
	well as intrinsics of where menu items will appear in the Start menu). An additional new
	feature is the ability to make available particular software Windows applications to particular
	users and/or groups.
	</para>

	<para>
<indexterm significance="normal"><primary>NTConfig.POL</primary></indexterm>
<indexterm significance="normal"><primary>NETLOGON</primary></indexterm>
<indexterm significance="normal"><primary>local registry values</primary></indexterm>
	Remember, NT4 policy files are named <filename moreinfo="none">NTConfig.POL</filename> and are stored in the root
	of the NETLOGON share on the domain controllers. A Windows NT4 user enters a username and password
	and selects the domain name to which the logon will attempt to take place. During the logon process,
	the client machine reads the <filename moreinfo="none">NTConfig.POL</filename> file from the NETLOGON share on
	the authenticating server and modifies the local registry values according to the settings in this file.
	</para>

	<para>
<indexterm significance="normal"><primary>SYSVOL</primary></indexterm>
<indexterm significance="normal"><primary>NETLOGON</primary></indexterm>
<indexterm significance="normal"><primary>replicated</primary></indexterm>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
<indexterm significance="normal"><primary>domain controllers</primary></indexterm>
<indexterm significance="normal"><primary>Group Policy Container</primary><see>GPC</see></indexterm>
<indexterm significance="normal"><primary>Group Policy Template</primary><see>GPT</see></indexterm>
<indexterm significance="normal"><primary>replicated SYSVOL</primary></indexterm>
	Windows 200x GPOs are feature-rich. They are not stored in the NETLOGON share, but rather part of
	a Windows 200x policy file is stored in the Active Directory itself and the other part is stored
	in a shared (and replicated) volume called the SYSVOL folder. This folder is present on all Active
	Directory domain controllers. The part that is stored in the Active Directory itself is called the
	Group Policy Container (GPC), and the part that is stored in the replicated share called SYSVOL is
	known as the Group Policy Template (GPT).
	</para>

	<para>
<indexterm significance="normal"><primary>GPOs</primary></indexterm>
	With NT4 clients, the policy file is read and executed only as each user logs onto the network.
	MS Windows 200x policies are much more complex <?latex --- ?> GPOs are processed and applied at client machine
	startup (machine specific part), and when the user logs onto the network, the user-specific part
	is applied. In MS Windows 200x-style policy management, each machine and/or user may be subject
	to any number of concurrently applicable (and applied) policy sets (GPOs). Active Directory allows
	the administrator to also set filters over the policy settings. No such equivalent capability
	exists with NT4-style policy files.
	</para>

		<sect3>
		<title>Administration of Windows 200x/XP Policies</title>

		<para>
		<indexterm significance="normal"><primary>GPOs</primary></indexterm>
		<indexterm significance="normal"><primary>System Policy Editor</primary></indexterm>
<indexterm significance="normal"><primary>poledit.exe</primary></indexterm>
<indexterm significance="normal"><primary>MMC snap-in</primary></indexterm>
<indexterm significance="normal"><primary>Poledit</primary></indexterm>
		Instead of using the tool called <application moreinfo="none">the System Policy Editor</application>, commonly called Poledit (from the
		executable name <literal>poledit.exe</literal>), <acronym>GPOs</acronym> are created and managed using a 
		<application moreinfo="none">Microsoft Management Console</application> <acronym>(MMC)</acronym> snap-in as follows:</para>
		<procedure>
			<step performance="required"><para>
			Go to the Windows 200x/XP menu <guimenu moreinfo="none">Start-&gt;Programs-&gt;Administrative Tools</guimenu>
			and select the MMC snap-in called <guimenuitem moreinfo="none">Active Directory Users and Computers</guimenuitem>
			</para></step>

			<step performance="required"><para>
<indexterm significance="normal"><primary>organizational unit</primary><see>OU</see></indexterm>
			Select the domain or organizational unit (OU) that you wish to manage, then right-click
			to open the context menu for that object, and select the <guibutton moreinfo="none">Properties</guibutton>.
			</para></step>

			<step performance="required"><para>
			Left-click on the <guilabel moreinfo="none">Group Policy</guilabel> tab, then
			left-click on the New tab. Type a name
			for the new policy you will create.
			</para></step>

			<step performance="required"><para>
			 Left-click on the <guilabel moreinfo="none">Edit</guilabel> tab to commence the steps needed to create the GPO.
			</para></step>
		</procedure>

		<para>
		All policy configuration options are controlled through the use of policy administrative
		templates. These files have an .adm extension, both in NT4 as well as in Windows 200x/XP.
		Beware, however, the .adm files are not interchangeable across NT4 and Windows 200x.
		The latter introduces many new features as well as extended definition capabilities. It is
		well beyond the scope of this documentation to explain how to program .adm files; for that,
		refer to the Microsoft Windows Resource Kit for your particular
		version of MS Windows.
		</para>

		<note>
		<para>
<indexterm significance="normal"><primary>gpolmig.exe</primary></indexterm>
<indexterm significance="normal"><primary>NTConfig.POL</primary></indexterm>
<indexterm significance="normal"><primary>resource kit</primary></indexterm>
		The MS Windows 2000 Resource Kit contains a tool called <literal>gpolmig.exe</literal>. This tool can be used
		to migrate an NT4 <filename moreinfo="none">NTConfig.POL</filename> file into a Windows 200x style GPO. Be VERY careful how you
		use this powerful tool. Please refer to the resource kit manuals for specific usage information.
		</para>
		</note>

		</sect3>

		<sect3>
		<title>Custom System Policy Templates</title>

		<para>
		Over the past year, there has been a bit of talk regarding the creation of customized
		templates for the Windows Sytem Policy Editor. A recent announcement on the Samba mailing
		list is worthy of mention.
		</para>

		<para>
		Mike Petersen has announced the availability of a template file he has created. This custom System Policy
		Editor Template will allow you to successfully control Microsoft Windows workstations from an SMB server, such
		as Samba. This template has been tested on a few networks, although if you find any problems with any of these
		policies, or have any ideas for additional policies, let me know at mailto:mgpeter@pcc-services.com.  This
		Template includes many policies for Windows XP to allow it to behave better in a professional environment.
		</para>

		<para>
		For further information please see the <ulink url="http://www.pcc-services.com/custom_poledit.html">Petersen</ulink> Computer Consulting web site. There is
		a download link for the template file.
		</para>

		</sect3>
	</sect2>
</sect1>

<sect1>
<title>Managing Account/User Policies</title>

<para>
<indexterm significance="normal"><primary>Policies</primary></indexterm>
<indexterm significance="normal"><primary>policy file </primary></indexterm>
<indexterm significance="normal"><primary>registry settings</primary></indexterm>
Policies can define a specific user's settings or the settings for a group of users. The resulting
policy file contains the registry settings for all users, groups, and computers that will be using
the policy file. Separate policy files for each user, group, or computer are not necessary.
</para>

<para>
<indexterm significance="normal"><primary>NTConfig.POL</primary></indexterm>
If you create a policy that will be automatically downloaded from validating domain controllers,
you should name the file <filename moreinfo="none">NTConfig.POL</filename>. As system administrator, you have the option of renaming the
policy file and, by modifying the Windows NT-based workstation, directing the computer to update
the policy from a manual path. You can do this by either manually changing the registry or by using
the System Policy Editor. This can even be a local path such that each machine has its own policy file,
but if a change is necessary to all machines, it must be made individually to each workstation.
</para>

<para>
<indexterm significance="normal"><primary>NTConfig.POL</primary></indexterm>
<indexterm significance="normal"><primary>NETLOGON</primary></indexterm>
When a Windows NT4/200x/XP machine logs onto the network, the client looks in the NETLOGON share on
the authenticating domain controller for the presence of the <filename moreinfo="none">NTConfig.POL</filename> file. If one exists, it is 
downloaded, parsed, and then applied to the user's part of the registry.
</para>

<para>
<indexterm significance="normal"><primary>GPOs</primary></indexterm>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
<indexterm significance="normal"><primary>NTConfig.POL</primary></indexterm>
<indexterm significance="normal"><primary>NT4 style policy updates</primary></indexterm>
MS Windows 200x/XP clients that log onto an MS Windows Active Directory security domain may additionally
acquire policy settings through GPOs that are defined and stored in Active Directory
itself. The key benefit of using AD GPOs is that they impose no registry <emphasis>spoiling</emphasis> effect.
This has considerable advantage compared with the use of <filename moreinfo="none">NTConfig.POL</filename> (NT4) style policy updates.
</para>

<para>
<indexterm significance="normal"><primary>account restrictions</primary></indexterm>
<indexterm significance="normal"><primary>Common restrictions</primary></indexterm>
In addition to user access controls that may be imposed or applied via system and/or group policies
in a manner that works in conjunction with user profiles, the user management environment under
MS Windows NT4/200x/XP allows per-domain as well as per-user account restrictions to be applied.
Common restrictions that are frequently used include:
</para>

<para>
<indexterm significance="normal"><primary>Account Controls</primary></indexterm>
<itemizedlist>
	<listitem><para>Logon hours</para></listitem>
	<listitem><para>Password aging</para></listitem>
	<listitem><para>Permitted logon from certain machines only</para></listitem>
	<listitem><para>Account type (local or global)</para></listitem>
	<listitem><para>User rights</para></listitem>
</itemizedlist>
</para>

<para>
<indexterm significance="normal"><primary>Domain User Manager</primary></indexterm>
<indexterm significance="normal"><primary>NTConfig.POL</primary></indexterm>
Samba-3.0.20 does not yet implement all account controls that are common to MS Windows NT4/200x/XP.
While it is possible to set many controls using the Domain User Manager for MS Windows NT4, only password
expiry is functional today. Most of the remaining controls at this time have only stub routines
that may eventually be completed to provide actual control. Do not be misled by the fact that a
parameter can be set using the NT4 Domain User Manager or in the <filename moreinfo="none">NTConfig.POL</filename>.
</para>

</sect1>
<sect1>
<title>Management Tools</title>

<para>
Anyone who wishes to create or manage Group Policies will need to be familiar with a number of tools.
The following sections describe a few key tools that will help you to create a low-maintenance user
environment.
</para>

	<sect2>
	<title>Samba Editreg Toolset</title>

	<para>
	<indexterm significance="normal"><primary>editreg</primary></indexterm>
	<indexterm significance="normal"><primary>NTUser.DAT</primary></indexterm>
	<indexterm significance="normal"><primary>NTConfig.POL</primary></indexterm>
	A new tool called <literal>editreg</literal> is under development. This tool can be used
	to edit registry files (called <filename moreinfo="none">NTUser.DAT</filename>) that are stored in user
	and group profiles. <filename moreinfo="none">NTConfig.POL</filename> files have the same structure as the
	<filename moreinfo="none">NTUser.DAT</filename> file and can be edited using this tool. <literal>editreg</literal>
	is being built with the intent to enable <filename moreinfo="none">NTConfig.POL</filename> files to be saved in text format and to
	permit the building of new <filename moreinfo="none">NTConfig.POL</filename> files with extended capabilities. It is proving difficult
	to realize this capability, so do not be surprised if this feature does not materialize. Formal
	capabilities will be announced at the time that this tool is released for production use.
	</para>

	</sect2>

	<sect2>
	<title>Windows NT4/200x</title>

	<para>
<indexterm significance="normal"><primary>regedt32.exe</primary></indexterm>
<indexterm significance="normal"><primary>Group Policy Editor</primary></indexterm>
<indexterm significance="normal"><primary>MMC</primary></indexterm>
	The tools that may be used to configure these types of controls from the MS Windows environment are
	the NT4 User Manager for Domains, the NT4 System and Group Policy Editor, and the Registry Editor (regedt32.exe).
	Under MS Windows 200x/XP, this is done using the MMC with appropriate
	<quote>snap-ins,</quote> the registry editor, and potentially also the NT4 System and Group Policy Editor.
	</para>
	</sect2>

	<sect2>
	<title>Samba PDC</title>

	<para>
<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
<indexterm significance="normal"><primary>pdbedit</primary></indexterm>
<indexterm significance="normal"><primary>NET</primary></indexterm>
<indexterm significance="normal"><primary>rpcclient</primary></indexterm>
	With a Samba domain controller, the new tools for managing user account and policy information include:
	<literal>smbpasswd</literal>, <literal>pdbedit</literal>, <literal>net</literal>, and <literal>rpcclient</literal>. 
	The administrator should read the man pages for these tools and become familiar with their use.
	</para>

	</sect2>
</sect1>

<sect1>
<title>System Startup and Logon Processing Overview</title>

<para>
The following attempts to document the order of processing the system and user policies following a system
reboot and as part of the user logon:
</para>

<orderedlist inheritnum="ignore" continuation="restarts">
	<listitem><para>
<indexterm significance="normal"><primary>Remote Procedure Call System Service</primary><see>RPCSS</see></indexterm>
<indexterm significance="normal"><primary>multiple universal naming convention provider</primary><see>MUP</see></indexterm>
	Network starts, then Remote Procedure Call System Service (RPCSS) and multiple universal naming
	convention provider (MUP) start.
	</para></listitem>

	<listitem><para>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
<indexterm significance="normal"><primary>GPOs</primary></indexterm>
	Where Active Directory is involved, an ordered list of GPOs is downloaded
	and applied. The list may include GPOs that:
<itemizedlist>
	<listitem><para>Apply to the location of machines in a directory.</para></listitem>
	<listitem><para>Apply only when settings have changed.</para></listitem>
	<listitem><para>Depend on configuration of the scope of applicability: local,
	site, domain, organizational unit, and so on.</para></listitem>
</itemizedlist>
	No desktop user interface is presented until the above have been processed.
	</para></listitem>

	<listitem><para>
	Execution of startup scripts (hidden and synchronous by default).
	</para></listitem>

	<listitem><para>
	A keyboard action to effect start of logon (Ctrl-Alt-Del).
	</para></listitem>

	<listitem><para>
	User credentials are validated, user profile is loaded (depends on policy settings).
	</para></listitem>

	<listitem><para>
	An ordered list of user GPOs is obtained. The list contents depends on what is configured in respect of:

<itemizedlist>
	<listitem><para>Is the user a domain member, thus subject to particular policies?</para></listitem>
	<listitem><para>Loopback enablement, and the state of the loopback policy (merge or replace).</para></listitem>
	<listitem><para>Location of the Active Directory itself.</para></listitem>
	<listitem><para>Has the list of GPOs changed? No processing is needed if not changed.</para></listitem>
</itemizedlist>
	</para></listitem>

	<listitem><para>
	User policies are applied from Active Directory. Note: There are several types.
	</para></listitem>

	<listitem><para>
	Logon scripts are run. New to Windows 200x and Active Directory, logon scripts may be obtained based on GPOs
	(hidden and executed synchronously). NT4-style logon scripts are then run in a normal
	window.
	</para></listitem>

	<listitem><para>
	The user interface as determined from the GPOs is presented. Note: In a Samba domain (like an NT4
	domain), machine (system) policies are applied at startup; user policies are applied at logon.
	</para></listitem>
</orderedlist>

</sect1>

<sect1>
<title>Common Errors</title>

<para>
Policy-related problems can be quite difficult to diagnose and even more difficult to rectify. The following
collection demonstrates only basic issues.
</para>

<sect2>
<title>Policy Does Not Work</title>

<para>
<quote>We have created the <filename moreinfo="none">Config.POL</filename> file and put it in the <emphasis>NETLOGON</emphasis> share.
It has made no difference to our Win XP Pro machines, they just do not see it. It worked fine with Win 98 but does not
work any longer since we upgraded to Win XP Pro. Any hints?</quote>
</para>

<para>
Policy files are not portable between Windows 9x/Me and MS Windows NT4/200x/XP-based platforms. You need to
use the NT4 Group Policy Editor to create a file called <filename moreinfo="none">NTConfig.POL</filename> so it is in the
correct format for your MS Windows XP Pro clients.
</para>

</sect2>

</sect1>

</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="ProfileMgmt">
<chapterinfo>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
    <pubdate>April 3 2003</pubdate>
</chapterinfo>

<title>Desktop Profile Management</title>

<sect1>
<title>Features and Benefits</title>

<para>
<indexterm significance="normal"><primary>roaming profiles</primary></indexterm>
Roaming profiles are feared by some, hated by a few, loved by many, and a godsend for
some administrators.
</para>

<para>
<indexterm significance="normal"><primary>manage roaming profiles</primary></indexterm>
Roaming profiles allow an administrator to make available a consistent user desktop
as the user moves from one machine to another. This chapter provides much information
regarding how to configure and manage roaming profiles.
</para>

<para>
<indexterm significance="normal"><primary>local profiles</primary></indexterm>
While roaming profiles might sound like nirvana to some, they are a real and tangible
problem to others. In particular, users of mobile computing tools, where often there may not
be a sustained network connection, are often better served by purely local profiles.
This chapter provides information to help the Samba administrator deal with those
situations.
</para>

</sect1>

<sect1>
<title>Roaming Profiles</title>

<warning>
<para>
Roaming profiles support is different for Windows 9x/Me and Windows NT4/200x.
</para>
</warning>

<para>
Before discussing how to configure roaming profiles, it is useful to see how
Windows 9x/Me and Windows NT4/200x clients implement these features.
</para>

<para>
<indexterm significance="normal"><primary>NetUserGetInfo</primary></indexterm>
Windows 9x/Me clients send a NetUserGetInfo request to the server to get the user's
profiles location. However, the response does not have room for a separate
profiles location field, only the user's home share. This means that Windows 9x/Me
profiles are restricted to being stored in the user's home directory.
</para>


<para>
<indexterm significance="normal"><primary>NetSAMLogon</primary></indexterm>
<indexterm significance="normal"><primary>RPC</primary></indexterm>
Windows NT4/200x  clients send a NetSAMLogon RPC request, which contains many fields
including a separate field for the location of the user's profiles.
</para>

<sect2>
<title>Samba Configuration for Profile Handling</title>

<para>
This section documents how to configure Samba for MS Windows client profile support.
</para>

<sect3>
<title>NT4/200x User Profiles</title>

<para>
For example, to support Windows NT4/200x clients, set the following in the [global] section of the <filename moreinfo="none">smb.conf</filename> file:
</para>

<simplelist>
	<member><indexterm>logon path</indexterm><parameter>logon path =  \\profileserver\profileshare\profilepath\%U\moreprofilepath</parameter></member>
</simplelist>

<para>
This is typically implemented like:
<simplelist>
<member><indexterm>logon path</indexterm><parameter>logon path = \\%L\Profiles\%U</parameter></member>
</simplelist>
where <quote>%L</quote> translates to the name of the Samba server and <quote>%U</quote> translates to the username.
</para>

<para>
The default for this option is <filename moreinfo="none">\\%N\%U\profile</filename>, namely, <filename moreinfo="none">\\sambaserver\username\profile</filename>. 
The <filename moreinfo="none">\\%N\%U</filename> service is created automatically by the [homes] service. If you are using
a Samba server for the profiles, you must make the share that is specified in the logon path
browseable. Please refer to the man page for <filename moreinfo="none">smb.conf</filename> regarding the different
semantics of <quote>%L</quote> and <quote>%N</quote>, as well as <quote>%U</quote> and <quote>%u</quote>.
</para>

<note><para>
<indexterm significance="normal"><primary>logons</primary></indexterm>
<indexterm significance="normal"><primary>disconnect a connection</primary></indexterm>
MS Windows NT/200x clients at times do not disconnect a connection to a server between logons. It is recommended
to not use the <parameter>homes</parameter> metaservice name as part of the profile share path.
</para></note>
</sect3>

<sect3>
<title>Windows 9x/Me User Profiles</title>

<para>
<indexterm significance="normal"><primary>net use /home</primary></indexterm>
<indexterm significance="normal"><primary>logon home</primary></indexterm>
To support Windows 9x/Me clients, you must use the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOGONHOME" xlink:href="smb.conf.5.html#LOGONHOME">logon home</link>
parameter. Samba has been fixed so <userinput moreinfo="none">net use /home</userinput> now works as well and it, too, relies
on the <parameter moreinfo="none">logon home</parameter> parameter.
</para>

<para>
<indexterm significance="normal"><primary>logon home</primary></indexterm>
<indexterm significance="normal"><primary>\\%L\%U\.profiles</primary></indexterm>
<indexterm significance="normal"><primary>.profiles</primary></indexterm>
By using the <parameter moreinfo="none">logon home</parameter> parameter, you are restricted to putting Windows 9x/Me profiles
in the user's home directory.  But wait! There is a trick you can use. If you set the following in the
<parameter>[global]</parameter> section of your <filename moreinfo="none">smb.conf</filename> file:
<simplelist>
<member><indexterm>logon home</indexterm><parameter>logon home = \\%L\%U\.profiles</parameter></member>
</simplelist>
then your Windows 9x/Me clients will dutifully put their clients in a subdirectory
of your home directory called <filename moreinfo="none">.profiles</filename> (making them hidden).
</para>

<para>
<indexterm significance="normal"><primary>net use /home</primary></indexterm>
Not only that, but <userinput moreinfo="none">net use /home</userinput> will also work because of a feature in
Windows 9x/Me. It removes any directory stuff off the end of the home directory area
and only uses the server and share portion. That is, it looks like you
specified <filename moreinfo="none">\\%L\%U</filename> for <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOGONHOME" xlink:href="smb.conf.5.html#LOGONHOME">logon home</link>.
</para>
</sect3>

<sect3>
<title>Mixed Windows Windows 9x/Me and NT4/200x User Profiles</title>

<para>
You can support profiles for Windows 9x and Windows NT clients by setting both the
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOGONHOME" xlink:href="smb.conf.5.html#LOGONHOME">logon home</link> and <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOGONPATH" xlink:href="smb.conf.5.html#LOGONPATH">logon path</link> parameters. For example,
</para>

<para><simplelist>
<member><indexterm>logon home</indexterm><parameter>logon home = \\%L\%U\.profiles</parameter></member>
<member><indexterm>logon path</indexterm><parameter>logon path = \\%L\profiles\%U</parameter></member>
</simplelist></para>

<para>
<indexterm significance="normal"><primary>mixed profile</primary></indexterm>
Windows 9x/Me and NT4 and later profiles should not be stored in the same location because
Windows NT4 and later will experience problems with mixed profile environments.
</para>

</sect3>
<sect3>
<title>Disabling Roaming Profile Support</title>

<para>
<indexterm significance="normal"><primary>disable roaming profiles</primary></indexterm>
The question often asked is, <quote>How may I enforce use of local profiles?</quote> or
<quote>How do I disable roaming profiles?</quote>
</para>

<para>
<indexterm significance="normal"><primary>roaming profiles</primary></indexterm>
There are three ways of doing this:
</para>

<indexterm significance="normal"><primary>windows registry settings</primary><secondary>roaming profiles</secondary></indexterm>

<variablelist>
	<varlistentry>
		<term>In <filename moreinfo="none">smb.conf</filename></term>:
		<listitem><para>
		Affect the following settings and ALL clients will be forced to use a local profile:
		<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOGONHOME" xlink:href="smb.conf.5.html#LOGONHOME">logon home =  </link> and <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOGONPATH" xlink:href="smb.conf.5.html#LOGONPATH">logon path =  </link>
		</para>

		<para>
		The arguments to these parameters must be left blank. It is necessary to include the <constant>=</constant> sign
		to specifically assign the empty value.
		</para></listitem>
	</varlistentry>

	<varlistentry>
		<term>MS Windows Registry:</term>
		<listitem><para>
<indexterm significance="normal"><primary>MMC</primary></indexterm>
<indexterm significance="normal"><primary>local profile</primary></indexterm>
		Use the Microsoft Management Console (MMC) <literal>gpedit.msc</literal> to instruct your MS Windows XP
		machine to use only a local profile. This, of course, modifies registry settings. The full
		path to the option is:
<screen format="linespecific">
Local Computer Policy\
	Computer Configuration\
		Administrative Templates\
			System\
				User Profiles\

Disable: Only Allow Local User Profiles 
Disable: Prevent Roaming Profile Change from Propagating to the Server
</screen>
	</para></listitem>
	</varlistentry>

	<varlistentry>
		<term>Change of Profile Type:</term>
<indexterm significance="normal"><primary>Profile Type</primary></indexterm>
		<listitem><para>From the start menu right-click on the <guiicon moreinfo="none">My Computer</guiicon> icon,
		select <guimenuitem moreinfo="none">Properties</guimenuitem>, click on the <guilabel moreinfo="none">User Profiles</guilabel>
		tab, select the profile you wish to change from
		<guimenu moreinfo="none">Roaming</guimenu> type to <guimenu moreinfo="none">Local</guimenu>, and click on
		<guibutton moreinfo="none">Change Type</guibutton>.
		</para></listitem>
	</varlistentry>
</variablelist>

<para>
Consult the MS Windows registry guide for your particular MS Windows version for more information
about which registry keys to change to enforce use of only local user profiles.
</para>

<note><para>
<indexterm significance="normal"><primary>Windows Resource Kit</primary></indexterm>
The specifics of how to convert a local profile to a roaming profile, or a roaming profile
to a local one, vary according to the version of MS Windows you are running. Consult the Microsoft MS
Windows Resource Kit for your version of Windows for specific information.
</para></note>

</sect3>
</sect2>

<sect2>
<title>Windows Client Profile Configuration Information</title>

<sect3>
<title>Windows 9x/Me Profile Setup</title>

<para>
When a user first logs in on Windows 9x, the file user.DAT is created, as are folders <filename moreinfo="none">Start
Menu</filename>, <filename moreinfo="none">Desktop</filename>, <filename moreinfo="none">Programs</filename>, and
<filename moreinfo="none">Nethood</filename>. These directories and their contents will be merged with the local versions
stored in <filename moreinfo="none">c:\windows\profiles\username</filename> on subsequent logins, taking the most recent from
each.	You will need to use the <parameter>[global]</parameter> options <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRESERVECASE" xlink:href="smb.conf.5.html#PRESERVECASE">preserve case = yes</link>, <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SHORTPRESERVECASE" xlink:href="smb.conf.5.html#SHORTPRESERVECASE">short preserve case = yes</link>, and <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="CASESENSITIVE" xlink:href="smb.conf.5.html#CASESENSITIVE">case sensitive = no</link> in order to maintain capital letters in shortcuts in any of the
profile folders.
</para>

<para>
<indexterm significance="normal"><primary>user.DAT</primary></indexterm>
<indexterm significance="normal"><primary>user.MAN</primary></indexterm>
The <filename moreinfo="none">user.DAT</filename> file contains all the user's preferences. If you wish to enforce a set of preferences,
rename their <filename moreinfo="none">user.DAT</filename> file to <filename moreinfo="none">user.MAN</filename>, and deny them write access to this file.
</para>

<orderedlist inheritnum="ignore" continuation="restarts">
	<listitem> <para>
	On the Windows 9x/Me machine, go to <guimenu moreinfo="none">Control Panel</guimenu> -&gt;
	<guimenuitem moreinfo="none">Passwords</guimenuitem> and select the <guilabel moreinfo="none">User Profiles</guilabel> tab.
	Select the required level of roaming preferences. Press <guibutton moreinfo="none">OK</guibutton>, but do not
	allow the computer to reboot.
	</para> </listitem>

	<listitem> <para>
	On the Windows 9x/Me machine, go to <guimenu moreinfo="none">Control Panel</guimenu> -&gt;
	<guimenuitem moreinfo="none">Network</guimenuitem> -&gt; <guimenuitem moreinfo="none">Client for Microsoft Networks</guimenuitem>
	-&gt; <guilabel moreinfo="none">Preferences</guilabel>. Select <guilabel moreinfo="none">Log on to NT Domain</guilabel>.	Then,
	ensure that the Primary Logon is <guilabel moreinfo="none">Client for Microsoft Networks</guilabel>. Press
	<guibutton moreinfo="none">OK</guibutton>, and this time allow the computer to reboot.
	</para> </listitem>
</orderedlist>

<para>
<indexterm significance="normal"><primary>Primary Logon</primary></indexterm>
<indexterm significance="normal"><primary>Client for Novell Networks</primary></indexterm>
<indexterm significance="normal"><primary>Novell</primary></indexterm>
<indexterm significance="normal"><primary>Windows Logon</primary></indexterm>
Under Windows 9x/Me, profiles are downloaded from the Primary Logon. If you have the Primary Logon
as <quote>Client for Novell Networks</quote>, then the profiles and logon script will be downloaded from
your Novell server. If you have the Primary Logon as <quote>Windows Logon</quote>, then the profiles will
be loaded from the local machine <?latex --- ?> a bit against the concept of roaming profiles, it would seem! 
</para>

<para>
<indexterm significance="normal"><primary>domain logon server</primary></indexterm>
You will now find that the Microsoft Networks Login box contains <constant>[user, password, domain]</constant> instead
of just <constant>[user, password]</constant>. Type in the Samba server's domain name (or any other domain known to exist,
but bear in mind that the user will be authenticated against this domain and profiles downloaded from it
if that domain logon server supports it), user name and user's password. 
</para>

<para>
Once the user has been successfully validated, the Windows 9x/Me machine informs you that
<computeroutput moreinfo="none">The user has not logged on before</computeroutput> and asks <computeroutput moreinfo="none">Do you
wish to save the user's preferences?</computeroutput> Select <guibutton moreinfo="none">Yes</guibutton>.
</para>

<para>
Once the Windows 9x/Me client comes up with the desktop, you should be able to examine the
contents of the directory specified in the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOGONPATH" xlink:href="smb.conf.5.html#LOGONPATH">logon path</link> on
the Samba server and verify that the <filename moreinfo="none">Desktop</filename>, <filename moreinfo="none">Start Menu</filename>,
<filename moreinfo="none">Programs</filename>, and <filename moreinfo="none">Nethood</filename> folders have been created.
</para>

<para>
<indexterm significance="normal"><primary>cached locally</primary></indexterm>
<indexterm significance="normal"><primary>shortcuts</primary></indexterm>
<indexterm significance="normal"><primary>profile directory</primary></indexterm>
These folders will be cached locally on the client and updated when the user logs off (if
you haven't made them read-only by then). You will find that if the user creates further folders or
shortcuts, the client will merge the profile contents downloaded with the contents of the profile
directory already on the local client, taking the newest folders and shortcut from each set.
</para>

<para>
<indexterm significance="normal"><primary>local profile</primary></indexterm>
<indexterm significance="normal"><primary>remote profile</primary></indexterm>
<indexterm significance="normal"><primary>ownership rights</primary></indexterm>
<indexterm significance="normal"><primary>profile directory</primary></indexterm>
If you have made the folders/files read-only on the Samba server, then you will get errors from
the Windows 9x/Me machine on logon and logout as it attempts to merge the local and remote profile.
Basically, if you have any errors reported by the Windows 9x/Me machine, check the UNIX file permissions
and ownership rights on the profile directory contents, on the Samba server.
</para>

<para>
<indexterm significance="normal"><primary>windows registry settings</primary></indexterm>
<indexterm significance="normal"><primary>profile path</primary></indexterm>
<indexterm significance="normal"><primary>user profiles</primary></indexterm>
<indexterm significance="normal"><primary>desktop cache</primary></indexterm>
<indexterm significance="normal"><primary>windows registry settings</primary><secondary>profile path</secondary></indexterm>
If you have problems creating user profiles, you can reset the user's local desktop cache, as shown below.
When this user next logs in, the user will be told that he/she is logging in <quote>for the first
time</quote>.
</para>


<orderedlist inheritnum="ignore" continuation="restarts">
	<listitem><para>
	Instead of logging in under the [user, password, domain] dialog, press <guibutton moreinfo="none">escape</guibutton>.
	</para> </listitem>

	<listitem><para>
	Run the <literal>regedit.exe</literal> program, and look in:
	</para>

	<para>
	<filename moreinfo="none">HKEY_LOCAL_MACHINE\Windows\CurrentVersion\ProfileList</filename>
	</para>

	<para>
	You will find an entry for each user of ProfilePath. Note the contents of this key
	(likely to be <filename moreinfo="none">c:\windows\profiles\username</filename>), then delete the key
	<parameter moreinfo="none">ProfilePath</parameter> for the required user.
	</para></listitem>

	<listitem><para>
	Exit the registry editor.
	</para></listitem>

	<listitem><para>
	Search for the user's .PWL password-caching file in the <filename moreinfo="none">c:\windows</filename> directory, and delete it.
	</para></listitem>

	<listitem><para>
	Log off the Windows 9x/Me client.
	</para></listitem>

	<listitem><para>
	Check the contents of the profile path (see <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOGONPATH" xlink:href="smb.conf.5.html#LOGONPATH">logon path</link>
	described above) and delete the <filename moreinfo="none">user.DAT</filename> or <filename moreinfo="none">user.MAN</filename>
	file for the user, making a backup if required. 
	</para></listitem>
</orderedlist>

<warning><para>
<indexterm significance="normal"><primary>ProfilePath</primary></indexterm>
Before deleting the contents of the directory listed in the <parameter moreinfo="none">ProfilePath</parameter>
(this is likely to be <filename moreinfo="none">c:\windows\profiles\username)</filename>, ask whether the owner has
any important files stored on his or her desktop or start menu. Delete the contents of the
directory <parameter moreinfo="none">ProfilePath</parameter> (making a backup if any of the files are needed).
</para>

<para>
This will have the effect of removing the local (read-only hidden system file) <filename moreinfo="none">user.DAT</filename>
in their profile directory, as well as the local <quote>desktop,</quote> <quote>nethood,</quote>
<quote>start menu,</quote> and <quote>programs</quote> folders.
</para></warning>

<para>
<indexterm significance="normal"><primary>log level</primary></indexterm>
<indexterm significance="normal"><primary>packet sniffer</primary></indexterm>
<indexterm significance="normal"><primary>ethereal</primary></indexterm>
<indexterm significance="normal"><primary>netmon.exe</primary></indexterm>
If all else fails, increase Samba's debug log levels to between 3 and 10, and/or run a packet
sniffer program such as ethereal or <literal>netmon.exe</literal>, and look for error messages.
</para>

<para>
<indexterm significance="normal"><primary>roaming profiles</primary></indexterm>
<indexterm significance="normal"><primary>packet trace</primary></indexterm>
If you have access to an Windows NT4/200x server, then first set up roaming profiles and/or
netlogons on the Windows NT4/200x server. Make a packet trace, or examine the example packet traces
provided with Windows NT4/200x server, and see what the differences are with the equivalent Samba trace.
</para>

</sect3>

<sect3>
<title>Windows NT4 Workstation</title>

<para>
When a user first logs in to a Windows NT workstation, the profile NTuser.DAT is created. The profile
location can be now specified through the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOGONPATH" xlink:href="smb.conf.5.html#LOGONPATH">logon path</link> parameter.
</para>

<para>
There is a parameter that is now available for use with NT Profiles: <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOGONDRIVE" xlink:href="smb.conf.5.html#LOGONDRIVE">logon drive</link>. 
This should be set to <filename moreinfo="none">H:</filename> or any other drive, and should be used in conjunction with
the new <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOGONHOME" xlink:href="smb.conf.5.html#LOGONHOME">logon home</link> parameter.
</para>

<para>
<indexterm significance="normal"><primary>.PDS extension</primary></indexterm>
<indexterm significance="normal"><primary>profile path</primary></indexterm>
The entry for the NT4 profile is a directory, not a file. The NT help on profiles mentions that a
directory is also created with a .PDS extension. The user, while logging in, must have write permission
to create the full profile path (and the folder with the .PDS extension for those situations where it
might be created). 
</para>

<para>
<indexterm significance="normal"><primary>NTuser.DAT</primary></indexterm>
In the profile directory, Windows NT4 creates more folders than Windows 9x/Me. It creates
<filename moreinfo="none">Application Data</filename> and others, as well as <filename moreinfo="none">Desktop</filename>,
<filename moreinfo="none">Nethood</filename>, <filename moreinfo="none">Start Menu,</filename> and <filename moreinfo="none">Programs</filename>.
The profile itself is stored in a file <filename moreinfo="none">NTuser.DAT</filename>. Nothing appears to be stored
in the .PDS directory, and its purpose is currently unknown.
</para>

<para>
<indexterm significance="normal"><primary>NTuser.DAT</primary></indexterm>
<indexterm significance="normal"><primary>NTuser.MAN</primary></indexterm>
You can use the <application moreinfo="none">System Control Panel</application> to copy a local profile onto
a Samba server (see NT help on profiles; it is also capable of firing up the correct location in the
<application moreinfo="none">System Control Panel</application> for you). The NT help file also mentions that renaming
<filename moreinfo="none">NTuser.DAT</filename> to <filename moreinfo="none">NTuser.MAN</filename> turns a profile into a mandatory one.
</para>

<para>
The case of the profile is significant.	The file must be called <filename moreinfo="none">NTuser.DAT</filename>
or, for a mandatory profile, <filename moreinfo="none">NTuser.MAN</filename>.
</para>

</sect3>

<sect3>
<title>Windows 2000/XP Professional</title>

<para>
You must first convert the profile from a local profile to a domain profile on the MS Windows
workstation as follows: </para>

<procedure>
	<step performance="required"><para> Log on as the <emphasis>local</emphasis> workstation administrator. </para></step>

	<step performance="required"><para> Right-click on the <guiicon moreinfo="none">My Computer</guiicon> icon, and select
	<guimenuitem moreinfo="none">Properties</guimenuitem>.</para></step>

	<step performance="required"><para> Click on the <guilabel moreinfo="none">User Profiles</guilabel> tab.</para></step>

	<step performance="required"><para> Select the profile you wish to convert (click it once).</para></step>

	<step performance="required"><para> Click on the <guibutton moreinfo="none">Copy To</guibutton> button.</para></step>

	<step performance="required"><para> In the <guilabel moreinfo="none">Permitted to use</guilabel> box, click on the
	<guibutton moreinfo="none">Change</guibutton> button. </para></step>

	<step performance="required"><para> Click on the <guilabel moreinfo="none">Look in</guilabel> area that lists the machine name. When you click here, it will
	open up a selection box. Click on the domain to which the profile must be accessible. </para>

	<note><para>You will need to log on if a logon box opens up. 
	For example, connect as <replaceable>DOMAIN</replaceable>\root, password:
	<replaceable>mypassword</replaceable>.</para></note> </step>

	<step performance="required"><para> To make the profile capable of being used by anyone, select <quote>Everyone</quote>. </para></step>

	<step performance="required"><para> Click on <guibutton moreinfo="none">OK</guibutton> and the Selection box will close. </para></step>

	<step performance="required"><para> Now click on <guibutton moreinfo="none">OK</guibutton> to create the profile in the path
	you nominated.	</para></step>
</procedure>

<para>
Done. You now have a profile that can be edited using the Samba <literal>profiles</literal> tool.
</para>

<note><para>
Under Windows NT/200x, the use of mandatory profiles forces the use of MS Exchange storage of mail
data and keeps it out of the desktop profile. That keeps desktop profiles from becoming unusable.
</para></note>

<sect4>
<title>Windows XP Service Pack 1</title>
	<para>
	There is a security check new to Windows XP (or maybe only Windows XP service pack 1).
	It can be disabled via a group policy in the Active Directory. The policy is called:
<screen format="linespecific">
Computer Configuration\Administrative Templates\System\User Profiles\
          Do not check for user ownership of Roaming Profile Folders
</screen>
	</para>

	<para>
	This should be set to <constant>Enabled</constant>.
	</para>

	<para>
	Does the new version of Samba have an Active Directory analogue?  If so, then you may be able to set the policy through this.
	</para>

	<para>If you cannot set group policies in Samba, then you may be able to set the policy locally on
	each machine. If you want to try this, then do the following:
	</para>


<procedure>
	<step performance="required"><para>On the XP workstation, log in with an administrative account.</para></step>

	<step performance="required"><para>Click on <guimenu moreinfo="none">Start</guimenu> -&gt; <guimenuitem moreinfo="none">Run</guimenuitem>.</para></step>
	<step performance="required"><para>Type <literal>mmc</literal>.</para></step>
	<step performance="required"><para>Click on <guibutton moreinfo="none">OK</guibutton>.</para></step>
	<step performance="required"><para>A Microsoft Management Console should appear.</para></step>
	<step performance="required"><para>Click on <guimenu moreinfo="none">File</guimenu> -&gt; <guimenuitem moreinfo="none">Add/Remove Snap-in</guimenuitem> -&gt; <guimenuitem moreinfo="none">Add</guimenuitem>.</para></step> 
	<step performance="required"><para>Double-click on <guiicon moreinfo="none">Group Policy</guiicon>.</para></step> 
	<step performance="required"><para>Click on <guibutton moreinfo="none">Finish</guibutton> -&gt; <guibutton moreinfo="none">Close</guibutton>.</para></step> 
	<step performance="required"><para>Click on <guibutton moreinfo="none">OK</guibutton>.</para></step>
	<step performance="required"><para>In the <quote>Console Root</quote> window expand <guiicon moreinfo="none">Local Computer Policy</guiicon> -&gt;
		<guiicon moreinfo="none">Computer Configuration</guiicon> -&gt; <guiicon moreinfo="none">Administrative Templates</guiicon> -&gt; 
		<guiicon moreinfo="none">System</guiicon> -&gt; <guiicon moreinfo="none">User Profiles</guiicon>.</para></step>
	<step performance="required"><para>Double-click on <guilabel moreinfo="none">Do not check for user ownership of Roaming Profile Folders</guilabel>.</para></step>
	<step performance="required"><para>Select <guilabel moreinfo="none">Enabled</guilabel>.</para></step>
	<step performance="required"><para>Click on <guibutton moreinfo="none">OK</guibutton>.</para></step>
	<step performance="required"><para>Close the whole console. You do not need to save the settings (this refers to the
	console settings rather than the policies you have changed).</para></step>
	<step performance="required"><para>Reboot.</para></step>
</procedure>
</sect4>
</sect3>
</sect2>

<sect2>
<title>User Profile Hive Cleanup Service</title>

<para>
There are certain situations that cause a cached local copy of roaming profile not to be deleted on exit, even if
the policy to force such deletion is set. To deal with that situation, a special service was created. The application 
<literal>UPHClean</literal> (User Profile Hive Cleanup) can be installed as a service on Windows NT4/2000/XP Professional
and Windows 2003.
</para>

<para>
The UPHClean software package can be downloaded from the User Profile Hive Cleanup
Service<footnote>http://www.microsoft.com/downloads/details.aspx?FamilyID=1B286E6D-8912-4E18-B570-42470E2F3582&amp;displaylang=en</footnote>
web site.
</para>

</sect2>

<sect2>
<title>Sharing Profiles between Windows 9x/Me and NT4/200x/XP Workstations</title>

<para>
<indexterm significance="normal"><primary>profile sharing</primary></indexterm>
<indexterm significance="normal"><primary>profile contents</primary></indexterm>
Sharing of desktop profiles between Windows versions is not recommended. Desktop profiles are an
evolving phenomenon, and profiles for later versions of MS Windows clients add features that may interfere
with earlier versions of MS Windows clients. Probably the more salient reason to not mix profiles is
that when logging off an earlier version of MS Windows, the older format of profile contents may overwrite
information that belongs to the newer version, resulting in loss of profile information content when that
user logs on again with the newer version of MS Windows.
</para>

<para>
If you then want to share the same Start Menu and Desktop with Windows 9x/Me, you must specify a common
location for the profiles. The <filename moreinfo="none">smb.conf</filename> parameters that need to be common are 
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOGONPATH" xlink:href="smb.conf.5.html#LOGONPATH">logon path</link> and <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOGONHOME" xlink:href="smb.conf.5.html#LOGONHOME">logon home</link>.
</para>

<para>
<indexterm significance="normal"><primary>user.DAT</primary></indexterm>
<indexterm significance="normal"><primary>NTuser.DAT</primary></indexterm>
If you have this set up correctly, you will find separate <filename moreinfo="none">user.DAT</filename> and
<filename moreinfo="none">NTuser.DAT</filename> files in the same profile directory.
</para>

</sect2>

<sect2>
<title>Profile Migration from Windows NT4/200x Server to Samba</title>

<para>
<indexterm significance="normal"><primary>encrypted passwords</primary></indexterm>
There is nothing to stop you from specifying any path that you like for the location of users' profiles.
Therefore, you could specify that the profile be stored on a Samba server or any other SMB server,
as long as that SMB server supports encrypted passwords.
</para>

<sect3 id="profilemigrn">
<title>Windows NT4 Profile Management Tools</title>

<para>
<indexterm significance="normal"><primary>resource kit</primary></indexterm>
Unfortunately, the resource kit information is specific to the version of MS Windows NT4/200x. The
correct resource kit is required for each platform.
</para>

<para>Here is a quick guide:</para>

<procedure>
<title>Profile Migration Procedure</title>

	<step performance="required"><para> On your NT4 domain controller, right-click on <guiicon moreinfo="none">My Computer</guiicon>, then select 
	<guilabel moreinfo="none">Properties</guilabel>, then the tab labeled <guilabel moreinfo="none">User Profiles</guilabel>. </para></step>

	<step performance="required"><para> Select a user profile you want to migrate and click on it. </para>

	<note><para>I am using the term <quote>migrate</quote> loosely. You can copy a profile to create a group
	profile. You can give the user <parameter moreinfo="none">Everyone</parameter> rights to the profile you copy this to. That
	is what you need to do, since your Samba domain is not a member of a trust relationship with your NT4
	PDC.</para></note></step>

	<step performance="required"><para>Click on the <guibutton moreinfo="none">Copy To</guibutton> button.</para></step>

	<step performance="required"><para>In the box labeled <guilabel moreinfo="none">Copy Profile to</guilabel> add your new path, such as,
	<filename moreinfo="none">c:\temp\foobar</filename></para></step>

	<step performance="required"><para>Click on <guibutton moreinfo="none">Change</guibutton> in the <guilabel moreinfo="none">Permitted to use</guilabel> box.</para></step>

	<step performance="required"><para>Click on the group <quote>Everyone</quote>, click on <guibutton moreinfo="none">OK</guibutton>. This
	closes the <quote>choose user</quote> box.</para></step>

	<step performance="required"><para>Now click on <guibutton moreinfo="none">OK</guibutton>.</para></step>
</procedure>

<para>
Follow these steps for every profile you need to migrate.
</para>

</sect3>

<sect3>
<title>Side Bar Notes</title>


<para>
<indexterm significance="normal"><primary>SID</primary></indexterm>
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>info</tertiary></indexterm>
You should obtain the SID of your NT4 domain. You can use the <literal>net rpc info</literal> to do this.
See <link linkend="NetCommand">The Net Command Chapter</link>, <link linkend="netmisc1">Other Miscellaneous Operations</link> for more information.
</para>

</sect3>

<sect3>
<title>moveuser.exe</title>

<para>
<indexterm significance="normal"><primary>moveuser.exe</primary></indexterm>
The Windows 200x professional resource kit has <literal>moveuser.exe</literal>.
<literal>moveuser.exe</literal> changes the security of a profile from one user to another. This allows the
account domain to change and/or the username to change.
</para>

<para>
This command is like the Samba <literal>profiles</literal> tool.
</para>

</sect3>

<sect3>
<title>Get SID</title>

<para>
<indexterm significance="normal"><primary>SID</primary></indexterm>
<indexterm significance="normal"><primary>GetSID.exe</primary></indexterm>
You can identify the SID by using <literal>GetSID.exe</literal> from the Windows NT Server 4.0 Resource Kit.
</para>

<para>
Windows NT 4.0 stores the local profile information in the registry under the following key:
<filename moreinfo="none">HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows NT\CurrentVersion\ProfileList</filename>
</para>

<para>
Under the ProfileList key, there will be subkeys named with the SIDs of the users who have logged
on to this computer. (To find the profile information for the user whose locally cached profile you want
to move, find the SID for the user with the <literal>GetSID.exe</literal> utility.) Inside the appropriate user's subkey,
you will see a string value named <parameter moreinfo="none">ProfileImagePath</parameter>.
</para>

</sect3>
</sect2>
</sect1>

<sect1>
<title>Mandatory Profiles</title>

<para>
<indexterm significance="normal"><primary>mandatory profiles</primary></indexterm>
A mandatory profile is a profile that the user does not have the ability to overwrite. During the
user's session, it may be possible to change the desktop environment; however, as the user logs out, all changes
made will be lost. If it is desired to not allow the user any ability to change the desktop environment,
then this must be done through policy settings. See <link linkend="PolicyMgmt">System and Account
Policies</link>.
</para>

<note><para> 
<indexterm significance="normal"><primary>fake-permissions module</primary></indexterm>
<indexterm significance="normal"><primary>VFS module</primary></indexterm>
<indexterm significance="normal"><primary>fake_perms</primary></indexterm>
Under NO circumstances should the profile directory (or its contents) be made read-only because this may
render the profile unusable.  Where it is essential to make a profile read-only within the UNIX file system,
this can be done, but then you absolutely must use the <literal>fake-permissions</literal> VFS module to
instruct MS Windows NT/200x/XP clients that the Profile has write permission for the user.  See <link linkend="fakeperms">fake_perms VFS module</link>.
</para></note>

<para>
<indexterm significance="normal"><primary>NTUser.MAN</primary></indexterm>
<indexterm significance="normal"><primary>NTUser.DAT</primary></indexterm>
For MS Windows NT4/200x/XP, the procedure shown in <link linkend="profilemigrn">Profile Migration from Windows
NT4/200x Server to Samba</link> can also be used to create mandatory profiles. To convert a group profile into
a mandatory profile, simply locate the <filename moreinfo="none">NTUser.DAT</filename> file in the copied profile and rename
it to <filename moreinfo="none">NTUser.MAN</filename>.
</para>

<para>
<indexterm significance="normal"><primary>User.MAN</primary></indexterm>
For MS Windows 9x/Me, it is the <filename moreinfo="none">User.DAT</filename> file that must be renamed to
<filename moreinfo="none">User.MAN</filename> to effect a mandatory profile.
</para>

</sect1>

<sect1>
<title>Creating and Managing Group Profiles</title>

<para>
<indexterm significance="normal"><primary>group profiles</primary></indexterm>
<indexterm significance="normal"><primary>template</primary></indexterm>
<indexterm significance="normal"><primary>profile migration tool</primary></indexterm>
<indexterm significance="normal"><primary>profile access rights</primary></indexterm>
Most organizations are arranged into departments. There is a nice benefit in this fact, since usually
most users in a department require the same desktop applications and the same desktop layout. MS
Windows NT4/200x/XP will allow the use of group profiles. A group profile is a profile that is created
first using a template (example) user. Then using the profile migration tool (see above), the profile is
assigned access rights for the user group that needs to be given access to the group profile.
</para>

<para>
<indexterm significance="normal"><primary>User Manager</primary></indexterm>
The next step is rather important. Instead of assigning a group profile to users (Using User Manager)
on a <quote>per-user</quote> basis, the group itself is assigned the now modified profile.
</para>

<note><para>
Be careful with group profiles. If the user who is a member of a group also has a personal
profile, then the result will be a fusion (merge) of the two.
</para></note>

</sect1>

<sect1>
<title>Default Profile for Windows Users</title>

<para>
<indexterm significance="normal"><primary>default profile</primary></indexterm>
<indexterm significance="normal"><primary>registry keys</primary></indexterm>
MS Windows 9x/Me and NT4/200x/XP will use a default profile for any user for whom a profile
does not already exist. Armed with a knowledge of where the default profile is located on the Windows
workstation, and knowing which registry keys affect the path from which the default profile is created,
it is possible to modify the default profile to one that has been optimized for the site. This has
significant administrative advantages.
</para>

<sect2>
<title>MS Windows 9x/Me</title>

<para>
<indexterm significance="normal"><primary>System Policy Editor</primary></indexterm>
<indexterm significance="normal"><primary>registry</primary></indexterm>
To enable default per-use profiles in Windows 9x/Me, you can either use the <application moreinfo="none">Windows
98 System Policy Editor</application> or change the registry directly.
</para>

<para>
To enable default per-user profiles in Windows 9x/Me, launch the <application moreinfo="none">System Policy
Editor</application>, then select <guimenu moreinfo="none">File</guimenu> -&gt; <guimenuitem moreinfo="none">Open Registry</guimenuitem>.
Next click on the <guiicon moreinfo="none">Local Computer</guiicon> icon, click on <guilabel moreinfo="none">Windows 98 System</guilabel>,
select <guilabel moreinfo="none">User Profiles</guilabel>, and click on the enable box. Remember to save the registry
changes.
</para>

<para>
<indexterm significance="normal"><primary>regedit.exe</primary></indexterm>
To modify the registry directly, launch the <application moreinfo="none">Registry Editor</application>
(<literal>regedit.exe</literal>) and select the hive <filename moreinfo="none">HKEY_LOCAL_MACHINE\Network\Logon</filename>.
Now add a DWORD type key with the name <quote>User Profiles.</quote> To enable user profiles to set the value
to 1; to disable user profiles set it to 0.
</para>

<sect3>
<title>User Profile Handling with Windows 9x/Me</title>

<para>
When a user logs on to a Windows 9x/Me machine, the local profile path,
<filename moreinfo="none">HKEY_LOCAL_MACHINE\Software\Microsoft\Windows\CurrentVersion\ProfileList</filename>, is checked
for an existing entry for that user.
</para>

<para>
If the user has an entry in this registry location, Windows 9x/Me checks for a locally cached
version of the user profile. Windows 9x/Me also checks the user's home directory (or other specified
directory if the location has been modified) on the server for the user profile. If a profile exists
in both locations, the newer of the two is used. If the user profile exists on the server but does not
exist on the local machine, the profile on the server is downloaded and used. If the user profile only
exists on the local machine, that copy is used.
</para>

<para>
If a user profile is not found in either location, the default user profile from the Windows
9x/Me machine is used and copied to a newly created folder for the logged on user. At log off, any
changes that the user made are written to the user's local profile. If the user has a roaming profile,
the changes are written to the user's profile on the server.
</para>

</sect3>
</sect2>

<sect2>
<title>MS Windows NT4 Workstation</title>

<para>
On MS Windows NT4, the default user profile is obtained from the location
<filename moreinfo="none">%SystemRoot%\Profiles</filename>, which in a default installation will translate to
<filename moreinfo="none">C:\Windows NT\Profiles</filename>. Under this directory on a clean install, there will be three
directories: <filename moreinfo="none">Administrator</filename>, <filename moreinfo="none">All
Users,</filename> and <filename moreinfo="none">Default
User</filename>.
</para>

<para>
The <filename moreinfo="none">All Users</filename> directory contains menu settings that are common across all
system users. The <filename moreinfo="none">Default User</filename> directory contains menu entries that are customizable
per user depending on the profile settings chosen/created.
</para>

<para>
When a new user first logs onto an MS Windows NT4 machine, a new profile is created from:
</para>

<itemizedlist>
	<listitem><para>All Users settings.</para></listitem>
	<listitem><para>Default User settings (contains the default <filename moreinfo="none">NTUser.DAT</filename> file).</para></listitem>
</itemizedlist>

<para>
<indexterm significance="normal"><primary>NTConfig.POL</primary></indexterm>
When a user logs on to an MS Windows NT4 machine that is a member of a Microsoft security domain,
the following steps are followed for profile handling: 
</para>

<procedure>
	<step performance="required"> <para> The user's account information that is obtained during the logon process
	contains the location of the user's desktop profile. The profile path may be local to
	the machine or it may be located on a network share. If there exists a profile at the
	location of the path from the user account, then this profile is copied to the location
	<filename moreinfo="none">%SystemRoot%\Profiles\%USERNAME%</filename>. This profile then inherits the settings
	in the <filename moreinfo="none">All Users</filename> profile in the <filename moreinfo="none">%SystemRoot%\Profiles</filename>
	location. </para> </step>

	<step performance="required"> <para> If the user account has a profile path, but at its location a profile does not
	exist, then a new profile is created in the <filename moreinfo="none">%SystemRoot%\Profiles\%USERNAME%</filename>
	directory from reading the <filename moreinfo="none">Default User</filename> profile. </para> </step>

	<step performance="required"> <para>
<indexterm significance="normal"><primary>NTConfig.POL</primary></indexterm>
<indexterm significance="normal"><primary>NETLOGON</primary></indexterm>
<indexterm significance="normal"><primary>authenticating server</primary></indexterm>
<indexterm significance="normal"><primary>logon server</primary></indexterm>
<indexterm significance="normal"><primary>HKEY_CURRENT_USER</primary></indexterm>
	If the NETLOGON share on the authenticating server (logon server) contains
	a policy file (<filename moreinfo="none">NTConfig.POL</filename>), then its contents are applied to the
	<filename moreinfo="none">NTUser.DAT</filename>, which is applied to the <filename moreinfo="none">HKEY_CURRENT_USER</filename>
	part of the registry. 
	</para> </step>

	<step performance="required"> <para> When the user logs out, if the profile is set to be a roaming profile, it will be
	written out to the location of the profile. The <filename moreinfo="none">NTuser.DAT</filename> file is then
	re-created from the contents of the <filename moreinfo="none">HKEY_CURRENT_USER</filename> contents. Thus,
	should there not exist in the NETLOGON share an <filename moreinfo="none">NTConfig.POL</filename> at the next
	logon, the effect of the previous <filename moreinfo="none">NTConfig.POL</filename> will still be held in the
	profile. The effect of this is known as tattooing.
	</para> </step>
</procedure>

<para>
MS Windows NT4 profiles may be <emphasis>local</emphasis> or <emphasis>roaming</emphasis>. A local
profile is stored in the <filename moreinfo="none">%SystemRoot%\Profiles\%USERNAME%</filename> location. A roaming
profile will also remain stored in the same way, unless the following registry key is created:
<screen format="linespecific">
HKEY_LOCAL_MACHINE\SYSTEM\Software\Microsoft\Windows NT\CurrentVersion\
winlogon\"DeleteRoamingCache"=dword:0000000
</screen>
In this case, the local copy (in <filename moreinfo="none">%SystemRoot%\Profiles\%USERNAME%</filename>) will be deleted
on logout.
</para>

<para>
<indexterm significance="normal"><primary>regedt32</primary></indexterm>
Under MS Windows NT4, default locations for common resources like <filename moreinfo="none">My Documents</filename>
may be redirected to a network share by modifying the following registry keys. These changes may be
made via use of the System Policy Editor. To do so may require that you create your own template
extension for the Policy Editor to allow this to be done through the GUI. Another way to do this is by
first creating a default user profile, then while logged in as that user, running <literal>regedt32</literal> to edit
the key settings.
</para>

<para>
The Registry Hive key that affects the behavior of folders that are part of the default user
profile are controlled by entries on Windows NT4 is:
<screen format="linespecific">
HKEY_CURRENT_USER
	\Software
		\Microsoft
			\Windows
				\CurrentVersion
					\Explorer
						\User Shell Folders
</screen>
<indexterm significance="normal"><primary>windows registry settings</primary><secondary>default profile locations</secondary></indexterm>
</para>

<para>  The above hive key contains a list of automatically managed
folders. The default entries are shown in <link linkend="ProfileLocs">the next table</link>.
</para>

<table frame="all" id="ProfileLocs">
	<title>User Shell Folder Registry Keys Default Values</title>
	<tgroup cols="2">
		<colspec align="left"/>
		<colspec align="left"/>
	<thead>
		<row><entry>Name</entry><entry>Default Value</entry></row>
	</thead>
	<tbody>
		<row><entry>AppData</entry><entry>%USERPROFILE%\Application Data</entry></row>
		<row><entry>Desktop</entry><entry>%USERPROFILE%\Desktop</entry></row>
		<row><entry>Favorites</entry><entry>%USERPROFILE%\Favorites</entry></row>
		<row><entry>NetHood</entry><entry>%USERPROFILE%\NetHood</entry></row>
		<row><entry>PrintHood</entry><entry>%USERPROFILE%\PrintHood</entry></row>
		<row><entry>Programs</entry><entry>%USERPROFILE%\Start Menu\Programs</entry></row>
		<row><entry>Recent</entry><entry>%USERPROFILE%\Recent</entry></row>
		<row><entry>SendTo</entry><entry>%USERPROFILE%\SendTo</entry></row>
		<row><entry>Start Menu </entry><entry>%USERPROFILE%\Start Menu</entry></row>
		<row><entry>Startup</entry><entry>%USERPROFILE%\Start Menu\Programs\Startup</entry></row>
	</tbody>
	</tgroup>
</table>

<para> The registry key that contains the location of the default profile settings is:
<screen format="linespecific">
HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\CurrentVersion\Explorer\
User Shell Folders
</screen>
</para>

<para>
The default entries are shown in <link linkend="regkeys">Defaults of Profile Settings Registry Keys</link>.
</para>

<table frame="all" id="regkeys">
	<title>Defaults of Profile Settings Registry Keys</title>
	<tgroup cols="2">
		<colspec align="left"/>
		<colspec align="left"/>
	<tbody>
		<row><entry>Common Desktop</entry><entry>%SystemRoot%\Profiles\All Users\Desktop</entry></row>
		<row><entry>Common Programs</entry><entry>%SystemRoot%\Profiles\All Users\Programs</entry></row>
		<row><entry>Common Start Menu</entry><entry>%SystemRoot%\Profiles\All Users\Start Menu</entry></row>
		<row><entry>Common Startup</entry><entry>%SystemRoot%\Profiles\All Users\Start Menu\Programs\Startup</entry></row>
	</tbody>
	</tgroup>
</table>

</sect2>

<sect2>
<title>MS Windows 200x/XP</title>

<note><para>
<indexterm significance="normal"><primary>GPOs</primary></indexterm>
<indexterm significance="normal"><primary>Windows XP Home Edition</primary></indexterm>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
<indexterm significance="normal"><primary>domain security</primary></indexterm>
MS Windows XP Home Edition does use default per-user profiles, but cannot participate
in domain security, cannot log onto an NT/ADS-style domain, and thus can obtain the profile only
from itself. While there are benefits in doing this, the beauty of those MS Windows clients that
can participate in domain logon processes is that they allow the administrator to create a global default
profile and enforce it through the use of Group Policy Objects (GPOs).
</para></note>

<para>
<indexterm significance="normal"><primary>Default User</primary></indexterm>
When a new user first logs onto an MS Windows 200x/XP machine, the default profile is obtained from
<filename moreinfo="none">C:\Documents and Settings\Default User</filename>. The administrator can modify or change the
contents of this location, and MS Windows 200x/XP will gladly use it. This is far from the optimum arrangement,
since it will involve copying a new default profile to every MS Windows 200x/XP client workstation.
</para>

<para>
<indexterm significance="normal"><primary>NETLOGON</primary></indexterm>
When MS Windows 200x/XP participates in a domain security context, and if the default user profile is not
found, then the client will search for a default profile in the NETLOGON share of the authenticating server.
In MS Windows parlance, it is <filename moreinfo="none">%LOGONSERVER%\NETLOGON\Default User,</filename>
and if one exists there, it will copy this to the workstation in the <filename moreinfo="none">C:\Documents and
Settings\</filename> under the Windows login name of the use.
</para>

<note> <para> This path translates, in Samba parlance, to the <filename moreinfo="none">smb.conf</filename>
<parameter>[NETLOGON]</parameter> share. The directory should be created at the root
of this share and must be called <filename moreinfo="none">Default User</filename>.
</para> </note>

<para> If a default profile does not exist in this location, then MS Windows 200x/XP will use the local
default profile. </para>

<para> On logging out, the user's desktop profile is stored to the location specified in the registry
settings that pertain to the user. If no specific policies have been created or passed to the client
during the login process (as Samba does automatically), then the user's profile is written to the
local machine only under the path <filename moreinfo="none">C:\Documents and Settings\%USERNAME%</filename>. </para>

<para> Those wishing to modify the default behavior can do so through these three methods: </para>

<itemizedlist>
	<listitem> <para> Modify the registry keys on the local machine manually and place the new
	default profile in the NETLOGON share root. This is not recommended because it is maintenance intensive.
	</para> </listitem>

	<listitem> <para> Create an NT4-style NTConfig.POL file that specifies this behavior and locate
	this file in the root of the NETLOGON share along with the new default profile. </para> </listitem>

	<listitem> <para> Create a GPO that enforces this through Active Directory, and place the new
	default profile in the NETLOGON share.	</para> </listitem>
</itemizedlist>

<para>The registry hive key that affects the behavior of folders that are part of the default user
profile are controlled by entries on Windows 200x/XP is: </para>

<para> <filename moreinfo="none">HKEY_CURRENT_USER\Software\Microsoft\Windows\CurrentVersion\Explorer\User Shell
Folders\</filename> </para>

<para>
This hive key contains a list of automatically managed folders. The default entries are shown
in <link linkend="defregpthkeys">the next table</link>
<indexterm significance="normal"><primary>windows registry settings</primary><secondary>default profile locations</secondary></indexterm>
</para>


<table frame="all" id="defregpthkeys">
	<title>Defaults of Default User Profile Paths Registry Keys</title>
	<tgroup cols="2">
		<colspec align="left"/>
		<colspec align="left"/>
	<thead>
		<row><entry>Name</entry><entry>Default Value</entry></row>
	</thead>
	<tbody>
		<row><entry>AppData</entry><entry>%USERPROFILE%\Application Data</entry></row>
		<row><entry>Cache</entry><entry>%USERPROFILE%\Local Settings\Temporary Internet Files</entry></row>
		<row><entry>Cookies</entry><entry>%USERPROFILE%\Cookies</entry></row>
		<row><entry>Desktop</entry><entry>%USERPROFILE%\Desktop</entry></row>
		<row><entry>Favorites</entry><entry>%USERPROFILE%\Favorites</entry></row>
		<row><entry>History</entry><entry>%USERPROFILE%\Local Settings\History</entry></row>
		<row><entry>Local AppData</entry><entry>%USERPROFILE%\Local Settings\Application Data</entry></row>
		<row><entry>Local Settings</entry><entry>%USERPROFILE%\Local Settings</entry></row>
		<row><entry>My Pictures</entry><entry>%USERPROFILE%\My Documents\My Pictures</entry></row>
		<row><entry>NetHood</entry><entry>%USERPROFILE%\NetHood</entry></row>
		<row><entry>Personal</entry><entry>%USERPROFILE%\My Documents</entry></row>
		<row><entry>PrintHood</entry><entry>%USERPROFILE%\PrintHood</entry></row>
		<row><entry>Programs</entry><entry>%USERPROFILE%\Start Menu\Programs</entry></row>
		<row><entry>Recent</entry><entry>%USERPROFILE%\Recent</entry></row>
		<row><entry>SendTo</entry><entry>%USERPROFILE%\SendTo</entry></row>
		<row><entry>Start Menu</entry><entry>%USERPROFILE%\Start Menu</entry></row>
		<row><entry>Startup</entry><entry>%USERPROFILE%\Start Menu\Programs\Startup</entry></row>
		<row><entry>Templates</entry><entry>%USERPROFILE%\Templates</entry></row>
	</tbody>
	</tgroup>
</table>

<para> There is also an entry called <quote>Default</quote> that has no value set. The default entry is
of type <constant>REG_SZ</constant>; all the others are of type <constant>REG_EXPAND_SZ</constant>. </para>

<para> It makes a huge difference to the speed of handling roaming user profiles if all the folders are
stored on a dedicated location on a network server. This means that it will not be necessary to write
the Outlook PST file over the network for every login and logout. </para>

<para>
To set this to a network location, you could use the following examples:
<screen format="linespecific">
%LOGONSERVER%\%USERNAME%\Default Folders
</screen>
This stores the folders in the user's home directory under a directory called <filename moreinfo="none">Default
Folders</filename>. You could also use:
<screen format="linespecific">
\\<replaceable>SambaServer</replaceable>\<replaceable>FolderShare</replaceable>\%USERNAME%
</screen>
</para>

<para>
in which case the default folders are stored in the server named <replaceable>SambaServer</replaceable>
in the share called <replaceable>FolderShare</replaceable> under a directory that has the name of the
MS Windows user as seen by the Linux/UNIX file system.	</para>

<para> Please note that once you have created a default profile share, you <emphasis>must</emphasis> migrate a user's profile
(default or custom) to it. </para>

<para> MS Windows 200x/XP profiles may be <emphasis>local</emphasis> or <emphasis>roaming</emphasis>.
	A roaming profile is cached locally unless the following registry key is created: 

<indexterm significance="normal"><primary>delete roaming profiles</primary></indexterm>
</para>


<para> <programlisting format="linespecific"> HKEY_LOCAL_MACHINE\SYSTEM\Software\Microsoft\Windows NT\CurrentVersion\
	winlogon\"DeleteRoamingCache"=dword:00000001</programlisting></para>

<para>
In this case, the local cache copy is deleted on logout.
</para> 
</sect2> 
</sect1>

<sect1> <title>Common Errors</title>

<para>
The following are some typical errors, problems, and questions that have been asked on the Samba mailing lists.
</para>

<sect2>
<title>Configuring Roaming Profiles for a Few Users or Groups</title>

<para>
With Samba-2.2.x, the choice you have is to enable or disable roaming profiles support. It is a
global-only setting. The default is to have roaming profiles, and the default path will locate them in
the user's home directory.
</para>

<para>
If disabled globally, then no one will have roaming profile ability. If enabled and you want it
to apply only to certain machines, then on those machines on which roaming profile support is not wanted,
it is necessary to disable roaming profile handling in the registry of each such machine.
</para>

<para>
With Samba-3, you can have a global profile setting in <filename moreinfo="none">smb.conf</filename>, and you can override this by
per-user settings using the Domain User Manager (as with MS Windows NT4/200x). </para>

<para> In any case, you can configure only one profile per user. That profile can be either: </para>

<itemizedlist>
	<listitem><para>A profile unique to that user.</para></listitem>
	<listitem><para>A mandatory profile (one the user cannot change).</para></listitem>
	<listitem><para>A group profile (really should be mandatory <?latex --- ?> that is, unchangable).</para></listitem>
</itemizedlist>

</sect2>

<sect2> <title>Cannot Use Roaming Profiles</title>

<para> A user requested the following: <quote> I do not want roaming profiles to be implemented. I want
to give users a local profile alone. I am totally lost with this error. For the past
two days I tried everything, I googled around but found no useful pointers. Please help me. </quote></para>

<para> The choices are: </para>

<variablelist>
	<varlistentry>
		<term>Local profiles</term> <listitem><para> I know of no registry keys that will allow
		autodeletion of LOCAL profiles on log out.</para></listitem>
	</varlistentry>

	<varlistentry>
		<term>Roaming profiles</term> <listitem><para> As a user logs onto the network, a centrally
		stored profile is copied to the workstation to form a local profile. This local profile
		will persist (remain on the workstation disk) unless a registry key is changed that will
		cause this profile to be automatically deleted on logout. </para></listitem>
	</varlistentry>
</variablelist>

<para>The roaming profile choices are: </para>

<variablelist>
	<varlistentry>
		<term>Personal roaming profiles</term> <listitem><para> These are typically stored in
		a profile share on a central (or conveniently located local) server. </para>

		<para> Workstations cache (store) a local copy of the profile. This cached
		copy is used when the profile cannot be downloaded at next logon. </para></listitem>
	</varlistentry>

	<varlistentry>
		<term>Group profiles</term> <listitem><para>These are loaded from a central profile
		server.</para></listitem>
	</varlistentry>

	<varlistentry>
		<term>Mandatory profiles</term> <listitem><para> Mandatory profiles can be created for
		a user as well as for any group that a user is a member of. Mandatory profiles cannot be
		changed by ordinary users. Only the administrator can change or reconfigure a mandatory
		profile. </para></listitem>
	</varlistentry>
</variablelist>

<para> A Windows NT4/200x/XP profile can vary in size from 130KB to very large. Outlook PST files are
most often part of the profile and can be many gigabytes in size. On average (in a well controlled environment),
roaming profile size of 2MB is a good rule of thumb to use for planning purposes. In an undisciplined
environment, I have seen up to 2GB profiles. Users tend to complain when it takes an hour to log onto a
workstation, but they harvest the fruits of folly (and ignorance). </para>

<para> The point of this discussion is to show that roaming profiles and good controls of how they can be
changed as well as good discipline make for a problem-free site. </para>

<para> Microsoft's answer to the PST problem is to store all email in an MS Exchange Server backend. This
removes the need for a PST file. </para>

<para>Local profiles mean: </para>

<itemizedlist>
	<listitem><para>If each machine is used by many users, then much local disk storage is needed
	for local profiles.</para></listitem> <listitem><para>Every workstation the user logs into has
	its own profile; these can be very different from machine to machine.</para></listitem>
</itemizedlist>

<para> On the other hand, use of roaming profiles means: </para>

<itemizedlist>
	<listitem><para>The network administrator can control the desktop environment of all users.</para></listitem>
	<listitem><para>Use of mandatory profiles drastically reduces network management overheads.</para></listitem>
	<listitem><para>In the long run, users will experience fewer problems.</para></listitem>
</itemizedlist>

</sect2>

<sect2>
<title>Changing the Default Profile</title>

<para><quote>When the client logs onto the domain controller, it searches
for a profile to download. Where do I put this default profile?</quote></para>

<para>
<indexterm significance="normal"><primary>default profile</primary></indexterm>
First, the Samba server needs to be configured as a domain controller. This can be done by
setting in <filename moreinfo="none">smb.conf</filename>: </para>

<simplelist>
<member><indexterm>security</indexterm><parameter>security = user</parameter></member>
<member><indexterm>os level</indexterm><parameter>os level = 32 (or more)</parameter></member>
<member><indexterm>domain logons</indexterm><parameter>domain logons = Yes</parameter></member>
</simplelist>

<para> There must be a <parameter>[netlogon]</parameter> share that is world readable. It is
a good idea to add a logon script to preset printer and drive connections. There is also a facility
for automatically synchronizing the workstation time clock with that of the logon server (another good
thing to do). </para>

<note><para> To invoke autodeletion of roaming profiles from the local workstation cache (disk storage), use
the <application moreinfo="none">Group Policy Editor</application> to create a file called <filename moreinfo="none">NTConfig.POL</filename>
with the appropriate entries. This file needs to be located in the <parameter>netlogon</parameter>
share root directory.</para></note>

<para> Windows clients need to be members of the domain. Workgroup machines do not use network logons,
so they do not interoperate with domain profiles. </para>

<para> For roaming profiles, add to <filename moreinfo="none">smb.conf</filename>: </para>

<simplelist>
<member><indexterm>logon path</indexterm><parameter>logon path = \\%N\profiles\%U</parameter></member>
<member># Default logon drive is Z:</member>
<member><indexterm>logon drive</indexterm><parameter>logon drive = H:</parameter></member>
<member># This requires a PROFILES share that is world writable.</member>
</simplelist>

</sect2>

<sect2>
<title>Debugging Roaming Profiles and NT4-style Domain Policies</title>

<para>
Roaming profiles and domain policies are implemented via <literal>USERENV.DLL</literal>. 
Microsoft Knowledge Base articles <ulink url="http://support.microsoft.com/default.aspx?scid=kb;en-us;221833">221833</ulink> and 
<ulink url="http://support.microsoft.com/default.aspx?scid=kb;en-us;154120">154120</ulink>
 describe how to instruct that DLL to debug the login process.
</para>

</sect2>
</sect1>
</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="pam">
<chapterinfo>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
	<author>
		<firstname>Stephen</firstname><surname>Langasek</surname>
		<affiliation>
			<address format="linespecific"><email>vorlon@netexpress.net</email></address>
		</affiliation>
	</author>
    <pubdate>May 31, 2003</pubdate>
</chapterinfo>
 
<title>PAM-Based Distributed Authentication</title>

<para>
<indexterm significance="normal"><primary>PAM-enabled</primary></indexterm>
<indexterm significance="normal"><primary>Winbind</primary></indexterm>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
<indexterm significance="normal"><primary>Winbind-based authentication</primary></indexterm>
This chapter should help you to deploy Winbind-based authentication on any PAM-enabled
UNIX/Linux system. Winbind can be used to enable user-level application access authentication
from any MS Windows NT domain, MS Windows 200x Active Directory-based
domain, or any Samba-based domain environment. It will also help you to configure PAM-based local host access
controls that are appropriate to your Samba configuration.
</para>

<para>
<indexterm significance="normal"><primary>PAM management</primary></indexterm>
<indexterm significance="normal"><primary>pam_smbpass.so</primary></indexterm>
In addition to knowing how to configure Winbind into PAM, you will learn generic PAM management
possibilities and in particular how to deploy tools like <filename moreinfo="none">pam_smbpass.so</filename> to your advantage.
</para>

<note><para>
The use of Winbind requires more than PAM configuration alone.
Please refer to <link linkend="winbind">Winbind: Use of Domain Accounts</link>, for further information regarding Winbind.
</para></note>

<sect1>
<title>Features and Benefits</title>

<para>
<indexterm significance="normal"><primary>Sun Solaris</primary></indexterm>
<indexterm significance="normal"><primary>xxxxBSD</primary></indexterm>
<indexterm significance="normal"><primary>Linux</primary></indexterm>
<indexterm significance="normal"><primary>Pluggable Authentication Modules</primary><see>PAM</see></indexterm>
<indexterm significance="normal"><primary>/etc/passwd</primary></indexterm>
<indexterm significance="normal"><primary>login</primary></indexterm>
<indexterm significance="normal"><primary>passwd</primary></indexterm>
<indexterm significance="normal"><primary>chown</primary></indexterm>
A number of UNIX systems (e.g., Sun Solaris), as well as the xxxxBSD family and Linux,
now utilize the Pluggable Authentication Modules (PAM) facility to provide all authentication, 
authorization, and resource control services. Prior to the introduction of PAM, a decision
to use an alternative to the system password database (<filename moreinfo="none">/etc/passwd</filename>) 
would require the provision of alternatives for all programs that provide security services.
Such a choice would involve provision of alternatives to programs such as <literal>login</literal>, 
<literal>passwd</literal>, <literal>chown</literal>, and so on.
</para>

<para>
<indexterm significance="normal"><primary>PAM</primary></indexterm>
<indexterm significance="normal"><primary>/etc/pam.conf</primary></indexterm>
<indexterm significance="normal"><primary>Solaris</primary></indexterm>
<indexterm significance="normal"><primary>/etc/pam.d</primary></indexterm>
PAM provides a mechanism that disconnects these security programs from the underlying
authentication/authorization infrastructure. PAM is configured by making appropriate modifications to one file,
<filename moreinfo="none">/etc/pam.conf</filename> (Solaris), or by editing individual control files that are
located in <filename moreinfo="none">/etc/pam.d</filename>.
</para>

<para>
<indexterm significance="normal"><primary>PAM-enabled</primary></indexterm>
<indexterm significance="normal"><primary>dynamically loadable library modules</primary></indexterm>
On PAM-enabled UNIX/Linux systems, it is an easy matter to configure the system to use any
authentication backend so long as the appropriate dynamically loadable library modules
are available for it. The backend may be local to the system or may be centralized on a
remote server.
</para>

<para>
PAM support modules are available for:
</para>

<variablelist>
	<varlistentry><term><filename moreinfo="none">/etc/passwd</filename></term><listitem>
		<para>
<indexterm significance="normal"><primary>/etc/passwd</primary></indexterm>
<indexterm significance="normal"><primary>PAM modules</primary></indexterm>
<indexterm significance="normal"><primary>pam_unix.so</primary></indexterm>
<indexterm significance="normal"><primary>pam_unix2.so</primary></indexterm>
<indexterm significance="normal"><primary>pam_pwdb.so</primary></indexterm>
<indexterm significance="normal"><primary>pam_userdb.so</primary></indexterm>
		There are several PAM modules that interact with this standard UNIX user database. The most common are called
		<filename moreinfo="none">pam_unix.so</filename>, <filename moreinfo="none">pam_unix2.so</filename>, <filename moreinfo="none">pam_pwdb.so</filename> and
		<filename moreinfo="none">pam_userdb.so</filename>.
		</para>
	</listitem></varlistentry>

	<varlistentry><term>Kerberos</term><listitem>
		<para>
<indexterm significance="normal"><primary>pam_krb5.so</primary></indexterm>
<indexterm significance="normal"><primary>Kerberos</primary></indexterm>
<indexterm significance="normal"><primary>Heimdal</primary></indexterm>
<indexterm significance="normal"><primary>MIT Kerberos</primary></indexterm>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
		The <filename moreinfo="none">pam_krb5.so</filename> module allows the use of any Kerberos-compliant server.
		This tool is used to access MIT Kerberos, Heimdal Kerberos, and potentially
		Microsoft Active Directory (if enabled).
		</para>
	</listitem></varlistentry>

	<varlistentry><term>LDAP</term><listitem>
		<para>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
<indexterm significance="normal"><primary>pam_ldap.so</primary></indexterm>
<indexterm significance="normal"><primary>OpenLDAP</primary></indexterm>
<indexterm significance="normal"><primary>Sun ONE iDentity server</primary></indexterm>
<indexterm significance="normal"><primary>Novell eDirectory server</primary></indexterm>
<indexterm significance="normal"><primary>Microsoft Active Directory</primary></indexterm>
		The <filename moreinfo="none">pam_ldap.so</filename> module allows the use of any LDAP v2- or v3-compatible backend
		server. Commonly used LDAP backend servers include OpenLDAP v2.0 and v2.1,
		Sun ONE iDentity server, Novell eDirectory server, and Microsoft Active Directory.
		</para>
	</listitem></varlistentry>

	<varlistentry><term>NetWare Bindery</term><listitem>
		<para>
<indexterm significance="normal"><primary>NetWare Bindery</primary></indexterm>
<indexterm significance="normal"><primary>pam_ncp_auth.so</primary></indexterm>
<indexterm significance="normal"><primary>bindery-enabled</primary></indexterm>
<indexterm significance="normal"><primary>NetWare Core Protocol-based server</primary></indexterm>
		The <filename moreinfo="none">pam_ncp_auth.so</filename> module allows authentication off any bindery-enabled
		NetWare Core Protocol-based server.
		</para>
	</listitem></varlistentry>

	<varlistentry><term>SMB Password</term><listitem>
		<para>
<indexterm significance="normal"><primary>SMB Password</primary></indexterm>
<indexterm significance="normal"><primary>pam_smbpass.so</primary></indexterm>
<indexterm significance="normal"><primary>passdb backend</primary></indexterm>
		This module, called <filename moreinfo="none">pam_smbpass.so</filename>, allows user authentication of
		the passdb backend that is configured in the Samba <filename moreinfo="none">smb.conf</filename> file.
		</para>
	</listitem></varlistentry>

	<varlistentry><term>SMB Server</term><listitem>
		<para>
<indexterm significance="normal"><primary>SMB Server</primary></indexterm>
<indexterm significance="normal"><primary>pam_smb_auth.so</primary></indexterm>
		The <filename moreinfo="none">pam_smb_auth.so</filename> module is the original MS Windows networking authentication
		tool. This module has been somewhat outdated by the Winbind module.
		</para>
	</listitem></varlistentry>

	<varlistentry><term>Winbind</term><listitem>
		<para>
<indexterm significance="normal"><primary>Winbind</primary></indexterm>
<indexterm significance="normal"><primary>pam_winbind.so</primary></indexterm>
<indexterm significance="normal"><primary>domain controller</primary></indexterm>
<indexterm significance="normal"><primary>authentication</primary></indexterm>
		The <filename moreinfo="none">pam_winbind.so</filename> module allows Samba to obtain authentication from any
		MS Windows domain controller. It can just as easily be used to authenticate
		users for access to any PAM-enabled application.
		</para>
	</listitem></varlistentry>

	<varlistentry><term>RADIUS</term><listitem>
		<para>
<indexterm significance="normal"><primary>Remote Access Dial-In User Service</primary><see>RADIUS</see></indexterm>
		There is a PAM RADIUS (Remote Access Dial-In User Service) authentication
		module. In most cases, administrators need to locate the source code
		for this tool and compile and install it themselves. RADIUS protocols are
		used by many routers and terminal servers.
		</para>
	</listitem></varlistentry>
</variablelist>

<para>
<indexterm significance="normal"><primary>pam_smbpasswd.so</primary></indexterm>
<indexterm significance="normal"><primary>pam_winbind.so</primary></indexterm>
Of the modules listed, Samba provides the <filename moreinfo="none">pam_smbpasswd.so</filename> and the
<filename moreinfo="none">pam_winbind.so</filename> modules alone.
</para>

<para>
<indexterm significance="normal"><primary>wide-area network bandwidth</primary></indexterm>
<indexterm significance="normal"><primary>efficient authentication</primary></indexterm>
<indexterm significance="normal"><primary>PAM-capable</primary></indexterm>
<indexterm significance="normal"><primary>centrally managed</primary></indexterm>
Once configured, these permit a remarkable level of flexibility in the location and use
of distributed Samba domain controllers that can provide wide-area network bandwidth,
efficient authentication services for PAM-capable systems. In effect, this allows the
deployment of centrally managed and maintained distributed authentication from a
single-user account database.
</para>

</sect1>

<sect1>
<title>Technical Discussion</title>

<para>
<indexterm significance="normal"><primary>PAM</primary></indexterm>
<indexterm significance="normal"><primary>privilege-granting applications</primary></indexterm>
<indexterm significance="normal"><primary>/etc/pam.conf</primary></indexterm>
<indexterm significance="normal"><primary>/etc/pam.d/</primary></indexterm>
PAM is designed to provide system administrators with a great deal of flexibility in
configuration of the privilege-granting applications of their system. The local
configuration of system security controlled by PAM is contained in one of two places:
either the single system file <filename moreinfo="none">/etc/pam.conf</filename> or the
<filename moreinfo="none">/etc/pam.d/</filename> directory.
</para>

<sect2>
<title>PAM Configuration Syntax</title>

<para>
<indexterm significance="normal"><primary>PAM-specific tokens</primary></indexterm>
<indexterm significance="normal"><primary>case sensitivity</primary></indexterm>
In this section we discuss the correct syntax of and generic options respected by entries to these files.
PAM-specific tokens in the configuration file are case insensitive. The module paths, however, are case
sensitive, since they indicate a file's name and reflect the case dependence of typical file systems.  The
case sensitivity of the arguments to any given module is defined for each module in turn.
</para>

<para>
In addition to the lines described below, there are two special characters provided for the convenience
of the system administrator: comments are preceded by a <quote>#</quote> and extend to the next end-of-line; also,
module specification lines may be extended with a <quote>\</quote>-escaped newline. 
</para>

<para>
<indexterm significance="normal"><primary>PAM authentication module</primary></indexterm>
<indexterm significance="normal"><primary>/lib/security</primary></indexterm>
If the PAM authentication module (loadable link library file) is located in the
default location, then it is not necessary to specify the path. In the case of
Linux, the default location is <filename moreinfo="none">/lib/security</filename>. If the module
is located outside the default, then the path must be specified as:
<programlisting format="linespecific">
auth  required  /other_path/pam_strange_module.so
</programlisting>
</para>

<sect3>
<title>Anatomy of <filename moreinfo="none">/etc/pam.d</filename> Entries</title>

<para>
The remaining information in this subsection was taken from the documentation of the Linux-PAM
project. For more information on PAM, see 
<ulink url="http://ftp.kernel.org/pub/linux/libs/pam/">the Official Linux-PAM home page</ulink>.
</para>

<para>
<indexterm significance="normal"><primary>/etc/pam.conf</primary></indexterm>
A general configuration line of the <filename moreinfo="none">/etc/pam.conf</filename> file has the following form:
<programlisting format="linespecific">
service-name   module-type   control-flag   module-path   args
</programlisting>
</para>

<para>
We explain the meaning of each of these tokens. The second (and more recently adopted)
way of configuring Linux-PAM is via the contents of the <filename moreinfo="none">/etc/pam.d/</filename> directory.
Once we have explained the meaning of the tokens, we describe this method.
</para>

<variablelist>
	<varlistentry><term>service-name</term><listitem>
		<para>
<indexterm significance="normal"><primary>ftpd</primary></indexterm>
<indexterm significance="normal"><primary>rlogind</primary></indexterm>
<indexterm significance="normal"><primary>su</primary></indexterm>
		The name of the service associated with this entry. Frequently, the service-name is the conventional
		name of the given application <?latex --- ?> for example, <literal>ftpd</literal>, <literal>rlogind</literal> and
		<literal>su</literal>, and so on.
		</para>

		<para>
		There is a special service-name reserved for defining a default authentication mechanism. It has
		the name <parameter moreinfo="none">OTHER</parameter> and may be specified in either lower- or uppercase characters.
		Note, when there is a module specified for a named service, the <parameter moreinfo="none">OTHER</parameter>
		entries are ignored.
		</para>
		</listitem>
	</varlistentry>

	<varlistentry><term>module-type</term><listitem>
                <para>
		One of (currently) four types of module. The four types are as follows:
		</para>

		<itemizedlist>
			<listitem><para>
<indexterm significance="normal"><primary>auth</primary></indexterm>
<indexterm significance="normal"><primary>/etc/groups</primary></indexterm>
			<parameter moreinfo="none">auth:</parameter> This module type provides two aspects of authenticating the user.
			It establishes that the user is who he or she claims to be by instructing the application
			to prompt the user for a password or other means of identification. Second, the module can
			grant group membership (independently of the <filename moreinfo="none">/etc/groups</filename> file)
			or other privileges through its credential-granting properties.
			</para></listitem>

			<listitem><para>
<indexterm significance="normal"><primary>account</primary></indexterm>
<indexterm significance="normal"><primary>non-authentication-based account management</primary></indexterm>
			<parameter moreinfo="none">account:</parameter> This module performs non-authentication-based account management.
			It is typically used to restrict/permit access to a service based on the time of day, currently
		 	available system resources (maximum number of users), or perhaps the location of the user 
			login. For example, the <quote>root</quote> login may be permitted only on the console.
			</para></listitem>

			<listitem><para>
<indexterm significance="normal"><primary>session</primary></indexterm>
			<parameter moreinfo="none">session:</parameter> Primarily, this module is associated with doing things that need
			to be done for the user before and after he or she can be given service. Such things include logging
			information concerning the opening and closing of some data exchange with a user, mounting
			directories, and so on.
			</para></listitem>

			<listitem><para>
<indexterm significance="normal"><primary>password</primary></indexterm>
			<parameter moreinfo="none">password:</parameter> This last module type is required for updating the authentication
			token associated with the user. Typically, there is one module for each
			<quote>challenge/response</quote> authentication <parameter moreinfo="none">(auth)</parameter> module type.
			</para></listitem>
		</itemizedlist>
		</listitem>
	</varlistentry>

	<varlistentry><term>control-flag</term><listitem>
		<para>
		The control-flag is used to indicate how the PAM library will react to the success or failure of the
		module it is associated with. Since modules can be stacked (modules of the same type execute in series,
		one after another), the control-flags determine the relative importance of each module. The application
		is not made aware of the individual success or failure of modules listed in the
		<filename moreinfo="none">/etc/pam.conf</filename> file. Instead, it receives a summary success or fail response from
		the Linux-PAM library. The order of execution of these modules is that of the entries in the
		<filename moreinfo="none">/etc/pam.conf</filename> file; earlier entries are executed before later ones.
		As of Linux-PAM v0.60, this control-flag can be defined with one of two syntaxes.
		</para>

		<para>
<indexterm significance="normal"><primary>required</primary></indexterm>
<indexterm significance="normal"><primary>requisite</primary></indexterm>
<indexterm significance="normal"><primary>sufficient</primary></indexterm>
<indexterm significance="normal"><primary>optional</primary></indexterm>
		The simpler (and historical) syntax for the control-flag is a single keyword defined to indicate the
		severity of concern associated with the success or failure of a specific module. There are four such
		keywords: <parameter moreinfo="none">required</parameter>, <parameter moreinfo="none">requisite</parameter>,
		<parameter moreinfo="none">sufficient</parameter>, and <parameter moreinfo="none">optional</parameter>.
		</para>

		<para>
		The Linux-PAM library interprets these keywords in the following manner:
		</para>

		<itemizedlist>
			<listitem><para>
			<parameter moreinfo="none">required:</parameter> This indicates that the success of the module is required for the
			module-type facility to succeed. Failure of this module will not be apparent to the user until all
			of the remaining modules (of the same module-type) have been executed.
			</para></listitem>

			<listitem><para>
			<parameter moreinfo="none">requisite:</parameter> Like required, except that if such a module returns a
			failure, control is directly returned to the application. The return value is that associated with
			the first required or requisite module to fail. This flag can be used to protect against the
			possibility of a user getting the opportunity to enter a password over an unsafe medium. It is
			conceivable that such behavior might inform an attacker of valid accounts on a system. This
			possibility should be weighed against the not insignificant concerns of exposing a sensitive
			password in a hostile environment.
			</para></listitem>

			<listitem><para>
			<parameter moreinfo="none">sufficient:</parameter> The success of this module is deemed <parameter moreinfo="none">sufficient</parameter> to satisfy
			the Linux-PAM library that this module-type has succeeded in its purpose. In the event that no
			previous required module has failed, no more <quote>stacked</quote> modules of this type are invoked.
			(In this case, subsequent required modules are not invoked). A failure of this module is not deemed
			as fatal to satisfying the application that this module-type has succeeded.
			</para></listitem>

            <listitem><para>
            <parameter moreinfo="none">optional:</parameter> As its name suggests, this control-flag marks the module as not
			being critical to the success or failure of the user's application for service. In general,
			Linux-PAM ignores such a module when determining if the module stack will succeed or fail.
			However, in the absence of any definite successes or failures of previous or subsequent stacked
			modules, this module will determine the nature of the response to the application. One example of
			this latter case is when the other modules return something like PAM_IGNORE.
			</para></listitem>
		</itemizedlist>

		<para>
		The more elaborate (newer) syntax is much more specific and gives the administrator a great deal of control
		over how the user is authenticated. This form of the control-flag is delimited with square brackets and
		consists of a series of <parameter moreinfo="none">value=action</parameter> tokens:
		</para>

<para><programlisting format="linespecific">
[value1=action1 value2=action2 ...]
</programlisting></para>

		<para>
		Here, <parameter moreinfo="none">value1</parameter> is one of the following return values:
<screen format="linespecific">
<parameter moreinfo="none">success; open_err; symbol_err; service_err; system_err; buf_err;</parameter>
<parameter moreinfo="none">perm_denied; auth_err; cred_insufficient; authinfo_unavail;</parameter>
<parameter moreinfo="none">user_unknown; maxtries; new_authtok_reqd; acct_expired; session_err;</parameter>
<parameter moreinfo="none">cred_unavail; cred_expired; cred_err; no_module_data; conv_err;</parameter>
<parameter moreinfo="none">authtok_err; authtok_recover_err; authtok_lock_busy;</parameter>
<parameter moreinfo="none">authtok_disable_aging; try_again; ignore; abort; authtok_expired;</parameter>
<parameter moreinfo="none">module_unknown; bad_item;</parameter> and <parameter moreinfo="none">default</parameter>.
</screen>
</para>

		<para>
		The last of these (<parameter moreinfo="none">default</parameter>) can be used to set the action for those return values that are not explicitly defined.
		</para>

		<para>
		The <parameter moreinfo="none">action1</parameter> can be a positive integer or one of the following tokens: 
		<parameter moreinfo="none">ignore</parameter>; <parameter moreinfo="none">ok</parameter>; <parameter moreinfo="none">done</parameter>;
		<parameter moreinfo="none">bad</parameter>; <parameter moreinfo="none">die</parameter>; and <parameter moreinfo="none">reset</parameter>.
		A positive integer, J, when specified as the action, can be used to indicate that the next J modules of the
		current module-type will be skipped. In this way, the administrator can develop a moderately sophisticated
		stack of modules with a number of different paths of execution. Which path is taken can be determined by the
		reactions of individual modules.
		</para>

		<itemizedlist>
			<listitem><para>
			<parameter moreinfo="none">ignore:</parameter> When used with a stack of modules, the module's return status will not
			contribute to the return code the application obtains.
			</para></listitem>

			<listitem><para>
            <parameter moreinfo="none">bad:</parameter> This action indicates that the return code should be thought of as indicative
			of the module failing. If this module is the first in the stack to fail, its status value will be used
			for that of the whole stack.
			</para></listitem>

            <listitem><para>
            <parameter moreinfo="none">die:</parameter> Equivalent to bad with the side effect of terminating the module stack and
			PAM immediately returning to the application.
			</para></listitem>

           <listitem><para>
           <parameter moreinfo="none">ok:</parameter> This tells PAM that the administrator thinks this return code should
			contribute directly to the return code of the full stack of modules. In other words, if the former
			state of the stack would lead to a return of PAM_SUCCESS, the module's return code will override
			this value. Note, if the former state of the stack holds some value that is indicative of a module's
			failure, this <parameter moreinfo="none">ok</parameter> value will not be used to override that value.
			</para></listitem>

            <listitem><para>
            <parameter moreinfo="none">done:</parameter> Equivalent to <parameter moreinfo="none">ok</parameter> with the side effect of terminating the module stack and
			PAM immediately returning to the application.
                        </para></listitem>

           <listitem><para>
           <parameter moreinfo="none">reset:</parameter> Clears all memory of the state of the module stack and starts again with
			the next stacked module.
			</para></listitem>
		</itemizedlist>

		<para>
		Each of the four keywords, <parameter moreinfo="none">required</parameter>; <parameter moreinfo="none">requisite</parameter>;
		<parameter moreinfo="none">sufficient</parameter>; and <parameter moreinfo="none">optional</parameter>, have an equivalent expression in terms
		of the [...] syntax. They are as follows:
		</para>

		<para>
		<itemizedlist>
			<listitem><para>
			<parameter moreinfo="none">required</parameter> is equivalent to <parameter moreinfo="none">[success=ok new_authtok_reqd=ok ignore=ignore default=bad]</parameter>.
			</para></listitem>

			<listitem><para>
			<parameter moreinfo="none">requisite</parameter> is equivalent to <parameter moreinfo="none">[success=ok new_authtok_reqd=ok ignore=ignore default=die]</parameter>.
			</para></listitem>

			<listitem><para>
			<parameter moreinfo="none">sufficient</parameter> is equivalent to <parameter moreinfo="none">[success=done  new_authtok_reqd=done  default=ignore]</parameter>.
			</para></listitem>

			<listitem><para>
			<parameter moreinfo="none">optional</parameter> is equivalent to <parameter moreinfo="none">[success=ok new_authtok_reqd=ok default=ignore]</parameter>.
			</para></listitem>
		</itemizedlist>
		</para>

		<para>
		Just to get a feel for the power of this new syntax, here is a taste of what you can do with it. With Linux-PAM-0.63,
		the notion of client plug-in agents was introduced. This makes it possible for PAM to support
		machine-machine authentication using the transport protocol inherent to the client/server application. With the
		<parameter moreinfo="none">[ ... value=action ... ]</parameter> control syntax, it is possible for an application to be configured
		to support binary prompts with compliant clients, but to gracefully fail over into an alternative authentication
		mode for legacy applications.
		</para>
		</listitem>
	</varlistentry>

	<varlistentry><term>module-path</term><listitem>
		<para>
		The pathname of the dynamically loadable object file; the pluggable module itself. If the first character of the
		module path is <quote>/</quote>, it is assumed to be a complete path. If this is not the case, the given module path is appended
		to the default module path: <filename moreinfo="none">/lib/security</filename> (but see the previous notes).
		</para>

		<para>
		The arguments are a list of tokens that are passed to the module when it is invoked, much like arguments to a typical
		Linux shell command. Generally, valid arguments are optional and are specific to any given module. Invalid arguments
		are ignored by a module; however, when encountering an invalid argument, the module is required to write an error
		to syslog(3). For a list of generic options, see the next section.
		</para>

		<para>
		If you wish to include spaces in an argument, you should surround that argument with square brackets. For example:
		</para>

<para><programlisting format="linespecific">
squid auth required pam_mysql.so user=passwd_query passwd=mada \
db=eminence [query=select user_name from internet_service where \
user_name=<quote>%u</quote> and password=PASSWORD(<quote>%p</quote>) and service=<quote>web_proxy</quote>]
</programlisting></para>

		<para>
		When using this convention, you can include <quote>[</quote> characters inside the string, and if you wish to have a <quote>]</quote>
		character inside the string that will survive the argument parsing, you should use <quote>\[</quote>. In other words,
		</para>

<para><programlisting format="linespecific">
[..[..\]..]    --&gt;   ..[..]..
</programlisting></para>

		<para>
		Any line in one of the configuration files that is not formatted correctly will generally tend (erring on the
		side of caution) to make the authentication process fail. A corresponding error is written to the system log files
		with a call to syslog(3). 
		</para>
		</listitem>
	</varlistentry>
</variablelist>

</sect3>

</sect2>

<sect2>
<title>Example System Configurations</title>

<para>
The following is an example <filename moreinfo="none">/etc/pam.d/login</filename> configuration file. 
This example had all options uncommented and is probably not usable 
because it stacks many conditions before allowing successful completion 
of the login process. Essentially, all conditions can be disabled 
by commenting them out, except the calls to <filename moreinfo="none">pam_pwdb.so</filename>.
</para>

<sect3>
<title>PAM: Original Login Config</title>

<para>
	<programlisting format="linespecific">
#%PAM-1.0
# The PAM configuration file for the <quote>login</quote> service
#
auth         required    pam_securetty.so
auth         required    pam_nologin.so
# auth       required    pam_dialup.so
# auth       optional    pam_mail.so
auth         required    pam_pwdb.so shadow md5
# account    requisite   pam_time.so
account      required    pam_pwdb.so
session      required    pam_pwdb.so
# session    optional    pam_lastlog.so
# password   required    pam_cracklib.so retry=3
password     required    pam_pwdb.so shadow md5
</programlisting>
</para>

</sect3>

<sect3>
<title>PAM: Login Using <filename moreinfo="none">pam_smbpass</filename></title>

<para>
PAM allows use of replaceable modules. Those available on a sample system include:
</para>

<para><prompt moreinfo="none">$</prompt><userinput moreinfo="none">/bin/ls /lib/security</userinput>
<programlisting format="linespecific">
pam_access.so    pam_ftp.so          pam_limits.so     
pam_ncp_auth.so  pam_rhosts_auth.so  pam_stress.so     
pam_cracklib.so  pam_group.so        pam_listfile.so   
pam_nologin.so   pam_rootok.so       pam_tally.so      
pam_deny.so      pam_issue.so        pam_mail.so       
pam_permit.so    pam_securetty.so    pam_time.so       
pam_dialup.so    pam_lastlog.so      pam_mkhomedir.so  
pam_pwdb.so      pam_shells.so       pam_unix.so       
pam_env.so       pam_ldap.so         pam_motd.so       
pam_radius.so    pam_smbpass.so      pam_unix_acct.so  
pam_wheel.so     pam_unix_auth.so    pam_unix_passwd.so
pam_userdb.so    pam_warn.so         pam_unix_session.so
</programlisting></para>

<para>
The following example for the login program replaces the use of 
the <filename moreinfo="none">pam_pwdb.so</filename> module that uses the system 
password database (<filename moreinfo="none">/etc/passwd</filename>,
<filename moreinfo="none">/etc/shadow</filename>, <filename moreinfo="none">/etc/group</filename>) with 
the module <filename moreinfo="none">pam_smbpass.so</filename>, which uses the Samba 
database containing the Microsoft MD4 encrypted password 
hashes. This database is stored either in 
<filename moreinfo="none">/usr/local/samba/private/smbpasswd</filename>, 
<filename moreinfo="none">/etc/samba/smbpasswd</filename> or in 
<filename moreinfo="none">/etc/samba.d/smbpasswd</filename>, depending on the 
Samba implementation for your UNIX/Linux system. The 
<filename moreinfo="none">pam_smbpass.so</filename> module is provided by 
Samba version 2.2.1 or later. It can be compiled by specifying the 
<option>--with-pam_smbpass</option> options when running Samba's
<literal>configure</literal> script. For more information
on the <filename moreinfo="none">pam_smbpass</filename> module, see the documentation
in the <filename moreinfo="none">source/pam_smbpass</filename> directory of the Samba 
source distribution.
</para>

<para>
	<programlisting format="linespecific">
#%PAM-1.0
# The PAM configuration file for the <quote>login</quote> service
#
auth        required    pam_smbpass.so nodelay
account     required    pam_smbpass.so nodelay
session     required    pam_smbpass.so nodelay
password    required    pam_smbpass.so nodelay
</programlisting></para>

<para>
The following is the PAM configuration file for a particular 
Linux system. The default condition uses <filename moreinfo="none">pam_pwdb.so</filename>.
</para>

<para>
	<programlisting format="linespecific">
#%PAM-1.0
# The PAM configuration file for the <quote>samba</quote> service
#
auth       required     pam_pwdb.so nullok nodelay shadow audit
account    required     pam_pwdb.so audit nodelay
session    required     pam_pwdb.so nodelay
password   required     pam_pwdb.so shadow md5
</programlisting></para>

<para>
In the following example, the decision has been made to use the 
<literal>smbpasswd</literal> database even for basic Samba authentication. Such a 
decision could also be made for the <literal>passwd</literal> program and would 
thus allow the <literal>smbpasswd</literal> passwords to be changed using the
<literal>passwd</literal> program:
</para>

<para>
		<programlisting format="linespecific">
#%PAM-1.0
# The PAM configuration file for the <quote>samba</quote> service
#
auth       required     pam_smbpass.so nodelay
account    required     pam_pwdb.so audit nodelay
session    required     pam_pwdb.so nodelay
password   required     pam_smbpass.so nodelay smbconf=/etc/samba.d/smb.conf
</programlisting>
</para>

<note><para>PAM allows stacking of authentication mechanisms. It is 
also possible to pass information obtained within one PAM module through 
to the next module in the PAM stack. Please refer to the documentation for 
your particular system implementation for details regarding the specific 
capabilities of PAM in this environment. Some Linux implementations also 
provide the <filename moreinfo="none">pam_stack.so</filename> module that allows all 
authentication to be configured in a single central file. The 
<filename moreinfo="none">pam_stack.so</filename> method has some devoted followers 
on the basis that it allows for easier administration. As with all issues in 
life, though, every decision has trade-offs, so you may want to examine the 
PAM documentation for further helpful information.
</para></note>

</sect3>

</sect2>

<sect2>
<title><filename moreinfo="none">smb.conf</filename> PAM Configuration</title>

<para>
There is an option in <filename moreinfo="none">smb.conf</filename> called <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="OBEYPAMRESTRICTIONS" xlink:href="smb.conf.5.html#OBEYPAMRESTRICTIONS">obey pam restrictions</link>.
The following is from the online help for this option in SWAT:
</para>

<blockquote>
<para>
When Samba is configured to enable PAM support (i.e., <option>--with-pam</option>), this parameter will 
control whether or not Samba should obey PAM's account and session management directives. The default behavior 
is to use PAM for clear-text authentication only and to ignore any account or session management. Samba always 
ignores PAM for authentication in the case of <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="ENCRYPTPASSWORDS" xlink:href="smb.conf.5.html#ENCRYPTPASSWORDS">encrypt passwords = yes</link>. 
The reason is that PAM modules cannot support the challenge/response authentication mechanism needed in the presence of SMB 
password encryption. 
</para>

<para>Default: <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="OBEYPAMRESTRICTIONS" xlink:href="smb.conf.5.html#OBEYPAMRESTRICTIONS">obey pam restrictions = no</link></para>
</blockquote>

</sect2>

<sect2>
<title>Remote CIFS Authentication Using <filename moreinfo="none">winbindd.so</filename></title>

<para>
All operating systems depend on the provision of user credentials acceptable to the platform.
UNIX requires the provision of a user identifier (UID) as well as a group identifier (GID).
These are both simple integer numbers that are obtained from a password backend such
as <filename moreinfo="none">/etc/passwd</filename>.
</para>

<para>
Users and groups on a Windows NT server are assigned a relative ID (RID) which is unique for
the domain when the user or group is created. To convert the Windows NT user or group into
a  UNIX user or group, a mapping between RIDs and UNIX user and group IDs is required. This
is one of the jobs that winbind performs.
</para>

<para>
As winbind users and groups are resolved from a server, user and group IDs are allocated
from a specified range. This is done on a first come, first served basis, although all
existing users and groups will be mapped as soon as a client performs a user or  group 
enumeration command. The allocated UNIX IDs are stored in a database file under the Samba
lock directory and will be remembered.
</para>

<para>
The astute administrator will realize from this that the combination of <filename moreinfo="none">pam_smbpass.so</filename>, 
<literal>winbindd</literal>, and a distributed <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PASSDBBACKEND" xlink:href="smb.conf.5.html#PASSDBBACKEND">passdb backend</link>
such as <parameter moreinfo="none">ldap</parameter> will allow the establishment of a centrally managed, distributed user/password
database that can also be used by all PAM-aware (e.g., Linux) programs and applications. This arrangement can have
particularly potent advantages compared with the use of Microsoft Active Directory Service (ADS) insofar as
the reduction of wide-area network authentication traffic.
</para>

<warning><para>
The RID to UNIX ID database is the only location where the user and group  mappings are 
stored by <literal>winbindd</literal>. If this file is deleted or corrupted, there is no way for <literal>winbindd</literal>
to determine which user and group IDs correspond to Windows NT user and group RIDs.
</para></warning>

</sect2>

<sect2>
<title>Password Synchronization Using <filename moreinfo="none">pam_smbpass.so</filename></title>

<para>
<filename moreinfo="none">pam_smbpass</filename> is a PAM module that can be used on conforming systems to
keep the <filename moreinfo="none">smbpasswd</filename> (Samba password) database in sync with the UNIX
password file. PAM is an API supported
under some UNIX operating systems, such as Solaris, HPUX, and Linux, that provides a
generic interface to authentication mechanisms.
</para>

<para>
This module authenticates a local <filename moreinfo="none">smbpasswd</filename> user database. If you require
support for authenticating against a remote SMB server, or if you are
concerned about the presence of SUID root binaries on your system, it is
recommended that you use <filename moreinfo="none">pam_winbind</filename> instead.
</para>

<para>
Options recognized by this module are shown in <link linkend="smbpassoptions">next table</link>.
<table frame="all" id="smbpassoptions">
	<title>Options recognized by <parameter moreinfo="none">pam_smbpass</parameter></title>
	<tgroup cols="2" align="left">
		<colspec align="left"/>
		<colspec align="justify" colwidth="1*"/>
	<tbody>
		<row><entry>debug</entry><entry>Log more debugging info.</entry></row>
		<row><entry>audit</entry><entry>Like debug, but also logs unknown usernames.</entry></row>
		<row><entry>use_first_pass</entry><entry>Do not prompt the user for passwords; take them from PAM_ items instead.</entry></row>
		<row><entry>try_first_pass</entry><entry>Try to get the password from a previous PAM module; fall back to prompting the user.</entry></row>
		<row><entry>use_authtok</entry>
			<entry>Like try_first_pass, but *fail* if the new PAM_AUTHTOK has not been previously set (intended for stacking password modules only).</entry></row>
		<row><entry>not_set_pass</entry><entry>Do not make passwords used by this module available to other modules.</entry></row>
		<row><entry>nodelay</entry><entry>dDo not insert ~1-second delays on authentication failure.</entry></row>
		<row><entry>nullok</entry><entry>Null passwords are allowed.</entry></row>
		<row><entry>nonull</entry><entry>Null passwords are not allowed. Used to override the Samba configuration.</entry></row>
		<row><entry>migrate</entry><entry>Only meaningful in an <quote>auth</quote> context; used to update smbpasswd file with a password used for successful authentication.</entry></row>
		<row><entry>smbconf=<replaceable>file</replaceable></entry><entry>Specify an alternate path to the <filename moreinfo="none">smb.conf</filename> file.</entry></row>
	</tbody>
</tgroup>
</table>
</para>

<para>
The following are examples of the use of <filename moreinfo="none">pam_smbpass.so</filename> in the format of the Linux
<filename moreinfo="none">/etc/pam.d/</filename> files structure. Those wishing to implement this
tool on other platforms will need to adapt this appropriately.
</para>

<sect3>
<title>Password Synchronization Configuration</title>

<para>
The following is a sample PAM configuration that shows the use of pam_smbpass to make
sure <filename moreinfo="none">private/smbpasswd</filename> is kept in sync when <filename moreinfo="none">/etc/passwd (/etc/shadow)</filename>
is changed. It is useful when an expired password might be changed by an
application (such as <literal>ssh</literal>).
</para>

<para>
	<programlisting format="linespecific">
#%PAM-1.0
# password-sync
#
auth       requisite    pam_nologin.so
auth       required     pam_unix.so
account    required     pam_unix.so
password   requisite    pam_cracklib.so retry=3
password   requisite    pam_unix.so shadow md5 use_authtok try_first_pass
password   required     pam_smbpass.so nullok use_authtok try_first_pass
session    required     pam_unix.so
</programlisting></para>
</sect3>

<sect3>
<title>Password Migration Configuration</title>

<para>
The following PAM configuration shows the use of <filename moreinfo="none">pam_smbpass</filename> to migrate
from plaintext to encrypted passwords for Samba. Unlike other methods,
this can be used for users who have never connected to Samba shares:
password migration takes place when users <literal>ftp</literal> in, login using <literal>ssh</literal>, pop
their mail, and so on.
</para>

<para>
	<programlisting format="linespecific">
#%PAM-1.0
# password-migration
#
auth       requisite   pam_nologin.so
# pam_smbpass is called IF pam_unix succeeds.
auth       requisite   pam_unix.so
auth       optional    pam_smbpass.so migrate
account    required    pam_unix.so
password   requisite   pam_cracklib.so retry=3
password   requisite   pam_unix.so shadow md5 use_authtok try_first_pass
password   optional    pam_smbpass.so nullok use_authtok try_first_pass
session    required    pam_unix.so
</programlisting></para>
</sect3>

<sect3>
<title>Mature Password Configuration</title>

<para>
The following is a sample PAM configuration for a mature <filename moreinfo="none">smbpasswd</filename> installation.
<filename moreinfo="none">private/smbpasswd</filename> is fully populated, and we consider it an error if
the SMB password does not exist or does not match the UNIX password.
</para>

<para>
<programlisting format="linespecific">
#%PAM-1.0
# password-mature
#
auth       requisite    pam_nologin.so
auth       required     pam_unix.so
account    required     pam_unix.so
password   requisite    pam_cracklib.so retry=3
password   requisite    pam_unix.so shadow md5 use_authtok try_first_pass
password   required     pam_smbpass.so use_authtok use_first_pass
session    required     pam_unix.so
</programlisting></para>
</sect3>

<sect3>
<title>Kerberos Password Integration Configuration</title>

<para>
The following is a sample PAM configuration that shows <parameter moreinfo="none">pam_smbpass</parameter> used together with
<parameter moreinfo="none">pam_krb5</parameter>. This could be useful on a Samba PDC that is also a member of
a Kerberos realm.
</para>

<para>
		<programlisting format="linespecific">
#%PAM-1.0
# kdc-pdc
#
auth       requisite   pam_nologin.so
auth       requisite   pam_krb5.so
auth       optional    pam_smbpass.so migrate
account    required    pam_krb5.so
password   requisite   pam_cracklib.so retry=3
password   optional    pam_smbpass.so nullok use_authtok try_first_pass
password   required    pam_krb5.so use_authtok try_first_pass
session    required    pam_krb5.so
</programlisting></para>

</sect3>

</sect2>

</sect1>

<sect1>
<title>Common Errors</title>

<para>
PAM can be fickle and sensitive to configuration glitches. Here we look at a few cases from
the Samba mailing list.
</para>

	<sect2>
	<title>pam_winbind Problem</title>

	<para>
	A user reported, <emphasis>I have the following PAM configuration</emphasis>:
	</para>

<para>
<programlisting format="linespecific">
auth required /lib/security/pam_securetty.so
auth sufficient /lib/security/pam_winbind.so
auth sufficient /lib/security/pam_unix.so use_first_pass nullok
auth required /lib/security/pam_stack.so service=system-auth
auth required /lib/security/pam_nologin.so
account required /lib/security/pam_stack.so service=system-auth
account required /lib/security/pam_winbind.so
password required /lib/security/pam_stack.so service=system-auth
</programlisting>
</para>

	<para>
	<emphasis>When I open a new console with [ctrl][alt][F1], I can't log in with my user <quote>pitie.</quote>
	I have tried with user <quote>scienceu\pitie</quote> also.</emphasis>
	</para>

	<para>
	The problem may lie with the inclusion of <parameter moreinfo="none">pam_stack.so
	service=system-auth</parameter>. That file often contains a lot of stuff that may
	duplicate what you are already doing. Try commenting out the <parameter moreinfo="none">pam_stack</parameter> lines
	for <parameter moreinfo="none">auth</parameter> and <parameter moreinfo="none">account</parameter> and see if things work. If they do, look at
	<filename moreinfo="none">/etc/pam.d/system-auth</filename> and copy only what you need from it into your
	<filename moreinfo="none">/etc/pam.d/login</filename> file. Alternatively, if you want all services to use
	Winbind, you can put the Winbind-specific stuff in <filename moreinfo="none">/etc/pam.d/system-auth</filename>.
	</para>

	</sect2>

	<sect2>
	<title>Winbind Is Not Resolving Users and Groups</title>

	<para>
	<quote>
	My <filename moreinfo="none">smb.conf</filename> file is correctly configured. I have specified 
	<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="IDMAPUID" xlink:href="smb.conf.5.html#IDMAPUID">idmap uid = 12000</link> 
	and <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="IDMAPGID" xlink:href="smb.conf.5.html#IDMAPGID">idmap gid = 3000-3500,</link>
	and <literal>winbind</literal> is running. When I do the following it all works fine.
	</quote>
	</para>

<para><screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">wbinfo -u</userinput>
MIDEARTH\maryo
MIDEARTH\jackb
MIDEARTH\ameds
...
MIDEARTH\root

<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">wbinfo -g</userinput>
MIDEARTH\Domain Users
MIDEARTH\Domain Admins
MIDEARTH\Domain Guests
...
MIDEARTH\Accounts

<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">getent passwd</userinput>
root:x:0:0:root:/root:/bin/bash
bin:x:1:1:bin:/bin:/bin/bash
...
maryo:x:15000:15003:Mary Orville:/home/MIDEARTH/maryo:/bin/false
</screen></para>

	<para>
	<quote>
	But this command fails:
	</quote>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">chown maryo a_file</userinput>
chown: 'maryo': invalid user
</screen>
	<quote>This is driving me nuts! What can be wrong?</quote>
	</para>

	<para>
	Your system is likely running <literal>nscd</literal>, the name service
	caching daemon. Shut it down, do not restart it! You will find your problem resolved.
	</para>

	</sect2>
</sect1>

</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="integrate-ms-networks">
 
<chapterinfo>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
        <pubdate> (Jan 01 2001) </pubdate>
</chapterinfo>
 
<title>Integrating MS Windows Networks with Samba</title>
 
<para>
<indexterm significance="normal"><primary>NetBIOS</primary></indexterm>
This chapter deals with NetBIOS over TCP/IP name to IP address resolution. If
your MS Windows clients are not configured to use NetBIOS over TCP/IP, then this
section does not apply to your installation. If your installation involves the use of
NetBIOS over TCP/IP, then this chapter may help you to resolve networking problems.
</para>

<note>
<para>
<indexterm significance="normal"><primary>NetBEUI</primary></indexterm>
<indexterm significance="normal"><primary>LLC</primary></indexterm>
NetBIOS over TCP/IP has nothing to do with NetBEUI. NetBEUI is NetBIOS
over Logical Link Control (LLC). On modern networks it is highly advised
to not run NetBEUI at all. Note also that there is no such thing as
NetBEUI over TCP/IP <?latex --- ?> the existence of such a protocol is a complete
and utter misapprehension.
</para>
</note>

<sect1>
<title>Features and Benefits</title>

<para>
Many MS Windows network administrators have never been exposed to basic TCP/IP
networking as it is implemented in a UNIX/Linux operating system. Likewise, many UNIX and
Linux administrators have not been exposed to the intricacies of MS Windows TCP/IP-based
networking (and may have no desire to be, either).
</para>

<para>
This chapter gives a short introduction to the basics of how a name can be resolved to 
its IP address for each operating system environment.
</para>

</sect1>

<sect1>
<title>Background Information</title>

<para>
<indexterm significance="normal"><primary>NetBIOS over TCP/IP</primary></indexterm>
<indexterm significance="normal"><primary>UDP port 137</primary></indexterm>
<indexterm significance="normal"><primary>TCP port 139</primary></indexterm>
<indexterm significance="normal"><primary>TCP port 445</primary></indexterm>
<indexterm significance="normal"><primary>UDP port 137</primary></indexterm>
Since the introduction of MS Windows 2000, it is possible to run MS Windows networking
without the use of NetBIOS over TCP/IP. NetBIOS over TCP/IP uses UDP port 137 for NetBIOS
name resolution and uses TCP port 139 for NetBIOS session services. When NetBIOS over
TCP/IP is disabled on MS Windows 2000 and later clients, then only the TCP port 445 is
used, and the UDP port 137 and TCP port 139 are not.
</para>

<note>
<para>
When using Windows 2000 or later clients, if NetBIOS over TCP/IP is not disabled, then
the client will use UDP port 137 (NetBIOS Name Service, also known as the Windows Internet
Name Service, or WINS), TCP port 139, and TCP port 445 (for actual file and print traffic).
</para>
</note>

<para>
<indexterm significance="normal"><primary>DNS</primary></indexterm>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
<indexterm significance="normal"><primary>DDNS</primary></indexterm>
<indexterm significance="normal"><primary>SRV RR</primary></indexterm>
<indexterm significance="normal"><primary>IXFR</primary></indexterm>
<indexterm significance="normal"><primary>DHCP</primary></indexterm>
When NetBIOS over TCP/IP is disabled, the use of DNS is essential. Most installations that disable NetBIOS
over TCP/IP today use MS Active Directory Service (ADS). ADS requires
<indexterm significance="normal"><primary>DNS</primary><secondary>Dynamic</secondary></indexterm> dynamic DNS with Service Resource
Records (SRV RR) and with Incremental Zone Transfers (IXFR).  <indexterm significance="normal"><primary>DHCP</primary></indexterm>
Use of DHCP with ADS is recommended as a further means of maintaining central control over the client
workstation network configuration.
</para>

</sect1>

<sect1>
<title>Name Resolution in a Pure UNIX/Linux World</title>

<para>
The key configuration files covered in this section are:
</para>

<indexterm significance="normal"><primary>/etc/hosts</primary></indexterm>
<indexterm significance="normal"><primary>/etc/resolv.conf</primary></indexterm>
<indexterm significance="normal"><primary>/etc/host.conf</primary></indexterm>
<indexterm significance="normal"><primary>/etc/nsswitch.conf</primary></indexterm>

<itemizedlist>
	<listitem><para><filename moreinfo="none">/etc/hosts</filename></para></listitem>
	<listitem><para><filename moreinfo="none">/etc/resolv.conf</filename></para></listitem>
	<listitem><para><filename moreinfo="none">/etc/host.conf</filename></para></listitem>
	<listitem><para><filename moreinfo="none">/etc/nsswitch.conf</filename></para></listitem>
</itemizedlist>

<sect2>
<title><filename moreinfo="none">/etc/hosts</filename></title>

<para>
This file contains a static list of IP addresses and names.
<programlisting format="linespecific">
127.0.0.1	localhost localhost.localdomain
192.168.1.1	bigbox.quenya.org	bigbox	alias4box
</programlisting>
</para>

<para>
<indexterm significance="normal"><primary>/etc/hosts&gt;</primary></indexterm>
<indexterm significance="normal"><primary>name resolution</primary></indexterm>
The purpose of <filename moreinfo="none">/etc/hosts</filename> is to provide a 
name resolution mechanism so users do not need to remember 
IP addresses.
</para>

<para>
<indexterm significance="normal"><primary>IP addresses</primary></indexterm>
<indexterm significance="normal"><primary>MAC address</primary></indexterm>
<indexterm significance="normal"><primary>physical network transport layer</primary></indexterm>
Network packets that are sent over the physical network transport 
layer communicate not via IP addresses but rather using the Media 
Access Control address, or MAC address. IP addresses are currently 
32 bits in length and are typically presented as four decimal 
numbers that are separated by a dot (or period) <?latex --- ?> for example, 168.192.1.1.
</para>

<para>
<indexterm significance="normal"><primary>MAC Addresses</primary></indexterm>
MAC addresses use 48 bits (or 6 bytes) and are typically represented 
as two-digit hexadecimal numbers separated by colons: 40:8e:0a:12:34:56.
</para>

<para>
Every network interface must have a MAC address. Associated with a MAC address may be one or more IP
addresses. There is no relationship between an IP address and a MAC address; all such assignments are
arbitrary or discretionary in nature. At the most basic level, all network communications take place using MAC
addressing. Since MAC addresses must be globally unique and generally remain fixed for any particular
interface, the assignment of an IP address makes sense from a network management perspective. More than one IP
address can be assigned per MAC address. One address must be the primary IP address <?latex --- ?> this is the
address that will be returned in the Address Resolution Protocol (ARP) reply.
</para>

<para>
<indexterm significance="normal"><primary>machine name</primary></indexterm>
When a user or a process wants to communicate with another machine, 
the protocol implementation ensures that the <quote>machine name</quote> or <quote>host 
name</quote> is resolved to an IP address in a manner that is controlled 
by the TCP/IP configuration control files. The file 
<filename moreinfo="none">/etc/hosts</filename> is one such file.
</para>

<para>
<indexterm significance="normal"><primary>ARP/RARP</primary></indexterm>
When the IP address of the destination interface has been determined, a protocol called ARP/RARP is used to
identify the MAC address of the target interface. ARP is a broadcast-oriented method that uses User Datagram
Protocol (UDP) to send a request to all interfaces on the local network segment using the all 1s MAC address.
Network interfaces are programmed to respond to two MAC addresses only; their own unique address and the
address ff:ff:ff:ff:ff:ff. The reply packet from an ARP request will contain the MAC address and the primary
IP address for each interface.
</para>

<para>
<indexterm significance="normal"><primary>/etc/hosts</primary></indexterm>
The <filename moreinfo="none">/etc/hosts</filename> file is foundational to all 
UNIX/Linux TCP/IP installations and as a minimum will contain 
the localhost and local network interface IP addresses and the 
primary names by which they are known within the local machine. 
This file helps to prime the pump so a basic level of name 
resolution can exist before any other method of name resolution 
becomes available.
</para>

</sect2>


<sect2>
<title><filename moreinfo="none">/etc/resolv.conf</filename></title>

<para>
This file tells the name resolution libraries:
</para>

<itemizedlist>
	<listitem><para>The name of the domain to which the machine 
	belongs.
	</para></listitem>
	
	<listitem><para>The name(s) of any domains that should be 
	automatically searched when trying to resolve unqualified 
	host names to their IP address.
	</para></listitem>
	
	<listitem><para>The name or IP address of available domain 
	name servers that may be asked to perform name-to-address 
	translation lookups.
	</para></listitem>
</itemizedlist>

</sect2>


<sect2>
<title><filename moreinfo="none">/etc/host.conf</filename></title>


<para>
<indexterm significance="normal"><primary>/etc/host.conf</primary></indexterm>
<filename moreinfo="none">/etc/host.conf</filename> is the primary means by which the setting in
<filename moreinfo="none">/etc/resolv.conf</filename> may be effected. It is a critical configuration file. This file controls
the order by which name resolution may proceed. The typical structure is:
<programlisting format="linespecific">
order hosts,bind
multi on
</programlisting></para>

<para>Both addresses should be returned. Please refer to the 
man page for <filename moreinfo="none">host.conf</filename> for further details.
</para>

</sect2>


<sect2>
<title><filename moreinfo="none">/etc/nsswitch.conf</filename></title>

<para>
<indexterm significance="normal"><primary>/etc/nsswitch.conf</primary></indexterm>
This file controls the actual name resolution targets. The 
file typically has resolver object specifications as follows:
<programlisting format="linespecific">
# /etc/nsswitch.conf
#
# Name Service Switch configuration file.
#

passwd:		compat
# Alternative entries for password authentication are:
# passwd:	compat files nis ldap winbind
shadow:		compat
group:		compat

hosts:		files nis dns
# Alternative entries for host name resolution are:
# hosts:	files dns nis nis+ hesiod db compat ldap wins
networks:	nis files dns

ethers:		nis files
protocols:	nis files
rpc:		nis files
services:	nis files
</programlisting></para>

<para>
Of course, each of these mechanisms requires that the appropriate 
facilities and/or services are correctly configured.
</para>

<para>
It should be noted that unless a network request/message must be 
sent, TCP/IP networks are silent. All TCP/IP communications assume a 
principal of speaking only when necessary.
</para>


<para>
<indexterm significance="normal"><primary>libnss_wins.so</primary></indexterm>
<indexterm significance="normal"><primary>NetBIOS names</primary></indexterm>
<indexterm significance="normal"><primary>make</primary></indexterm>
<indexterm significance="normal"><primary>/etc/nsswitch.conf</primary></indexterm>
<indexterm significance="normal"><primary>wins</primary></indexterm>
Starting with version 2.2.0, Samba has Linux support for extensions to 
the name service switch infrastructure so Linux clients will 
be able to obtain resolution of MS Windows NetBIOS names to IP 
addresses. To gain this functionality, Samba needs to be compiled 
with appropriate arguments to the make command (i.e., <userinput moreinfo="none">make 
nsswitch/libnss_wins.so</userinput>). The resulting library should 
then be installed in the <filename moreinfo="none">/lib</filename> directory, and 
the <parameter moreinfo="none">wins</parameter> parameter needs to be added to the <quote>hosts:</quote> line in 
the <filename moreinfo="none">/etc/nsswitch.conf</filename> file. At this point, it 
will be possible to ping any MS Windows machine by its NetBIOS 
machine name, as long as that machine is within the workgroup to 
which both the Samba machine and the MS Windows machine belong.
</para>

</sect2>
</sect1>


<sect1>
<title>Name Resolution as Used within MS Windows Networking</title>

<para>
<indexterm significance="normal"><primary>computer name</primary></indexterm>
<indexterm significance="normal"><primary>machine name</primary></indexterm>
<indexterm significance="normal"><primary>NetBIOS name</primary></indexterm>
<indexterm significance="normal"><primary>SMB name</primary></indexterm>
MS Windows networking is predicated on the name each machine is given. This name is known variously (and
inconsistently) as the <quote>computer name,</quote> <quote>machine name,</quote> <quote>networking
name,</quote> <quote>NetBIOS name,</quote> or <quote>SMB name.</quote> All terms mean the same thing with the
exception of <quote>NetBIOS name,</quote> which can also apply to the name of the workgroup or the domain
name. The terms <quote>workgroup</quote> and <quote>domain</quote> are really just a simple name with which
the machine is associated. All NetBIOS names are exactly 16 characters in length. The
16<superscript>th</superscript> character is reserved.  It is used to store a 1-byte value that indicates
service level information for the NetBIOS name that is registered. A NetBIOS machine name is therefore
registered for each service type that is provided by the client/server.
</para>

<para>
<link linkend="uniqnetbiosnames">Unique NetBIOS names</link> and <link linkend="netbiosnamesgrp">group names</link> tables 
list typical NetBIOS name/service type registrations.
</para>

<table frame="all" id="uniqnetbiosnames">
<title>Unique NetBIOS Names</title>
<tgroup cols="2">
<colspec align="left"/>
<colspec align="justify"/>
<tbody>
<row><entry>MACHINENAME&lt;00&gt;</entry><entry>Server Service is running on MACHINENAME</entry></row>
<row><entry>MACHINENAME&lt;03&gt;</entry><entry>Generic machine name (NetBIOS name)</entry></row>
<row><entry>MACHINENAME&lt;20&gt;</entry><entry>LanMan server service is running on MACHINENAME</entry></row>
<row><entry>WORKGROUP&lt;1b&gt;</entry><entry>Domain master browser</entry></row>
</tbody>
</tgroup>
</table>

<table frame="all" id="netbiosnamesgrp">
<title>Group Names</title>
<tgroup cols="2">
<colspec align="left"/>
<colspec align="justify"/>
<tbody>
<row><entry>WORKGROUP&lt;03&gt;</entry><entry>Generic name registered by all members of WORKGROUP</entry></row>
<row><entry>WORKGROUP&lt;1c&gt;</entry><entry>Domain cntrollers/netlogon servers</entry></row>
<row><entry>WORKGROUP&lt;1d&gt;</entry><entry>Local master browsers</entry></row>
<row><entry>WORKGROUP&lt;1e&gt;</entry><entry>Browser election service</entry></row>
</tbody>
</tgroup>
</table>

<para>
<indexterm significance="normal"><primary>NetBIOS</primary></indexterm>
It should be noted that all NetBIOS machines register their own 
names as per <link linkend="uniqnetbiosnames">Unique NetBIOS names</link> and <link linkend="netbiosnamesgrp">group names</link>. This is in vast contrast to TCP/IP 
installations where the system administrator traditionally 
determines in the <filename moreinfo="none">/etc/hosts</filename> or in the DNS database what names 
are associated with each IP address.
</para>

<para>
<indexterm significance="normal"><primary>NetBIOS</primary></indexterm>
<indexterm significance="normal"><primary>/etc/hosts</primary></indexterm>
<indexterm significance="normal"><primary>NetBIOS name</primary></indexterm>
One further point of clarification should be noted. The <filename moreinfo="none">/etc/hosts</filename> 
file and the DNS records do not provide the NetBIOS name information 
that MS Windows clients depend on to locate the type of service that may 
be needed. An example of this is what happens when an MS Windows client 
wants to locate a domain logon server. It finds this service and the IP 
address of a server that provides it by performing a lookup (via a 
NetBIOS broadcast) for enumeration of all machines that have 
registered the name type *&lt;1C&gt;. A logon request is then sent to each 
IP address that is returned in the enumerated list of IP addresses.
Whichever machine first replies, it then ends up providing the logon services.
</para>

<para>
<indexterm significance="normal"><primary>domain</primary></indexterm>
<indexterm significance="normal"><primary>workgroup</primary></indexterm>
The name <quote>workgroup</quote> or <quote>domain</quote> really can be confusing, since these 
have the added significance of indicating what is the security 
architecture of the MS Windows network. The term <quote>workgroup</quote> indicates 
that the primary nature of the network environment is that of a 
peer-to-peer design. In a workgroup, all machines are responsible for 
their own security, and generally such security is limited to the use of 
just a password (known as share-level security). In most situations 
with peer-to-peer networking, the users who control their own machines 
will simply opt to have no security at all. It is possible to have 
user-level security in a workgroup environment, thus requiring the use 
of a username and a matching password.
</para>

<para>
<indexterm significance="normal"><primary>SMB</primary></indexterm>
<indexterm significance="normal"><primary>Network Basic Input/Output System</primary><see>NetBIOS</see></indexterm>
<indexterm significance="normal"><primary>Logical Link Control</primary><see>LLC</see></indexterm>
<indexterm significance="normal"><primary>Network Basic Extended User Interface</primary><see>NetBEUI</see></indexterm>
<indexterm significance="normal"><primary>Internetworking Packet Exchange</primary><see>IPX</see></indexterm>
<indexterm significance="normal"><primary>NetWare</primary></indexterm>
<indexterm significance="normal"><primary>NetBT</primary></indexterm>
<indexterm significance="normal"><primary>NBT</primary></indexterm>
MS Windows networking is thus predetermined to use machine names 
for all local and remote machine message passing. The protocol used is 
called Server Message Block (SMB), and this is implemented using 
the NetBIOS protocol (Network Basic Input/Output System). NetBIOS can 
be encapsulated using LLC (Logical Link Control) protocol <?latex --- ?> in which case 
the resulting protocol is called NetBEUI (Network Basic Extended User 
Interface). NetBIOS can also be run over IPX (Internetworking Packet 
Exchange) protocol as used by Novell NetWare, and it can be run 
over TCP/IP protocols <?latex --- ?> in which case the resulting protocol is called 
NBT or NetBT, the NetBIOS over TCP/IP.
</para>

<para>
MS Windows machines use a complex array of name resolution mechanisms. 
Since we are primarily concerned with TCP/IP, this demonstration is 
limited to this area.
</para>

<sect2>
<title>The NetBIOS Name Cache</title>

<para>
<indexterm significance="normal"><primary>n-memory buffer</primary></indexterm>
<indexterm significance="normal"><primary>local cache</primary></indexterm>
<indexterm significance="normal"><primary/></indexterm>
All MS Windows machines employ an in-memory buffer in which is 
stored the NetBIOS names and IP addresses for all external 
machines that machine has communicated with over the 
past 10 to 15 minutes. It is more efficient to obtain an IP address 
for a machine from the local cache than it is to go through all the 
configured name resolution mechanisms.
</para>

<para>
<indexterm significance="normal"><primary>name lookup</primary></indexterm>
If a machine whose name is in the local name cache is shut 
down before the name is expired and flushed from the cache, then 
an attempt to exchange a message with that machine will be subject 
to timeout delays. Its name is in the cache, so a name resolution 
lookup will succeed, but the machine cannot respond. This can be 
frustrating for users but is a characteristic of the protocol.
</para>

<para>
<indexterm significance="normal"><primary>nbtstat</primary></indexterm>
<indexterm significance="normal"><primary>nmblookup</primary></indexterm>
<indexterm significance="normal"><primary>NetBIOS</primary></indexterm>
The MS Windows utility that allows examination of the NetBIOS 
name cache is called <quote>nbtstat.</quote> The Samba equivalent
is called <literal>nmblookup</literal>.
</para>

</sect2>

<sect2>
<title>The LMHOSTS File</title>

<para>
<indexterm significance="normal"><primary>LMHOSTS</primary></indexterm>
This file is usually located in MS Windows NT 4.0 or Windows 200x/XP in the directory
<filename moreinfo="none">%SystemRoot%\SYSTEM32\DRIVERS\ETC</filename> and contains the IP address
and the machine name in matched pairs. The <filename moreinfo="none">LMHOSTS</filename> file
performs NetBIOS name to IP address mapping.
</para>

<para>
It typically looks like this:
</para>

<para><programlisting format="linespecific">
# Copyright (c) 1998 Microsoft Corp.
#
# This is a sample LMHOSTS file used by the Microsoft Wins Client (NetBIOS
# over TCP/IP) stack for Windows98
#
# This file contains the mappings of IP addresses to NT computer names
# (NetBIOS) names. Each entry should be kept on an individual line.
# The IP address should be placed in the first column followed by the
# corresponding computer name. The address and the computer name
# should be separated by at least one space or tab. The "#" character
# is generally used to denote the start of a comment (see the exceptions
# below).
#
# This file is compatible with Microsoft LAN Manager 2.x TCP/IP lmhosts
# files and offers the following extensions:
#
#      #PRE
#      #DOM:&lt;domain&gt;
#      #INCLUDE &lt;filename&gt;
#      #BEGIN_ALTERNATE
#      #END_ALTERNATE
#      \0xnn (non-printing character support)
#
# Following any entry in the file with the characters "#PRE" will cause
# the entry to be preloaded into the name cache. By default, entries are
# not preloaded, but are parsed only after dynamic name resolution fails.
#
# Following an entry with the "#DOM:&lt;domain&gt;" tag will associate the
# entry with the domain specified by &lt;domain&gt;. This effects how the
# browser and logon services behave in TCP/IP environments. To preload
# the host name associated with #DOM entry, it is necessary to also add a
# #PRE to the line. The &lt;domain&gt; is always pre-loaded although it will not
# be shown when the name cache is viewed.
#
# Specifying "#INCLUDE &lt;filename&gt;" will force the RFC NetBIOS (NBT)
# software to seek the specified &lt;filename&gt; and parse it as if it were
# local. &lt;filename&gt; is generally a UNC-based name, allowing a
# centralized lmhosts file to be maintained on a server.
# It is ALWAYS necessary to provide a mapping for the IP address of the
# server prior to the #INCLUDE. This mapping must use the #PRE directive.
# In addition the share "public" in the example below must be in the
# LanMan Server list of "NullSessionShares" in order for client machines to
# be able to read the lmhosts file successfully. This key is under
# \machine\system\currentcontrolset\services\lanmanserver\
# parameters\nullsessionshares
# in the registry. Simply add "public" to the list found there.
#
# The #BEGIN_ and #END_ALTERNATE keywords allow multiple #INCLUDE
# statements to be grouped together. Any single successful include
# will cause the group to succeed.
#
# Finally, non-printing characters can be embedded in mappings by
# first surrounding the NetBIOS name in quotations, then using the
# \0xnn notation to specify a hex value for a non-printing character.
#
# The following example illustrates all of these extensions:
#
# 102.54.94.97     rhino     #PRE #DOM:networking  #net group's DC
# 102.54.94.102    "appname  \0x14"       #special app server
# 102.54.94.123    popular   #PRE         #source server
# 102.54.94.117    localsrv  #PRE         #needed for the include
#
# #BEGIN_ALTERNATE
# #INCLUDE \\localsrv\public\lmhosts
# #INCLUDE \\rhino\public\lmhosts
# #END_ALTERNATE
#
# In the above example, the "appname" server contains a special
# character in its name, the "popular" and "localsrv" server names are
# pre-loaded, and the "rhino" server name is specified so it can be used
# to later #INCLUDE a centrally maintained lmhosts file if the "localsrv"
# system is unavailable.
#
# Note that the whole file is parsed including comments on each lookup,
# so keeping the number of comments to a minimum will improve performance.
# Therefore it is not advisable to simply add lmhosts file entries onto the
# end of this file.
</programlisting></para>

</sect2>

<sect2>
<title>HOSTS File</title>

<para>
This file is usually located in MS Windows NT 4.0 or Windows 200x/XP in 
the directory <filename moreinfo="none">%SystemRoot%\SYSTEM32\DRIVERS\ETC</filename> and contains 
the IP address and the IP hostname in matched pairs. It can be 
used by the name resolution infrastructure in MS Windows, depending 
on how the TCP/IP environment is configured. This file is in 
every way the equivalent of the UNIX/Linux <filename moreinfo="none">/etc/hosts</filename> file.
</para>
</sect2>


<sect2>
<title>DNS Lookup</title>


<para>
<indexterm significance="normal"><primary>DNS</primary></indexterm>
This capability is configured in the TCP/IP setup area in the network 
configuration facility. If enabled, an elaborate name resolution sequence 
is followed, the precise nature of which is dependent on how the NetBIOS 
Node Type parameter is configured. A Node Type of 0 means that
NetBIOS broadcast (over UDP broadcast) is used if the name 
that is the subject of a name lookup is not found in the NetBIOS name 
cache. If that fails, then DNS, HOSTS, and LMHOSTS are checked. If set to 
Node Type 8, then a NetBIOS Unicast (over UDP Unicast) is sent to the 
WINS server to obtain a lookup before DNS, HOSTS, LMHOSTS, or broadcast 
lookup is used.
</para>

</sect2>

<sect2>
<title>WINS Lookup</title>


<para>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
<indexterm significance="normal"><primary>Windows Internet Name Server</primary><see>WINS</see></indexterm>
<indexterm significance="normal"><primary>NetBIOS Name Server</primary><see>NBNS</see></indexterm>
A WINS (Windows Internet Name Server) service is the equivalent of the 
rfc1001/1002 specified NBNS (NetBIOS Name Server). A WINS server stores 
the names and IP addresses that are registered by a Windows client 
if the TCP/IP setup has been given at least one WINS server IP address.
</para>

<para>
To configure Samba to be a WINS server, the following parameter needs 
to be added to the <filename moreinfo="none">smb.conf</filename> file:
</para>

<para><simplelist>
<member><indexterm>wins support</indexterm><parameter>wins support = Yes</parameter></member>
</simplelist></para>

<para>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
To configure Samba to use a WINS server, the following parameters are 
needed in the <filename moreinfo="none">smb.conf</filename> file:
</para>

<para><simplelist>
<member><indexterm>wins support</indexterm><parameter>wins support = No</parameter></member>
<member><indexterm>wins server</indexterm><parameter>wins server = xxx.xxx.xxx.xxx</parameter></member>
</simplelist></para>

<para>
where <replaceable>xxx.xxx.xxx.xxx</replaceable> is the IP address 
of the WINS server.
</para>

<para>For information about setting up Samba as a WINS server, read 
<link linkend="NetworkBrowsing">Network Browsing</link>.</para>

</sect2>
</sect1>

<sect1>
<title>Common Errors</title>

<para>
TCP/IP network configuration problems find every network administrator sooner or later.
The cause can be anything from keyboard mishaps to forgetfulness to simple mistakes to
carelessness. Of course, no one is ever deliberately careless!
</para>

	<sect2>
		<title>Pinging Works Only One Way</title>

	<para>
	<quote>I can ping my Samba server from Windows, but I cannot ping my Windows
	machine from the Samba server.</quote>
	</para>

	<para>
	The Windows machine was at IP address 192.168.1.2 with netmask 255.255.255.0, the
	Samba server (Linux) was at IP address 192.168.1.130 with netmask 255.255.255.128.
	The machines were on a local network with no external connections.
	</para>

	<para>
	Due to inconsistent netmasks, the Windows machine was on network 192.168.1.0/24, while
	the Samba server was on network 192.168.1.128/25 <?latex --- ?> logically a different network.
	</para>

	</sect2>

	<sect2>
	<title>Very Slow Network Connections</title>

	<para>
	A common cause of slow network response includes:
	</para>

	<itemizedlist>
		<listitem><para>Client is configured to use DNS and the DNS server is down.</para></listitem>
		<listitem><para>Client is configured to use remote DNS server, but the
		remote connection is down.</para></listitem>
		<listitem><para>Client is configured to use a WINS server, but there is no WINS server.</para></listitem>
		<listitem><para>Client is not configured to use a WINS server, but there is a WINS server.</para></listitem>
		<listitem><para>Firewall is filtering out DNS or WINS traffic.</para></listitem>
	</itemizedlist>

	</sect2>

	<sect2>
	<title>Samba Server Name-Change Problem</title>

	<para>
	<quote>The name of the Samba server was changed, Samba was restarted, and now the Samba server cannot be
	pinged by its new name from an MS Windows NT4 workstation, but it does still respond to pinging using
	the old name. Why?</quote>
	</para>

	<para>
	From this description, three things are obvious:
	</para>

	<itemizedlist>
		<listitem><para>WINS is not in use; only broadcast-based name resolution is used.</para></listitem>
		<listitem><para>The Samba server was renamed and restarted within the last 10 or 15 minutes.</para></listitem>
		<listitem><para>The old Samba server name is still in the NetBIOS name cache on the MS Windows NT4 workstation.</para></listitem>
	</itemizedlist>

	<para>
	To find what names are present in the NetBIOS name cache on the MS Windows NT4 machine,
	open a <literal>cmd</literal> shell and then:
	</para>

	<para>
<screen format="linespecific">
<prompt moreinfo="none">C:\&gt; </prompt><userinput moreinfo="none">nbtstat -n</userinput>

              NetBIOS Local Name Table

   Name                 Type          Status
------------------------------------------------
FRODO            &lt;03&gt;  UNIQUE      Registered
ADMINISTRATOR     &lt;03&gt;  UNIQUE      Registered
FRODO            &lt;00&gt;  UNIQUE      Registered
SARDON           &lt;00&gt;  GROUP       Registered
FRODO            &lt;20&gt;  UNIQUE      Registered
FRODO            &lt;1F&gt;  UNIQUE      Registered


<prompt moreinfo="none">C:\&gt; </prompt>nbtstat -c

             NetBIOS Remote Cache Name Table

   Name                 Type       Host Address     Life [sec]
--------------------------------------------------------------
GANDALF	&lt;20&gt;  UNIQUE      192.168.1.1          240

<prompt moreinfo="none">C:\&gt; </prompt>
</screen>
	</para>

	<para>
	In this example, GANDALF is the Samba server and FRODO is the MS Windows NT4 workstation.
	The first listing shows the contents of the Local Name Table (i.e., identity information on
	the MS Windows workstation), and the second shows the NetBIOS name in the NetBIOS name cache.
	The name cache contains the remote machines known to this workstation.
	</para>

	</sect2>

</sect1>

</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="unicode">
<chapterinfo>
	<author>
<firstname>Jelmer</firstname><othername>R.</othername><surname>Vernooij</surname><othername>R.</othername>
<affiliation>
	<orgname>The Samba Team</orgname>
	<address format="linespecific"><email>jelmer@samba.org</email></address>
</affiliation></author>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
	<author>
		<firstname>TAKAHASHI</firstname><surname>Motonobu</surname>
		<affiliation>
		<address format="linespecific"><email>monyo@home.monyo.com</email></address>
		</affiliation>
		<contrib>Japanese character support</contrib>
	</author>
	<pubdate>25 March 2003</pubdate>
</chapterinfo>

<title>Unicode/Charsets</title>

<sect1>
<title>Features and Benefits</title>

<para>
<indexterm significance="normal"><primary>use computer anywhere</primary></indexterm>
Every industry eventually matures. One of the great areas of maturation is in
the focus that has been given over the past decade to make it possible for anyone
anywhere to use a computer. It has not always been that way. In fact, not so long
ago, it was common for software to be written for exclusive use in the country of
origin.
</para>

<para>
Of all the effort that has been brought to bear on providing native
language support for all computer users, the efforts of the
<ulink url="http://www.openi18n.org/">Openi18n organization</ulink>
is deserving of special mention.
</para>

<para>
<indexterm significance="normal"><primary>codepages</primary></indexterm>
Samba-2.x supported a single locale through a mechanism called 
<emphasis>codepages</emphasis>. Samba-3 is destined to become a truly transglobal
file- and printer-sharing platform.
</para>

</sect1>

<sect1>
<title>What Are Charsets and Unicode?</title>

<para>
<indexterm significance="normal"><primary>character set</primary></indexterm>
Computers communicate in numbers. In texts, each number is 
translated to a corresponding letter. The meaning that will be assigned 
to a certain number depends on the <emphasis>character set (charset)
</emphasis> that is used. 
</para>

<para>
<indexterm significance="normal"><primary>charset</primary></indexterm>
<indexterm significance="normal"><primary>ASCII</primary></indexterm>
A charset can be seen as a table that is used to translate numbers to 
letters. Not all computers use the same charset (there are charsets 
with German umlauts, Japanese characters, and so on). The American Standard Code
for Information Interchange (ASCII) encoding system has been the normative character
encoding scheme used by computers to date. This employs a charset that contains 
256 characters. Using this mode of encoding, each character takes exactly one byte.
</para>

<para>
<indexterm significance="normal"><primary>multibyte charsets</primary></indexterm>
<indexterm significance="normal"><primary>extended characters</primary></indexterm>
There are also charsets that support extended characters, but those need at least
twice as much storage space as does ASCII encoding. Such charsets can contain
<literal>256 * 256 = 65536</literal> characters, which is more than all possible
characters one could think of. They are called multibyte charsets because they use
more then one byte to store one character. 
</para>

<para>
<indexterm significance="normal"><primary>unicode</primary></indexterm>
One standardized multibyte charset encoding scheme is known as
<ulink url="http://www.unicode.org/">unicode</ulink>.  A big advantage of using a
multibyte charset is that you only need one. There is no need to make sure two
computers use the same charset when they are communicating.
</para>

<para>
<indexterm significance="normal"><primary>single-byte charsets</primary></indexterm>
<indexterm significance="normal"><primary>SMB/CIFS</primary></indexterm>
<indexterm significance="normal"><primary>negotiating the charset</primary></indexterm>
Old Windows clients use single-byte charsets, named 
<parameter moreinfo="none">codepages</parameter>, by Microsoft. However, there is no support for 
negotiating the charset to be used in the SMB/CIFS protocol. Thus, you 
have to make sure you are using the same charset when talking to an older client.
Newer clients (Windows NT, 200x, XP) talk Unicode over the wire.
</para>
</sect1>

<sect1>
<title>Samba and Charsets</title>

<para>
<indexterm significance="normal"><primary>Unicode</primary></indexterm>
<indexterm significance="normal"><primary>character sets</primary></indexterm>
As of Samba-3, Samba can (and will) talk Unicode over the wire. Internally, 
Samba knows of three kinds of character sets: 
</para>

<variablelist>
	<varlistentry>
		<term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="UNIXCHARSET" xlink:href="smb.conf.5.html#UNIXCHARSET">unix charset</link></term>
		<listitem><para>
<indexterm significance="normal"><primary>UTF-8</primary></indexterm>
<indexterm significance="normal"><primary>CP850</primary></indexterm>
		This is the charset used internally by your operating system. 
		The default is <constant>UTF-8</constant>, which is fine for most 
		systems and covers all characters in all languages. The default
		in previous Samba releases was to save filenames in the encoding of the 
		clients <?latex --- ?> for example, CP850 for Western European countries.
		</para></listitem>
	</varlistentry>

	<varlistentry>
		<term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="DISPLAYCHARSET" xlink:href="smb.conf.5.html#DISPLAYCHARSET">display charset</link></term>
		<listitem><para>This is the charset Samba uses to print messages
		on your screen. It should generally be the same as the <parameter moreinfo="none">unix charset</parameter>.
		</para></listitem>
	</varlistentry>

	<varlistentry>
		<term><link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="DOSCHARSET" xlink:href="smb.conf.5.html#DOSCHARSET">dos charset</link></term>
		<listitem><para>This is the charset Samba uses when communicating with 
		DOS and Windows 9x/Me clients. It will talk Unicode to all newer clients.
		The default depends on the charsets you have installed on your system.
		Run <literal>testparm -v | grep "dos charset"</literal> to see 
		what the default is on your system. 
		</para></listitem>
	</varlistentry>
</variablelist>

</sect1>

<sect1>
<title>Conversion from Old Names</title>

<para>
<indexterm significance="normal"><primary>charset conversion</primary></indexterm>
Because previous Samba versions did not do any charset conversion, 
characters in filenames are usually not correct in the UNIX charset but only 
for the local charset used by the DOS/Windows clients.
</para>

<para>Bjoern Jacke has written a utility named <ulink url="http://j3e.de/linux/convmv/">convmv</ulink>
that can convert whole directory structures to different charsets with one single command. 
</para>

</sect1>

<sect1>
<title>Japanese Charsets</title>

<para>
Setting up Japanese charsets is quite difficult. This is mainly because:
</para>

<itemizedlist>
	<listitem><para>
<indexterm significance="normal"><primary>JIS X 0208</primary></indexterm>
		The Windows character set is extended from the original legacy Japanese
		standard (JIS X 0208) and is not standardized. This means that the strictly
		standardized implementation cannot support the full Windows character set.
	</para></listitem>

	<listitem><para>
<indexterm significance="normal"><primary>Shift_JIS</primary></indexterm>
<indexterm significance="normal"><primary>EUC-JP</primary></indexterm>
<indexterm significance="normal"><primary>CAP</primary></indexterm>
<indexterm significance="normal"><primary>HEX</primary></indexterm>
<indexterm significance="normal"><primary>Japanese</primary></indexterm>
		Mainly for historical reasons, there are several encoding methods in
		Japanese, which are not fully compatible with each other. There are
		two major encoding methods. One is the Shift_JIS series used in Windows
		and some UNIXes. The other is the EUC-JP series used in most UNIXes
		and Linux. Moreover, Samba previously also offered several unique encoding
		methods, named CAP and HEX, to keep interoperability with CAP/NetAtalk and
		UNIXes that can't use Japanese filenames.  Some implementations of the
		EUC-JP series can't support the full Windows character set.
	</para></listitem>

	<listitem><para>There are some code conversion tables between Unicode and legacy
		Japanese character sets. One is compatible with Windows, another one
		is based on the reference of the Unicode consortium, and others are 
		a mixed implementation. The Unicode consortium does not officially
		define any conversion tables between Unicode and legacy character
		sets, so there cannot be standard one.
	</para></listitem>

	<listitem><para>The character set and conversion tables available in iconv() depend
		on the iconv library that is available. Next to that, the Japanese locale 
		names may be different on different systems.  This means that the value of 
		the charset parameters depends on the implementation of iconv() you are using.
		</para>

		<para>
<indexterm significance="normal"><primary>UCS-2</primary></indexterm>
<indexterm significance="normal"><primary>Shift_JIS</primary></indexterm>
<indexterm significance="normal"><primary>ASCII</primary></indexterm>
<indexterm significance="normal"><primary>English</primary></indexterm>
		Though 2-byte fixed UCS-2 encoding is used in Windows internally,
		Shift_JIS series encoding is usually used in Japanese environments
		as ASCII encoding is in English environments.
	</para></listitem>
</itemizedlist>

<sect2><title>Basic Parameter Setting</title>

	<para>
<indexterm significance="normal"><primary>CP932</primary></indexterm>
	The <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="DOSCHARSET" xlink:href="smb.conf.5.html#DOSCHARSET">dos charset</link> and 
	<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="DISPLAYCHARSET" xlink:href="smb.conf.5.html#DISPLAYCHARSET">display charset</link>
	should be set to the locale compatible with the character set 
	and encoding method used on Windows. This is usually CP932
	but sometimes has a different name.
	</para>

	<para>
<indexterm significance="normal"><primary>Shift_JIS</primary></indexterm>
<indexterm significance="normal"><primary>UTF-8</primary></indexterm>
<indexterm significance="normal"><primary>EUC-JP</primary></indexterm>
	The <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="UNIXCHARSET" xlink:href="smb.conf.5.html#UNIXCHARSET">unix charset</link> can be either Shift_JIS series,
	EUC-JP series, or UTF-8. UTF-8 is always available, but the availability of other locales
	and the name itself depends on the system.
	</para>

	<para>
	Additionally, you can consider using the Shift_JIS series as the
	value of the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="UNIXCHARSET" xlink:href="smb.conf.5.html#UNIXCHARSET">unix charset</link>
	parameter by using the vfs_cap module, which does the same thing as
	setting <quote>coding system = CAP</quote> in the Samba 2.2 series.
	</para>

	<para>
	Where to set <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="UNIXCHARSET" xlink:href="smb.conf.5.html#UNIXCHARSET">unix charset</link>
	to is a difficult question. Here is a list of details, advantages, and
	disadvantages of using a certain value.
	</para>

	<variablelist>
		<varlistentry><term>Shift_JIS series</term>
			<listitem><para>
			Shift_JIS series means a locale that is equivalent to <constant>Shift_JIS</constant>,
			used as a standard on Japanese Windows. In the case of <constant>Shift_JIS</constant>,
			for example, if a Japanese filename consists of 0x8ba4 and 0x974c
			(a 4-bytes Japanese character string meaning <quote>share</quote>) and <quote>.txt</quote>
			is written from Windows on Samba, the filename on UNIX becomes
			0x8ba4, 0x974c, <quote>.txt</quote> (an 8-byte BINARY string), same as Windows.
			</para>

			<para>Since Shift_JIS series is usually used on some commercial-based
			UNIXes; hp-ux and AIX as the Japanese locale (however, it is also possible
			to use the EUC-JP locale series). To use Shift_JIS series on these platforms,
			Japanese filenames created from Windows can be referred to also on
			UNIX.</para>

			<para>
			If your UNIX is already working with Shift_JIS and there is a user 
			who needs to use Japanese filenames written from Windows, the
			Shift_JIS series is the best choice.  However, broken filenames
			may be displayed, and some commands that cannot handle non-ASCII
			filenames may be aborted during parsing filenames. Especially, there
			may be <quote>\ (0x5c)</quote> in filenames, which need to be handled carefully.
			It is best to not touch filenames written from Windows on UNIX.
			</para>

			<para>
			Note that most Japanized free software actually works with EUC-JP
			only. It is good practice to verify that the Japanized free software can work
			with Shift_JIS.
			</para>
			</listitem>
		</varlistentry>

		<varlistentry><term>EUC-JP series</term>
			<listitem><para>
<indexterm significance="normal"><primary>EUC-JP</primary></indexterm>
<indexterm significance="normal"><primary>Japanese UNIX</primary></indexterm>
			EUC-JP series means a locale that is equivalent to the industry
			standard called EUC-JP, widely used in Japanese UNIX (although EUC
			contains specifications for languages other than Japanese, such as
			EUC-KR). In the case of EUC-JP series, for example, if a Japanese
			filename consists of 0x8ba4 and 0x974c and <quote>.txt</quote> is written from
			Windows on Samba, the filename on UNIX becomes 0xb6a6, 0xcdad,
			<quote>.txt</quote> (an 8-byte BINARY string). 
			</para>

			<para>
<indexterm significance="normal"><primary>EUC-JP</primary></indexterm>
<indexterm significance="normal"><primary>UNIX</primary></indexterm>
<indexterm significance="normal"><primary>Linux</primary></indexterm>
<indexterm significance="normal"><primary>FreeBSD</primary></indexterm>
<indexterm significance="normal"><primary>Solaris</primary></indexterm>
<indexterm significance="normal"><primary>IRIX</primary></indexterm>
<indexterm significance="normal"><primary>Tru64 UNIX</primary></indexterm>
<indexterm significance="normal"><primary>Japanese locale</primary></indexterm>
<indexterm significance="normal"><primary>Shift_JIS</primary></indexterm>
<indexterm significance="normal"><primary>UTF-8</primary></indexterm>
			Since EUC-JP is usually used on open source UNIX, Linux, and FreeBSD, and on commercial-based UNIX, Solaris,
			IRIX, and Tru64 UNIX as Japanese locale (however, it is also possible on Solaris to use Shift_JIS and UTF-8,
			and on Tru64 UNIX it is possible to use Shift_JIS). To use EUC-JP series, most Japanese filenames created from
			Windows can be referred to also on UNIX. Also, most Japanized free software works mainly with EUC-JP only.
			</para>

			<para>
			It is recommended to choose EUC-JP series when using Japanese filenames on UNIX.
			</para>

			<para>
			Although there is no character that needs to be carefully treated
			like <quote>\ (0x5c)</quote>, broken filenames may be displayed and some
			commands that cannot handle non-ASCII filenames may be aborted
			during parsing filenames.
			</para>

			<para>
<indexterm significance="normal"><primary>eucJP-ms locale</primary></indexterm>
			Moreover, if you built Samba using differently installed libiconv,
			the eucJP-ms locale included in libiconv and EUC-JP series locale
			included in the operating system may not be compatible. In this case, you may need to
			avoid using incompatible characters for filenames.
			</para>
			</listitem>
		</varlistentry>

		<varlistentry><term>UTF-8</term>
			<listitem><para>
			UTF-8 means a locale equivalent to UTF-8, the international standard defined by the Unicode consortium. In
			UTF-8, a <parameter moreinfo="none">character</parameter> is expressed using 1 to 3 bytes. In case of the Japanese language,
			most characters are expressed using 3 bytes. Since on Windows Shift_JIS, where a character is expressed with 1
			or 2 bytes is used to express Japanese, basically a byte length of a UTF-8 string the length of the UTF-8
			string is 1.5 times that of the original Shift_JIS string. In the case of UTF-8, for example, if a Japanese
			filename consists of 0x8ba4 and 0x974c, and <quote>.txt</quote> is written from Windows on Samba, the filename
			on UNIX becomes 0xe585, 0xb1e6, 0x9c89, <quote>.txt</quote> (a 10-byte BINARY string).
			</para>

			<para>
			For systems where iconv() is not available or where iconv()'s locales
			are not compatible with Windows, UTF-8 is the only locale available.
			</para>

			<para> 
			There are no systems that use UTF-8 as the default locale for Japanese.
			</para>

			<para>
			Some broken filenames may be displayed, and some commands that
			cannot handle non-ASCII filenames may be aborted during parsing
			filenames. Especially, there may be <quote>\ (0x5c)</quote> in filenames, which
			must be handled carefully, so you had better not touch filenames
			written from Windows on UNIX.
			</para>

			<para>
<indexterm significance="normal"><primary>Windows</primary></indexterm>
<indexterm significance="normal"><primary>Java</primary></indexterm>
<indexterm significance="normal"><primary>Unicode UTF-8</primary></indexterm>
			In addition, although it is not directly concerned with Samba, since
			there is a delicate difference between the iconv() function, which is
			generally used on UNIX, and the functions used on other platforms,
			such as Windows and Java, so far is concerens the conversion between
			Shift_JIS and Unicode UTF-8 must be done with care and recognition
			of the limitations involved in the process.
			</para>

			<para>
<indexterm significance="normal"><primary>Mac OS X </primary></indexterm>
			Although Mac OS X uses UTF-8 as its encoding method for filenames,
			it uses an extended UTF-8 specification that Samba cannot handle, so
			UTF-8 locale is not available for Mac OS X.
			</para>
			</listitem>
		</varlistentry>

		<varlistentry><term>Shift_JIS series + vfs_cap (CAP encoding)</term>
			<listitem><para>
<indexterm significance="normal"><primary>CAP</primary></indexterm>
<indexterm significance="normal"><primary>NetAtalk</primary></indexterm>
<indexterm significance="normal"><primary>Macintosh</primary></indexterm>
			CAP encoding means a specification used in CAP and NetAtalk, file
			server software for Macintosh. In the case of CAP encoding, for
			example, if a Japanese filename consists of 0x8ba4 and 0x974c, and
			<quote>.txt</quote> is written from Windows on Samba, the filename on UNIX
			becomes <quote>:8b:a4:97L.txt</quote> (a 14 bytes ASCII string). 
			</para>

			<para>
			For CAP encoding, a byte that cannot be expressed as an ASCII
			character (0x80 or above) is encoded in an <quote>:xx</quote> form. You need to take
			care of containing a <quote>\(0x5c)</quote> in a filename, but filenames are not
			broken in a system that cannot handle non-ASCII filenames.
			</para>

			<para>
			The greatest merit of CAP encoding is the compatibility of encoding
			filenames with CAP or NetAtalk. These are respectively the Columbia Appletalk
			Protocol, and the NetAtalk Open Source software project.
			Since these software applications write a file name on UNIX with CAP encoding, if a
			directory is shared with both Samba and NetAtalk, you need to use
			CAP encoding to avoid non-ASCII filenames from being broken.
			</para>

			<para>
			However, recently, NetAtalk has been
			patched on some systems to write filenames with EUC-JP (e.g., Japanese original Vine Linux).
			In this case, you need to choose EUC-JP series instead of CAP encoding.
			</para>

			<para>
			vfs_cap itself is available for non-Shift_JIS series locales for
			systems that cannot handle non-ASCII characters or systems that
			share files with NetAtalk.
			</para>

			<para>
			To use CAP encoding on Samba-3, you should use the unix charset parameter and VFS 
			as in <link linkend="vfscap-intl">the VFS CAP smb.conf file</link>.
			</para>

<example id="vfscap-intl">
<title>VFS CAP</title>
	<simplelist>
<member> </member><member><parameter>[global]</parameter></member>
<member># the locale name "CP932" may be different</member>
<member><indexterm>dos charset</indexterm><parameter>dos charset = CP932</parameter></member>
<member><indexterm>unix charset</indexterm><parameter>unix charset = CP932</parameter></member>

<member> </member><member><parameter>[cap-share]</parameter></member>
<member><indexterm>vfs option</indexterm><parameter>vfs option = cap</parameter></member>
</simplelist>
</example>

			<para>
<indexterm significance="normal"><primary>CP932</primary></indexterm>
<indexterm significance="normal"><primary>libiconv</primary></indexterm>
<indexterm significance="normal"><primary>unix charset</primary></indexterm>
<indexterm significance="normal"><primary>cap-share</primary></indexterm>
			You should set CP932 if using GNU libiconv for unix charset. With this setting,
			filenames in the <quote>cap-share</quote> share are written with CAP encoding.
			</para>
			</listitem>
		</varlistentry>
	</variablelist>

</sect2>

<sect2><title>Individual Implementations</title>

<para>
Here is some additional information regarding individual implementations:
</para>

	<variablelist>
		<varlistentry><term>GNU libiconv</term>
			<listitem><para>
			To handle Japanese correctly, you should apply the patch
			<ulink url="http://www2d.biglobe.ne.jp/~msyk/software/libiconv-patch.html">libiconv-1.8-cp932-patch.diff.gz</ulink>
			to libiconv-1.8.
			</para>
			
			<para>
			Using the patched libiconv-1.8, these settings are available:
			</para>

<programlisting format="linespecific">
dos charset = CP932
unix charset = CP932 / eucJP-ms / UTF-8
		|       |
		|       +-- EUC-JP series
		+-- Shift_JIS series
display charset = CP932
</programlisting>

			<para>
			Other Japanese locales (for example, Shift_JIS and EUC-JP) should not
			be used because of the lack of the compatibility with Windows.
			</para>
			</listitem>
		</varlistentry>

		<varlistentry><term>GNU glibc</term>
			<listitem><para>
			To handle Japanese correctly, you should apply a <ulink url="http://www2d.biglobe.ne.jp/~msyk/software/glibc/">patch</ulink>
			to glibc-2.2.5/2.3.1/2.3.2 or should use the patch-merged versions, glibc-2.3.3 or later.
			</para>

			<para>
			Using the above glibc, these setting are available:
			<simplelist>
			<member><indexterm>dos charset</indexterm><parameter>dos charset = CP932</parameter></member>
			<member><indexterm>unix charset</indexterm><parameter>unix charset = CP932 / eucJP-ms / UTF-8</parameter></member>
			<member><indexterm>display charset</indexterm><parameter>display charset = CP932</parameter></member>
			</simplelist>
			</para>

			<para>
			Other Japanese locales (for example, Shift_JIS and EUC-JP) should not
			be used because of the lack of the compatibility with Windows.
			</para>
			</listitem>
		</varlistentry>
	</variablelist>

</sect2>

<sect2>
	<title>Migration from Samba-2.2 Series</title>

<para> 
Prior to Samba-2.2 series, the <quote>coding system</quote> parameter was used. The default codepage in Samba
2.x was code page 850. In the Samba-3 series this has been replaced with the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="UNIXCHARSET" xlink:href="smb.conf.5.html#UNIXCHARSET">unix charset</link> parameter.  <link linkend="japancharsets">Japanese Character Sets in Samba-2.2 and Samba-3</link>
shows the mapping table when migrating from the Samba-2.2 series to Samba-3.
</para>

	<table frame="all" id="japancharsets">
		<title>Japanese Character Sets in Samba-2.2 and Samba-3</title>

		<tgroup cols="2" align="center">
			<colspec align="center"/>
			<colspec align="center"/>
			<thead>
				<row><entry>Samba-2.2 Coding System</entry><entry>Samba-3 unix charset</entry></row>
			</thead>
			<tbody>
				<row><entry>SJIS</entry><entry>Shift_JIS series</entry></row>
				<row><entry>EUC</entry><entry>EUC-JP series</entry></row>
				<row><entry>EUC3<footnote><para>Only exists in Japanese Samba version</para></footnote></entry><entry>EUC-JP series</entry></row>
				<row><entry>CAP</entry><entry>Shift_JIS series + VFS</entry></row>
				<row><entry>HEX</entry><entry>currently none</entry></row>
				<row><entry>UTF8</entry><entry>UTF-8</entry></row>
				<row><entry>UTF8-Mac<footnote><para>Only exists in Japanese Samba version</para></footnote></entry><entry>currently none</entry></row>
				<row><entry>others</entry><entry>none</entry></row>
			</tbody>
		</tgroup>
	</table>

</sect2>

</sect1>

<sect1>
	<title>Common Errors</title>

	<sect2>
		<title>CP850.so Can't Be Found</title>

		<para><quote>Samba is complaining about a missing <filename moreinfo="none">CP850.so</filename> file.</quote></para>

		<para>
		CP850 is the default <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="DOSCHARSET" xlink:href="smb.conf.5.html#DOSCHARSET">dos charset</link>.
		The <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="DOSCHARSET" xlink:href="smb.conf.5.html#DOSCHARSET">dos charset</link> is used to convert data to the codepage used by your DOS clients.
		If you do not have any DOS clients, you can safely ignore this message. </para>

		<para>
		CP850 should be supported by your local iconv implementation. Make sure you have all the required packages installed.
		If you compiled Samba from source, make sure that the configure process found iconv. This can be
		confirmed by checking the <filename moreinfo="none">config.log</filename> file that is generated when
		<literal>configure</literal> is executed.</para> 
	</sect2>
</sect1>

</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="Backup">
<chapterinfo>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
</chapterinfo>

<title>Backup Techniques</title>

<sect1>
<title>Features and Benefits</title>

<para>
<indexterm significance="normal"><primary>backup</primary></indexterm>
<indexterm significance="normal"><primary>UNIX system files</primary></indexterm>
<indexterm significance="normal"><primary>system tools</primary></indexterm>
<indexterm significance="normal"><primary>Samba mailing lists</primary></indexterm>
The Samba project is over 10 years old. During the early history
of Samba, UNIX administrators were its key implementors. UNIX administrators
use UNIX system tools to backup UNIX system files. Over the past
4 years, an increasing number of Microsoft network administrators have
taken an interest in Samba. This is reflected in the questions about backup
in general on the Samba mailing lists.
</para>

</sect1>

<sect1>
<title>Discussion of Backup Solutions</title>

<para>
<indexterm significance="normal"><primary>Meccano set</primary></indexterm>
<indexterm significance="normal"><primary>training course</primary></indexterm>
During discussions at a Microsoft Windows training course, one of
the pro-UNIX delegates stunned the class when he pointed out that Windows
NT4 is limiting compared with UNIX. He likened UNIX to a Meccano set
that has an unlimited number of tools that are simple, efficient,
and, in combination, capable of achieving any desired outcome.
</para>

<para>
<indexterm significance="normal"><primary>networking advocates</primary></indexterm>
<indexterm significance="normal"><primary>clear purpose preferred</primary></indexterm>
One of the Windows networking advocates retorted that if she wanted a
Meccano set, she would buy one. She made it clear that a complex single
tool that does more than is needed but does it with a clear purpose and
intent is preferred by some like her.
</para>

<para>
<indexterm significance="normal"><primary>due diligence</primary></indexterm>
<indexterm significance="normal"><primary>research</primary></indexterm>
<indexterm significance="normal"><primary>backup solution</primary></indexterm>
Please note that all information here is provided as is and without recommendation
of fitness or suitability. The network administrator is strongly encouraged to
perform due diligence research before implementing any backup solution, whether free
software or commercial.
</para>

<para>
A useful Web site I recently stumbled across that you might like to refer to
is located at <ulink noescape="1" url="http://www.allmerchants.com/Software/Backup_Software/">
www.allmerchants.com</ulink>.
</para>

<para>
The following three free software projects might also merit consideration.
</para>

	<sect2>
	<title>BackupPC</title>


	<para>
	<indexterm significance="normal"><primary>BackupPC</primary></indexterm>
<indexterm significance="normal"><primary>rsync</primary></indexterm>
<indexterm significance="normal"><primary>rsyncd</primary></indexterm>
	BackupPC version 2.0.0 has been released on <ulink url="http://backuppc.sourceforge.net">SourceForge</ulink>.
	 New features include support for <literal>rsync/rsyncd</literal> and internationalization of the CGI interface
	(including English, French, Spanish, and German).
	</para>

	<para>
<indexterm significance="normal"><primary>BackupPC</primary></indexterm>
<indexterm significance="normal"><primary>laptops</primary></indexterm>
<indexterm significance="normal"><primary>SMB</primary></indexterm>
<indexterm significance="normal"><primary>smbclient</primary></indexterm>
<indexterm significance="normal"><primary>tar</primary></indexterm>
<indexterm significance="normal"><primary>rsh</primary></indexterm>
<indexterm significance="normal"><primary>ssh</primary></indexterm>
<indexterm significance="normal"><primary>rsync</primary></indexterm>
	BackupPC is a high-performance Perl-based package for backing up Linux,
	UNIX, and Windows PCs and laptops to a server's disk. BackupPC is highly
	configurable and easy to install and maintain. SMB (via smbclient),
	<literal>tar</literal> over <literal>rsh/ssh</literal>, or <literal>rsync/rsyncd</literal>
	 are used to extract client data.
	</para>

	<para>
<indexterm significance="normal"><primary>RAID</primary></indexterm>
<indexterm significance="normal"><primary>local disk</primary></indexterm>
<indexterm significance="normal"><primary>network storage</primary></indexterm>
	Given the ever-decreasing cost of disks and RAID systems, it is now
	practical and cost effective to backup a large number of machines onto
	a server's local disk or network storage. This is what BackupPC does.
	</para>

	<para>
	Key features are pooling of identical files (big savings in server disk
	space), compression, and a comprehensive CGI interface that allows users
	to browse backups and restore files.
	</para>

	<para>
<indexterm significance="normal"><primary>GNU GPL</primary></indexterm>
	BackupPC is free software distributed under a GNU GPL license.
	BackupPC runs on Linux/UNIX/freenix servers and has been tested
	on Linux, UNIX, Windows 9x/Me, Windows 98, Windows 200x, Windows XP, and Mac OSX clients.
	</para>

	</sect2>

	<sect2>
	<title>Rsync</title>

	<para>
<indexterm significance="normal"><primary>rsync</primary></indexterm>
<indexterm significance="normal"><primary>ftp</primary></indexterm>
<indexterm significance="normal"><primary>http</primary></indexterm>
<indexterm significance="normal"><primary>scp</primary></indexterm>
<indexterm significance="normal"><primary>rcp</primary></indexterm>
<indexterm significance="normal"><primary>checksum-search</primary></indexterm>
	<literal>rsync</literal> is a flexible program for efficiently copying files or
		directory trees.</para>

	<para><literal>rsync</literal> has many options to select which files will be copied
	  and how they are to be transferred. It may be used as an
	  alternative to <literal>ftp, http, scp</literal>, or <literal>rcp</literal>.</para>

	<para>
<indexterm significance="normal"><primary>remote-update protocol</primary></indexterm>
<indexterm significance="normal"><primary>transfer differences</primary></indexterm>
<indexterm significance="normal"><primary>differences</primary></indexterm>
	The rsync remote-update protocol allows rsync to transfer just
	  the differences between two sets of files across the network link,
	  using an efficient checksum-search algorithm described in the
	  technical report that accompanies the rsync package.</para>

	<para>Some of the additional features of rsync are:</para>

	<itemizedlist>

		<listitem>
		  <para>
		    Support for copying links, devices, owners, groups, and permissions.
		  </para>
		</listitem>

		<listitem>
		  <para>
		    Exclude and exclude-from options are similar to GNU tar.
		  </para>
		</listitem>

		<listitem>
		  <para>
		    A CVS exclude mode for ignoring the same files that CVS would ignore.
		  </para>
		</listitem>

		<listitem>
		  <para>
		    Can use any transparent remote shell, including rsh or ssh.
		  </para>
		</listitem>

		<listitem>
		  <para>
		    Does not require root privileges.
		  </para>
		</listitem>

		<listitem>
		  <para>
		    Pipelining of file transfers to minimize latency costs.
		  </para>
		</listitem>

		<listitem>
		  <para>
		    Support for anonymous or authenticated rsync servers (ideal for
		    mirroring).
		  </para>
		</listitem>
	</itemizedlist>

	</sect2>

	<sect2>
	<title>Amanda</title>


	<para>
	<indexterm significance="normal"><primary>Amanda</primary></indexterm>
<indexterm significance="normal"><primary>native dump</primary></indexterm>
<indexterm significance="normal"><primary>GNU tar</primary></indexterm>
	Amanda, the Advanced Maryland Automatic Network Disk Archiver, is a backup system that
	allows the administrator of a LAN to set up a single master backup server to back up
	multiple hosts to a single large capacity tape drive. Amanda uses native dump and/or
	GNU tar facilities and can back up a large number of workstations running multiple
	versions of UNIX. Recent versions can also use Samba to back up Microsoft Windows hosts.
	</para>

	<para>
	For more information regarding Amanda, please check the <ulink url="http://www.amanda.org/">
	www.amanda.org/ site</ulink>.
	</para>

	</sect2>

	<sect2>
	<title>BOBS: Browseable Online Backup System</title>

	
	<para>
	<indexterm significance="normal"><primary>BOBS</primary></indexterm>
	Browseable Online Backup System (BOBS) is a complete online backup system. Uses large
	disks for storing backups and lets users browse the files using a Web browser. Handles
	some special files like AppleDouble and icon files.
	</para>

	<para>
	The home page for BOBS is located at <ulink url="http://bobs.sourceforge.net/">
	bobs.sourceforge.net</ulink>.
	</para>

	</sect2>

</sect1>

</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="SambaHA">
<chapterinfo>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
	
<author>
	<firstname>Jeremy</firstname><surname>Allison</surname>
	<affiliation>
		<orgname>Samba Team</orgname>
		<address format="linespecific"><email>jra@samba.org</email></address>
	</affiliation>
</author>
</chapterinfo>

<title>High Availability</title>

<sect1>
<title>Features and Benefits</title>

<para>
<indexterm significance="normal"><primary>availability</primary></indexterm>
<indexterm significance="normal"><primary>intolerance</primary></indexterm>
<indexterm significance="normal"><primary>vital task</primary></indexterm>
Network administrators are often concerned about the availability of file and print
services. Network users are inclined toward intolerance of the services they depend
on to perform vital task responsibilities.
</para>

<para>
A sign in a computer room served to remind staff of their responsibilities. It read:
</para>

<blockquote>
<para>
<indexterm significance="normal"><primary>fail</primary></indexterm>
<indexterm significance="normal"><primary>managed by humans</primary></indexterm>
<indexterm significance="normal"><primary>economically wise</primary></indexterm>
<indexterm significance="normal"><primary>anticipate failure</primary></indexterm>
All humans fail, in both great and small ways we fail continually. Machines fail too.
Computers are machines that are managed by humans, the fallout from failure
can be spectacular. Your responsibility is to deal with failure, to anticipate it
and to eliminate it as far as is humanly and economically wise to achieve.
Are your actions part of the problem or part of the solution?
</para>
</blockquote>

<para>
If we are to deal with failure in a planned and productive manner, then first we must
understand the problem. That is the purpose of this chapter.
</para>

<para>
<indexterm significance="normal"><primary>high availability</primary></indexterm>
<indexterm significance="normal"><primary>CIFS/SMB</primary></indexterm>
<indexterm significance="normal"><primary>state of knowledge</primary></indexterm>
Parenthetically, in the following discussion there are seeds of information on how to
provision a network infrastructure against failure. Our purpose here is not to provide
a lengthy dissertation on the subject of high availability. Additionally, we have made
a conscious decision to not provide detailed working examples of high availability
solutions; instead we present an overview of the issues in the hope that someone will
rise to the challenge of providing a detailed document that is focused purely on
presentation of the current state of knowledge and practice in high availability as it
applies to the deployment of Samba and other CIFS/SMB technologies.
</para>

</sect1>

<sect1>
<title>Technical Discussion</title>

<para>
<indexterm significance="normal"><primary>SambaXP conference</primary></indexterm>
<indexterm significance="normal"><primary>Germany</primary></indexterm>
<indexterm significance="normal"><primary>inspired structure</primary></indexterm>
The following summary was part of a presentation by Jeremy Allison at the SambaXP 2003
conference that was held at Goettingen, Germany, in April 2003. Material has been added
from other sources, but it was Jeremy who inspired the structure that follows.
</para>

	<sect2>
	<title>The Ultimate Goal</title>

	<para>
<indexterm significance="normal"><primary>clustering technologies</primary></indexterm>
<indexterm significance="normal"><primary>affordable power</primary></indexterm>
<indexterm significance="normal"><primary>unstoppable services</primary></indexterm>
	All clustering technologies aim to achieve one or more of the following:
	</para>

	<itemizedlist>
		<listitem><para>Obtain the maximum affordable computational power.</para></listitem>
		<listitem><para>Obtain faster program execution.</para></listitem>
		<listitem><para>Deliver unstoppable services.</para></listitem>
		<listitem><para>Avert points of failure.</para></listitem>
		<listitem><para>Exact most effective utilization of resources.</para></listitem>
	</itemizedlist>

	<para>
	A clustered file server ideally has the following properties:
<indexterm significance="normal"><primary>clustered file server</primary></indexterm>
<indexterm significance="normal"><primary>connect transparently</primary></indexterm>
<indexterm significance="normal"><primary>transparently reconnected</primary></indexterm>
<indexterm significance="normal"><primary>distributed file system</primary></indexterm>
	</para>

	<itemizedlist>
		<listitem><para>All clients can connect transparently to any server.</para></listitem>
		<listitem><para>A server can fail and clients are transparently reconnected to another server.</para></listitem>
		<listitem><para>All servers serve out the same set of files.</para></listitem>
		<listitem><para>All file changes are immediately seen on all servers.</para>
			<itemizedlist><listitem><para>Requires a distributed file system.</para></listitem></itemizedlist></listitem>
		<listitem><para>Infinite ability to scale by adding more servers or disks.</para></listitem>
	</itemizedlist>

	</sect2>

	<sect2>
	<title>Why Is This So Hard?</title>

	<para>
	In short, the problem is one of <emphasis>state</emphasis>.
	</para>

	<itemizedlist>
		<listitem>
			<para>
<indexterm significance="normal"><primary>state information</primary></indexterm>
			All TCP/IP connections are dependent on state information.
			</para>
			<para>
<indexterm significance="normal"><primary>TCP failover</primary></indexterm>
			The TCP connection involves a packet sequence number. This
			sequence number would need to be dynamically updated on all
			machines in the cluster to effect seamless TCP failover.
			</para>
		</listitem>
		<listitem>
			<para>
<indexterm significance="normal"><primary>CIFS/SMB</primary></indexterm>
<indexterm significance="normal"><primary>TCP</primary></indexterm>
			CIFS/SMB (the Windows networking protocols) uses TCP connections.
			</para>
			<para>
			This means that from a basic design perspective, failover is not
			seriously considered.
			<itemizedlist>
				<listitem><para>
				All current SMB clusters are failover solutions
				<?latex --- ?> they rely on the clients to reconnect. They provide server
				failover, but clients can lose information due to a server failure.
<indexterm significance="normal"><primary>server failure</primary></indexterm>
				</para></listitem>
			</itemizedlist>
			</para>
		</listitem>
		<listitem>
			<para>
			Servers keep state information about client connections.
			<itemizedlist>
<indexterm significance="normal"><primary>state</primary></indexterm>
				<listitem><para>CIFS/SMB involves a lot of state.</para></listitem>
				<listitem><para>Every file open must be compared with other open files
						to check share modes.</para></listitem>
			</itemizedlist>
			</para>
		</listitem>
	</itemizedlist>

		<sect3>
		<title>The Front-End Challenge</title>

		<para>
<indexterm significance="normal"><primary>cluster servers</primary></indexterm>
<indexterm significance="normal"><primary>single server</primary></indexterm>
<indexterm significance="normal"><primary>TCP data streams</primary></indexterm>
<indexterm significance="normal"><primary>front-end virtual server</primary></indexterm>
<indexterm significance="normal"><primary>virtual server</primary></indexterm>
<indexterm significance="normal"><primary>de-multiplex</primary></indexterm>
<indexterm significance="normal"><primary>SMB</primary></indexterm>
		To make it possible for a cluster of file servers to appear as a single server that has one
		name and one IP address, the incoming TCP data streams from clients must be processed by the
		front-end virtual server. This server must de-multiplex the incoming packets at the SMB protocol
		layer level and then feed the SMB packet to different servers in the cluster.
		</para>

		<para>
<indexterm significance="normal"><primary>IPC$ connections</primary></indexterm>
<indexterm significance="normal"><primary>RPC calls</primary></indexterm>
		One could split all IPC$ connections and RPC calls to one server to handle printing and user
		lookup requirements. RPC printing handles are shared between different IPC4 sessions <?latex --- ?> it is
		hard to split this across clustered servers!
		</para>

		<para>
		Conceptually speaking, all other servers would then provide only file services. This is a simpler
		problem to concentrate on.
		</para>

		</sect3>

		<sect3>
		<title>Demultiplexing SMB Requests</title>

		<para>
<indexterm significance="normal"><primary>SMB requests</primary></indexterm>
<indexterm significance="normal"><primary>SMB state information</primary></indexterm>
<indexterm significance="normal"><primary>front-end virtual server</primary></indexterm>
<indexterm significance="normal"><primary>complicated problem</primary></indexterm>
		De-multiplexing of SMB requests requires knowledge of SMB state information,
		all of which must be held by the front-end <emphasis>virtual</emphasis> server.
		This is a perplexing and complicated problem to solve.
		</para>

		<para>
<indexterm significance="normal"><primary>vuid</primary></indexterm>
<indexterm significance="normal"><primary>tid</primary></indexterm>
<indexterm significance="normal"><primary>fid</primary></indexterm>
		Windows XP and later have changed semantics so state information (vuid, tid, fid)
		must match for a successful operation. This makes things simpler than before and is a
		positive step forward.
		</para>

		<para>
<indexterm significance="normal"><primary>SMB requests</primary></indexterm>
<indexterm significance="normal"><primary>Terminal Server</primary></indexterm>
		SMB requests are sent by vuid to their associated server. No code exists today to
		effect this solution. This problem is conceptually similar to the problem of
		correctly handling requests from multiple requests from Windows 2000
		Terminal Server in Samba.
		</para>

		<para>
<indexterm significance="normal"><primary>de-multiplexing</primary></indexterm>
		One possibility is to start by exposing the server pool to clients directly.
		This could eliminate the de-multiplexing step.
		</para>

		</sect3>

		<sect3>
		<title>The Distributed File System Challenge</title>

		<para>
<indexterm significance="normal"><primary>Distributed File Systems</primary></indexterm>
		There exists many distributed file systems for UNIX and Linux.
		</para>

		<para>
<indexterm significance="normal"><primary>backend</primary></indexterm>
<indexterm significance="normal"><primary>SMB semantics</primary></indexterm>
<indexterm significance="normal"><primary>share modes</primary></indexterm>
<indexterm significance="normal"><primary>locking</primary></indexterm>
<indexterm significance="normal"><primary>oplock</primary></indexterm>
<indexterm significance="normal"><primary>distributed file systems</primary></indexterm>
		Many could be adopted to backend our cluster, so long as awareness of SMB
		semantics is kept in mind (share modes, locking, and oplock issues in particular).
		Common free distributed file systems include:
<indexterm significance="normal"><primary>NFS</primary></indexterm>
<indexterm significance="normal"><primary>AFS</primary></indexterm>
<indexterm significance="normal"><primary>OpenGFS</primary></indexterm>
<indexterm significance="normal"><primary>Lustre</primary></indexterm>
		</para>

		<itemizedlist>
			<listitem><para>NFS</para></listitem>
			<listitem><para>AFS</para></listitem>
			<listitem><para>OpenGFS</para></listitem>
			<listitem><para>Lustre</para></listitem>
		</itemizedlist>

		<para>
<indexterm significance="normal"><primary>server pool</primary></indexterm>
		The server pool (cluster) can use any distributed file system backend if all SMB
		semantics are performed within this pool.
		</para>

		</sect3>

		<sect3>
		<title>Restrictive Constraints on Distributed File Systems</title>

		<para>
<indexterm significance="normal"><primary>SMB services</primary></indexterm>
<indexterm significance="normal"><primary>oplock handling</primary></indexterm>
<indexterm significance="normal"><primary>server pool</primary></indexterm>
<indexterm significance="normal"><primary>backend file system pool</primary></indexterm>
		Where a clustered server provides purely SMB services, oplock handling
		may be done within the server pool without imposing a need for this to
		be passed to the backend file system pool.
		</para>

		<para>
<indexterm significance="normal"><primary>NFS</primary></indexterm>
<indexterm significance="normal"><primary>interoperability</primary></indexterm>
		On the other hand, where the server pool also provides NFS or other file services,
		it will be essential that the implementation be oplock-aware so it can
		interoperate with SMB services. This is a significant challenge today. A failure
		to provide this interoperability will result in a significant loss of performance that will be
		sorely noted by users of Microsoft Windows clients.
		</para>

		<para>
		Last, all state information must be shared across the server pool.
		</para>

		</sect3>

		<sect3>
		<title>Server Pool Communications</title>

		<para>
<indexterm significance="normal"><primary>POSIX semantics</primary></indexterm>
<indexterm significance="normal"><primary>SMB</primary></indexterm>
<indexterm significance="normal"><primary>POSIX locks</primary></indexterm>
<indexterm significance="normal"><primary>SMB locks</primary></indexterm>
		Most backend file systems support POSIX file semantics. This makes it difficult
		to push SMB semantics back into the file system. POSIX locks have different properties
		and semantics from SMB locks.
		</para>

		<para>
<indexterm significance="normal"><primary>smbd</primary></indexterm>
<indexterm significance="normal"><primary>tdb</primary></indexterm>
<indexterm significance="normal"><primary>Clustered smbds</primary></indexterm>
		All <literal>smbd</literal> processes in the server pool must of necessity communicate
		very quickly. For this, the current <parameter moreinfo="none">tdb</parameter> file structure that Samba
		uses is not suitable for use across a network. Clustered <literal>smbd</literal>s must use something else.
		</para>

		</sect3>

		<sect3>
		<title>Server Pool Communications Demands</title>

		<para>
		High-speed interserver communications in the server pool is a design prerequisite
		for a fully functional system. Possibilities for this include:
		</para>

		<itemizedlist>
<indexterm significance="normal"><primary>Myrinet</primary></indexterm>
<indexterm significance="normal"><primary>scalable coherent interface</primary><see>SCI</see></indexterm>
			<listitem><para>
			Proprietary shared memory bus (example: Myrinet or SCI [scalable coherent interface]).
			These are high-cost items.
			</para></listitem>
		
			<listitem><para>
			Gigabit Ethernet (now quite affordable).
			</para></listitem>
		
			<listitem><para>
			Raw Ethernet framing (to bypass TCP and UDP overheads).
			</para></listitem>
		</itemizedlist>

		<para>
		We have yet to identify metrics for  performance demands to enable this to happen
		effectively.
		</para>

		</sect3>

		<sect3>
		<title>Required Modifications to Samba</title>

		<para>
		Samba needs to be significantly modified to work with a high-speed server interconnect
		system to permit transparent failover clustering.
		</para>

		<para>
		Particular functions inside Samba that will be affected include:
		</para>

		<itemizedlist>
			<listitem><para>
			The locking database, oplock notifications,
			and the share mode database.
			</para></listitem>

			<listitem><para>
<indexterm significance="normal"><primary>failure semantics</primary></indexterm>
<indexterm significance="normal"><primary>oplock messages</primary></indexterm>
			Failure semantics need to be defined. Samba behaves the same way as Windows.
			When oplock messages fail, a file open request is allowed, but this is 
			potentially dangerous in a clustered environment. So how should interserver
			pool failure semantics function, and how should such functionality be implemented?
			</para></listitem>

			<listitem><para>
			Should this be implemented using a point-to-point lock manager, or can this
			be done using multicast techniques?
			</para></listitem>

		</itemizedlist>

		</sect3>
	</sect2>

	<sect2>
	<title>A Simple Solution</title>

	<para>
<indexterm significance="normal"><primary>failover servers</primary></indexterm>
<indexterm significance="normal"><primary>exported file system</primary></indexterm>
<indexterm significance="normal"><primary>distributed locking protocol</primary></indexterm>
	Allowing failover servers to handle different functions within the exported file system
	removes the problem of requiring a distributed locking protocol.
	</para>

	<para>
<indexterm significance="normal"><primary>high-speed server interconnect</primary></indexterm>
<indexterm significance="normal"><primary>complex file name space</primary></indexterm>
	If only one server is active in a pair, the need for high-speed server interconnect is avoided.
	This allows the use of existing high-availability solutions, instead of inventing a new one.
	This simpler solution comes at a price <?latex --- ?> the cost of which is the need to manage a more
	complex file name space. Since there is now not a single file system, administrators
	must remember where all services are located <?latex --- ?> a complexity not easily dealt with.
	</para>

	<para>
<indexterm significance="normal"><primary>virtual server</primary></indexterm>
	The <emphasis>virtual server</emphasis> is still needed to redirect requests to backend
	servers. Backend file space integrity is the responsibility of the administrator.
	</para>

	</sect2>

	<sect2>
	<title>High-Availability Server Products</title>

	<para>
<indexterm significance="normal"><primary>resource failover</primary></indexterm>
<indexterm significance="normal"><primary>high-availability services</primary></indexterm>
<indexterm significance="normal"><primary>dedicated heartbeat</primary></indexterm>
<indexterm significance="normal"><primary>LAN</primary></indexterm>
<indexterm significance="normal"><primary>failover process</primary></indexterm>
	Failover servers must communicate in order to handle resource failover. This is essential
	for high-availability services. The use of a dedicated heartbeat is a common technique to
	introduce some intelligence into the failover process. This is often done over a dedicated
	link (LAN or serial).
	</para>

	<para>
<indexterm significance="normal"><primary>SCSI</primary></indexterm>
<indexterm significance="normal"><primary>Red Hat Cluster Manager</primary></indexterm>
<indexterm significance="normal"><primary>Microsoft Wolfpack</primary></indexterm>
<indexterm significance="normal"><primary>Fiber Channel</primary></indexterm>
<indexterm significance="normal"><primary>failover communication</primary></indexterm>
	Many failover solutions (like Red Hat Cluster Manager and Microsoft Wolfpack)
	can use a shared SCSI of Fiber Channel disk storage array for failover communication.
	Information regarding Red Hat high availability solutions for Samba may be obtained from
	<ulink url="http://www.redhat.com/docs/manuals/enterprise/RHEL-AS-2.1-Manual/cluster-manager/s1-service-samba.html">www.redhat.com</ulink>.
	</para>

	<para>
<indexterm significance="normal"><primary>Linux High Availability project</primary></indexterm>
	The Linux High Availability project is a resource worthy of consultation if your desire is
	to build a highly available Samba file server solution. Please consult the home page at
	<ulink url="http://www.linux-ha.org/">www.linux-ha.org/</ulink>.
	</para>

	<para>
<indexterm significance="normal"><primary>backend failures</primary></indexterm>
<indexterm significance="normal"><primary>continuity of service</primary></indexterm>
	Front-end server complexity remains a challenge for high availability because it must deal
	gracefully with backend failures, while at the same time providing continuity of service
	to all network clients.
	</para>
	
	</sect2>

	<sect2>
	<title>MS-DFS: The Poor Man's Cluster</title>

	<para>
<indexterm significance="normal"><primary>MS-DFS</primary></indexterm>
<indexterm significance="normal"><primary>DFS</primary><see>MS-DFS, Distributed File Systems</see></indexterm>
	MS-DFS links can be used to redirect clients to disparate backend servers. This pushes
	complexity back to the network client, something already included by Microsoft.
	MS-DFS creates the illusion of a simple, continuous file system name space that works even
	at the file level.
	</para>

	<para>
	Above all, at the cost of complexity of management, a distributed system (pseudo-cluster) can
	be created using existing Samba functionality.
	</para>

	</sect2>

	<sect2>
	<title>Conclusions</title>

	<itemizedlist>
		<listitem><para>Transparent SMB clustering is hard to do!</para></listitem>
		<listitem><para>Client failover is the best we can do today.</para></listitem>
		<listitem><para>Much more work is needed before a practical and manageable high-availability transparent cluster solution will be possible.</para></listitem>
		<listitem><para>MS-DFS can be used to create the illusion of a single transparent cluster.</para></listitem>
	</itemizedlist>

	</sect2>

</sect1>
</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="largefile">
<chapterinfo>
	
<author>
	<firstname>Jeremy</firstname><surname>Allison</surname>
	<affiliation>
		<orgname>Samba Team</orgname>
		<address format="linespecific"><email>jra@samba.org</email></address>
	</affiliation>
</author>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
	<pubdate>March 5, 2005</pubdate>
</chapterinfo>
<title>Handling Large Directories</title>

<para>
<indexterm significance="normal"><primary>performance degradation</primary></indexterm>
<indexterm significance="normal"><primary>large numbers of files</primary></indexterm>
<indexterm significance="normal"><primary>large directory</primary></indexterm>
Samba-3.0.12 and later implements a solution for sites that have experienced performance degradation due to the
problem of using Samba-3 with applications that need large numbers of files (100,000 or more) per directory.
</para>

<para>
<indexterm significance="normal"><primary>read directory into memory</primary></indexterm>
<indexterm significance="normal"><primary>strange delete semantics</primary></indexterm>
The key was fixing the directory handling to read only the current list requested instead of the old
(up to samba-3.0.11) behavior of reading the entire directory into memory before doling out names.
Normally this would have broken OS/2 applications, which have very strange delete semantics, but by
stealing logic from Samba4 (thanks, Tridge), the current code in 3.0.12 handles this correctly.
</para>

<para>
<indexterm significance="normal"><primary>large directory</primary></indexterm>
<indexterm significance="normal"><primary>performance</primary></indexterm>
To set up an application that needs large numbers of files per directory in a way that does not
damage performance unduly, follow these steps:
</para>

<para>
<indexterm significance="normal"><primary>canonicalize files</primary></indexterm>
First, you need to canonicalize all the files in the directory to have one case, upper or lower <?latex --- ?> take your
pick (I chose upper because all my files were already uppercase names). Then set up a new custom share for the
application as follows:
<simplelist>
<member> </member><member><parameter>[bigshare]</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /data/manyfilesdir</parameter></member>
<member><indexterm>read only</indexterm><parameter>read only = no</parameter></member>
<member><indexterm>case sensitive</indexterm><parameter>case sensitive = True</parameter></member>
<member><indexterm>default case</indexterm><parameter>default case = upper</parameter></member>
<member><indexterm>preserve case</indexterm><parameter>preserve case = no</parameter></member>
<member><indexterm>short preserve case</indexterm><parameter>short preserve case = no</parameter></member>
</simplelist>
</para>

<para>
<indexterm significance="normal"><primary>case options</primary></indexterm>
<indexterm significance="normal"><primary>match case</primary></indexterm>
<indexterm significance="normal"><primary>uppercase</primary></indexterm>
Of course, use your own path and settings, but set the case options to match the case of all the files in your
directory. The path should point at the large directory needed for the application <?latex --- ?> any new files created in
there and in any paths under it will be forced by smbd into uppercase, but smbd will no longer have to scan
the directory for names: it knows that if a file does not exist in uppercase, then it doesn't exist at all.
</para>

<para>
<indexterm significance="normal"><primary>case-insensitive</primary></indexterm>
<indexterm significance="normal"><primary>consistent case</primary></indexterm>
<indexterm significance="normal"><primary>smbd</primary></indexterm>
The secret to this is really in the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="CASESENSITIVE" xlink:href="smb.conf.5.html#CASESENSITIVE">case sensitive = True</link>
line. This tells smbd never to scan for case-insensitive versions of names. So if an application asks for a file
called <filename moreinfo="none">FOO</filename>, and it cannot be found by a simple stat call, then smbd will return file not
found immediately without scanning the containing directory for a version of a different case. The other
<filename moreinfo="none">xxx case xxx</filename> lines make this work by forcing a consistent case on all files created by
<application moreinfo="none">smbd</application>.
</para>

<para>
<indexterm significance="normal"><primary>uppercase</primary></indexterm>
<indexterm significance="normal"><primary>stanza</primary></indexterm>
<indexterm significance="normal"><primary>lowercase filenames</primary></indexterm>
Remember, all files and directories under the <parameter moreinfo="none">path</parameter> directory must be in uppercase
with this <filename moreinfo="none">smb.conf</filename> stanza because <application moreinfo="none">smbd</application> will not be able to find lowercase filenames with these settings. Also
note that this is done on a per-share basis, allowing this parameter to be set only for a share servicing an application with
this problematic behavior (using large numbers of entries in a directory) <?latex --- ?> the rest of your <application moreinfo="none">smbd</application> shares
don't need to be affected.
</para>

<para>
This makes smbd much faster when dealing with large directories.  My test case has over 100,000 files, and
smbd now deals with this very efficiently.
</para>

</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="cfgsmarts">
<chapterinfo>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
	<pubdate>June 30, 2005</pubdate>
</chapterinfo>
<title>Advanced Configuration Techniques</title>

<para>
<indexterm significance="normal"><primary>configuration techniques</primary></indexterm>
<indexterm significance="normal"><primary>include</primary></indexterm>
Since the release of the first edition of this book there have been repeated requests to better document
configuration techniques that may help a network administrator to get more out of Samba. Some users have asked
for documentation regarding the use of the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="INCLUDE" xlink:href="smb.conf.5.html#INCLUDE">include = file-name</link> parameter.
</para>

<para>
<indexterm significance="normal"><primary>multiple servers</primary></indexterm>
<indexterm significance="normal"><primary>multiple server personalities</primary></indexterm>
Commencing around mid-2004 there has been increasing interest in the ability to host multiple Samba servers on
one machine. There has also been an interest in the hosting of multiple Samba server personalities on one
server.
</para>

<para>
<indexterm significance="normal"><primary>technical reviewers</primary></indexterm>
<indexterm significance="normal"><primary>reviewers</primary></indexterm>
Feedback from technical reviewers made the inclusion of this chapter a necessity. So, here is an 
answer the questions that have to date not been adequately addressed. Additional user input is welcome as
it will help this chapter to mature. What is presented here is just a small beginning.
</para>

<para>
<indexterm significance="normal"><primary>multiple servers</primary></indexterm>
<indexterm significance="normal"><primary>multiple hosting</primary></indexterm>
<indexterm significance="normal"><primary>domain controllers</primary></indexterm>
There are a number of ways in which multiple servers can be hosted on a single Samba server. Multiple server
hosting makes it possible to host multiple domain controllers on one machine. Each such machine is
independent, and each can be stopped or started without affecting another.
</para>

<para>
<indexterm significance="normal"><primary>multiple servers</primary></indexterm>
<indexterm significance="normal"><primary>DMS</primary></indexterm>
<indexterm significance="normal"><primary>anonymous server</primary></indexterm>
Sometimes it is desirable to host multiple servers, each with its own security mode. For example, a single
UNIX/Linux host may be a domain member server (DMS) as well as a generic anonymous print server. In this case,
only domain member machines and domain users can access the DMS, but even guest users can access the generic
print server. Another example of a situation where it may be beneficial to host a generic (anonymous) server
is to host a CDROM server.
</para>

<para>
<indexterm significance="normal"><primary>separate servers</primary></indexterm>
<indexterm significance="normal"><primary/></indexterm>
Some environments dictate the need to have separate servers, each with their own resources, each of which are
accessible only by certain users or groups. This is one of the simple, but highly effective, ways that Samba
can replace many physical Windows servers in one Samba installation.
</para>

<sect1>
<title>Implementation</title>

<para>
</para>

<sect2>
<title>Multiple Server Hosting</title>

<para>
<indexterm significance="normal"><primary>multiple server hosting</primary></indexterm>
<indexterm significance="normal"><primary>separate instances</primary></indexterm>
<indexterm significance="normal"><primary>nmbd</primary></indexterm>
<indexterm significance="normal"><primary>smbd</primary></indexterm>
<indexterm significance="normal"><primary>winbindd</primary></indexterm>
<indexterm significance="normal"><primary>recompiling</primary></indexterm>
<indexterm significance="normal"><primary>TDB</primary></indexterm>
The use of multiple server hosting involves running multiple separate instances of Samba, each with it's own
configuration file. This method is complicated by the fact that each instance of <application moreinfo="none">nmbd</application>, <application moreinfo="none">smbd</application> and <application moreinfo="none">winbindd</application>
must have write access to entirely separate TDB files. The ability to keep separate the TDB files used by
<application moreinfo="none">nmbd</application>, <application moreinfo="none">smbd</application> and <application moreinfo="none">winbindd</application> can be enabled either by recompiling Samba for each server hosted so each has its
own default TDB directories, or by configuring these in the <filename moreinfo="none">smb.conf</filename> file, in which case each instance of
<application moreinfo="none">nmbd</application>, <application moreinfo="none">smbd</application> and <application moreinfo="none">winbindd</application> must be told to start up with its own <filename moreinfo="none">smb.conf</filename> configuration file.
</para>

<para>
<indexterm significance="normal"><primary>independent</primary></indexterm>
<indexterm significance="normal"><primary>listen own socket</primary></indexterm>
<indexterm significance="normal"><primary>socket</primary></indexterm>
<indexterm significance="normal"><primary>SID</primary></indexterm>
Each instance should operate on its own IP address (that independent IP address can be an IP Alias).
Each instance of <application moreinfo="none">nmbd</application>, <application moreinfo="none">smbd</application> and <application moreinfo="none">winbindd</application> should listen only on its own IP socket. This can be secured
using the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SOCKETADDRESS" xlink:href="smb.conf.5.html#SOCKETADDRESS">socket address</link> parameter. Each instance of the Samba server will have its
own SID also, this means that the servers are discrete and independent of each other.
</para>

<para>
<indexterm significance="normal"><primary>multiple server hosting</primary></indexterm>
<indexterm significance="normal"><primary>private dir</primary></indexterm>
<indexterm significance="normal"><primary>pid directory</primary></indexterm>
<indexterm significance="normal"><primary>lock directory</primary></indexterm>
<indexterm significance="normal"><primary>interfaces</primary></indexterm>
<indexterm significance="normal"><primary>bind interfaces only</primary></indexterm>
<indexterm significance="normal"><primary>netbios name</primary></indexterm>
<indexterm significance="normal"><primary>workgroup</primary></indexterm>
<indexterm significance="normal"><primary>socket address</primary></indexterm>
The user of multiple server hosting is non-trivial, and requires careful configuration of each aspect of
process management and start up. The <filename moreinfo="none">smb.conf</filename> parameters that must be carefully configured includes:
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PRIVATEDIR" xlink:href="smb.conf.5.html#PRIVATEDIR">private dir</link>, <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PIDDIRECTORY" xlink:href="smb.conf.5.html#PIDDIRECTORY">pid directory</link>,<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOCKDIRECTORY" xlink:href="smb.conf.5.html#LOCKDIRECTORY">lock directory</link>, <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="INTERFACES" xlink:href="smb.conf.5.html#INTERFACES">interfaces</link>, <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="BINDINTERFACESONLY" xlink:href="smb.conf.5.html#BINDINTERFACESONLY">bind interfaces only</link>, <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="NETBIOSNAME" xlink:href="smb.conf.5.html#NETBIOSNAME">netbios name</link>, <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="WORKGROUP" xlink:href="smb.conf.5.html#WORKGROUP">workgroup</link>, <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SOCKETADDRESS" xlink:href="smb.conf.5.html#SOCKETADDRESS">socket address</link>.
</para>

<para>
<indexterm significance="normal"><primary>multiple servers</primary></indexterm>
<indexterm significance="normal"><primary>contribute</primary></indexterm>
<indexterm significance="normal"><primary>comprehensive documentation</primary></indexterm>
Those who elect to create multiple Samba servers should have the ability to read and follow
the Samba source code, and to modify it as needed. This mode of deployment is considered beyond the scope of
this book. However, if someone will contribute more comprehensive documentation we will gladly review it, and
if it is suitable extend this section of this chapter. Until such documentation becomes available the hosting
of multiple samba servers on a single host is considered not supported for Samba-3 by the Samba Team.
</para>

</sect2>

<sect2>
<title>Multiple Virtual Server Personalities</title>

<para>
<indexterm significance="normal"><primary>multiple virtual servers</primary></indexterm>
<indexterm significance="normal"><primary>netbios alias</primary></indexterm>
<indexterm significance="normal"><primary>meta-services</primary></indexterm>
Samba has the ability to host multiple virtual servers, each of which have their own personality.  This is
achieved by configuring an <filename moreinfo="none">smb.conf</filename> file that is common to all personalities hosted.  Each server
personality is hosted using its own <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="NETBIOSALIAS" xlink:href="smb.conf.5.html#NETBIOSALIAS">netbios alias</link> name, and each has its own distinct
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="[GLOBAL]" xlink:href="smb.conf.5.html#[GLOBAL]">[global]</link> section. Each server may have its own stanzas for services and meta-services.
</para>

<para>
<indexterm significance="normal"><primary>workgroup</primary></indexterm>
<indexterm significance="normal"><primary>security</primary></indexterm>
<indexterm significance="normal"><primary>netbios aliases</primary></indexterm>
When hosting multiple virtual servers, each with their own personality, each can be in a different workgroup.
Only the primary server can be a domain member or a domain controller. The personality is defined by the
combination of the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SECURITY" xlink:href="smb.conf.5.html#SECURITY">security</link> mode it is operating in, the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="NETBIOSALIASES" xlink:href="smb.conf.5.html#NETBIOSALIASES">netbios aliases</link> it has, and the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="WORKGROUP" xlink:href="smb.conf.5.html#WORKGROUP">workgroup</link> that is defined for it.
</para>

<para>
<indexterm significance="normal"><primary>NetBIOS name</primary></indexterm>
<indexterm significance="normal"><primary>NetBIOS-less SMB</primary></indexterm>
<indexterm significance="normal"><primary>smb ports</primary></indexterm>
<indexterm significance="normal"><primary>TCP port 139</primary></indexterm>
<indexterm significance="normal"><primary>TCP port 445</primary></indexterm>
<indexterm significance="normal"><primary>%L</primary></indexterm>
This configuration style can be used either with NetBIOS names, or using NetBIOS-less SMB over TCP services.
If run using NetBIOS mode (the most common method) it is important that the parameter <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SMBPORTS" xlink:href="smb.conf.5.html#SMBPORTS">smb ports = 139</link> should be specified in the primary <filename moreinfo="none">smb.conf</filename> file. Failure to do this will result
in Samba operating over TCP port 445 and problematic operation at best, and at worst only being able to obtain
the functionality that is specified in the primary <filename moreinfo="none">smb.conf</filename> file. The use of NetBIOS over TCP/IP using only
TCP port 139 means that the use of the <literal moreinfo="none">%L</literal> macro is fully enabled. If the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SMBPORTS" xlink:href="smb.conf.5.html#SMBPORTS">smb ports = 139</link> is not specified (the default is <parameter moreinfo="none">445 139</parameter>, or if
the value of this parameter is set at <parameter moreinfo="none">139 445</parameter> then the <literal moreinfo="none">%L</literal> macro
is not serviceable.
</para>

<para>
<indexterm significance="normal"><primary>host multiple servers</primary></indexterm>
<indexterm significance="normal"><primary>multiple personality</primary></indexterm>
<indexterm significance="normal"><primary>NetBIOS-less</primary></indexterm>
<indexterm significance="normal"><primary>%i macro</primary></indexterm>
It is possible to host multiple servers, each with their own personality, using port 445 (the NetBIOS-less SMB
port), in which case the <literal moreinfo="none">%i</literal> macro can be used to provide separate server identities (by
IP Address). Each can have its own <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SECURITY" xlink:href="smb.conf.5.html#SECURITY">security</link> mode. It will be necessary to use the
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="INTERFACES" xlink:href="smb.conf.5.html#INTERFACES">interfaces</link>, <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="BINDINTERFACESONLY" xlink:href="smb.conf.5.html#BINDINTERFACESONLY">bind interfaces only</link> and IP aliases in addition to
the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="NETBIOSNAME" xlink:href="smb.conf.5.html#NETBIOSNAME">netbios name</link> parameters to create the virtual servers. This method is considerably
more complex than that using NetBIOS names only using TCP port 139.
</para>

<para>
<indexterm significance="normal"><primary>anonymous file server</primary></indexterm>
Consider an example environment that consists of a standalone, user-mode security Samba server and a read-only
Windows 95 file server that has to be replaced. Instead of replacing the Windows 95 machine with a new PC, it
is possible to add this server as a read-only anonymous file server that is hosted on the Samba server. Here
are some parameters:
</para>

<para>
The Samba server is called <literal moreinfo="none">ELASTIC</literal>, its workgroup name is <literal moreinfo="none">ROBINSNEST</literal>.
The CDROM server is called <literal moreinfo="none">CDSERVER</literal> and its workgroup is <literal moreinfo="none">ARTSDEPT</literal>. A
possible implementation is shown here:
</para>

<para>
<indexterm significance="normal"><primary>/etc/samba</primary></indexterm>
<indexterm significance="normal"><primary>nmbd</primary></indexterm>
<indexterm significance="normal"><primary>smbd</primary></indexterm>
<indexterm significance="normal"><primary>smb.conf</primary></indexterm>
The <filename moreinfo="none">smb.conf</filename> file for the master server is shown in <link linkend="elastic">Elastic smb.conf File</link>.
This file is placed in the <filename moreinfo="none">/etc/samba</filename> directory. Only the <application moreinfo="none">nmbd</application> and the <application moreinfo="none">smbd</application> daemons
are needed. When started the server will appear in Windows Network Neighborhood as the machine
<literal moreinfo="none">ELASTIC</literal> under the workgroup <literal moreinfo="none">ROBINSNEST</literal>. It is helpful if the Windows
clients that must access this server are also in the workgroup <literal moreinfo="none">ROBINSNEST</literal> as this will make
browsing much more reliable.
</para>

<example id="elastic">
<title>Elastic smb.conf File</title>
<simplelist>
<member># Global parameters</member>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>workgroup</indexterm><parameter>workgroup = ROBINSNEST</parameter></member>
<member><indexterm>netbios name</indexterm><parameter>netbios name = ELASTIC</parameter></member>
<member><indexterm>netbios aliases</indexterm><parameter>netbios aliases = CDSERVER</parameter></member>
<member><indexterm>smb ports</indexterm><parameter>smb ports = 139</parameter></member>
<member><indexterm>printcap name</indexterm><parameter>printcap name = cups</parameter></member>
<member><indexterm>disable spoolss</indexterm><parameter>disable spoolss = Yes</parameter></member>
<member><indexterm>show add printer wizard</indexterm><parameter>show add printer wizard = No</parameter></member>
<member><indexterm>printing</indexterm><parameter>printing = cups</parameter></member>
<member><indexterm>include</indexterm><parameter>include = /etc/samba/smb-%L.conf</parameter></member>

<member> </member><member><parameter>[homes]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = Home Directories</parameter></member>
<member><indexterm>valid users</indexterm><parameter>valid users = %S</parameter></member>
<member><indexterm>read only</indexterm><parameter>read only = No</parameter></member>
<member><indexterm>browseable</indexterm><parameter>browseable = No</parameter></member>

<member> </member><member><parameter>[office]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = Data</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /data</parameter></member>
<member><indexterm>read only</indexterm><parameter>read only = No</parameter></member>

<member> </member><member><parameter>[printers]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = All Printers</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /var/spool/samba</parameter></member>
<member><indexterm>create mask</indexterm><parameter>create mask = 0600</parameter></member>
<member><indexterm>guest ok</indexterm><parameter>guest ok = Yes</parameter></member>
<member><indexterm>printable</indexterm><parameter>printable = Yes</parameter></member>
<member><indexterm>use client driver</indexterm><parameter>use client driver = Yes</parameter></member>
<member><indexterm>browseable</indexterm><parameter>browseable = No</parameter></member>
</simplelist>
</example>

<para>
<indexterm significance="normal"><primary>smb-cdserver.conf</primary></indexterm>
The configuration file for the CDROM server is listed in <link linkend="cdserver">CDROM Server
smb-cdserver.conf file</link>. This file is called <filename moreinfo="none">smb-cdserver.conf</filename> and it should be
located in the <filename moreinfo="none">/etc/samba</filename> directory. Machines that are in the workgroup
<literal moreinfo="none">ARTSDEPT</literal> will be able to browse this server freely.
</para>

<example id="cdserver">
<title>CDROM Server smb-cdserver.conf file</title>
<simplelist>
<member># Global parameters</member>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>workgroup</indexterm><parameter>workgroup = ARTSDEPT</parameter></member>
<member><indexterm>netbios name</indexterm><parameter>netbios name = CDSERVER</parameter></member>
<member><indexterm>map to guest</indexterm><parameter>map to guest = Bad User</parameter></member>
<member><indexterm>guest ok</indexterm><parameter>guest ok = Yes</parameter></member>

<member> </member><member><parameter>[carousel]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = CDROM Share</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /export/cddata</parameter></member>
<member><indexterm>read only</indexterm><parameter>read only = Yes</parameter></member>
<member><indexterm>guest ok</indexterm><parameter>guest ok = Yes</parameter></member>
</simplelist>
</example>

<para>
<indexterm significance="normal"><primary>different resources</primary></indexterm>
<indexterm significance="normal"><primary>separate workgroups</primary></indexterm>
<indexterm significance="normal"><primary>read-only access</primary></indexterm>
<indexterm significance="normal"><primary>nobody account</primary></indexterm>
The two servers have different resources and are in separate workgroups. The server <literal moreinfo="none">ELASTIC</literal>
can only be accessed by uses who have an appropriate account on the host server. All users will be able to
access the CDROM data that is stored in the <filename moreinfo="none">/export/cddata</filename> directory. File system
permissions should set so that the <literal moreinfo="none">others</literal> user has read-only access to the directory and its
contents. The files can be owned by root (any user other than the nobody account).
</para>

</sect2>

<sect2>
<title>Multiple Virtual Server Hosting</title>

<para>
<indexterm significance="normal"><primary>primary domain controller</primary></indexterm>
<indexterm significance="normal"><primary>extra machine</primary></indexterm>
<indexterm significance="normal"><primary>same domain/workgroup</primary></indexterm>
In this example, the requirement is for a primary domain controller for the domain called
<literal moreinfo="none">MIDEARTH</literal>. The PDC will be called <literal moreinfo="none">MERLIN</literal>. An extra machine called
<literal moreinfo="none">SAURON</literal> is required. Each machine will have only its own shares. Both machines belong to the
same domain/workgroup.
</para>

<para>
<indexterm significance="normal"><primary>master smb.conf</primary></indexterm>
<indexterm significance="normal"><primary>/etc/samba</primary></indexterm>
<indexterm significance="normal"><primary/></indexterm>
The master <filename moreinfo="none">smb.conf</filename> file is shown in <link linkend="mastersmbc">the Master smb.conf File Global Section</link>.
The two files that specify the share information for each server are shown in <link linkend="merlinsmbc">the
smb-merlin.conf File Share Section</link>, and <link linkend="sauronsmbc">the smb-sauron.conf File Share
Section</link>. All three files are locate in the <filename moreinfo="none">/etc/samba</filename> directory.
</para>

<example id="mastersmbc">
<title>Master smb.conf File Global Section</title>
<simplelist>
<member># Global parameters</member>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>workgroup</indexterm><parameter>workgroup = MIDEARTH</parameter></member>
<member><indexterm>netbios name</indexterm><parameter>netbios name = MERLIN</parameter></member>
<member><indexterm>netbios aliases</indexterm><parameter>netbios aliases = SAURON</parameter></member>
<member><indexterm>passdb backend</indexterm><parameter>passdb backend = tdbsam</parameter></member>
<member><indexterm>smb ports</indexterm><parameter>smb ports = 139</parameter></member>
<member><indexterm>syslog</indexterm><parameter>syslog = 0</parameter></member>
<member><indexterm>printcap name</indexterm><parameter>printcap name = CUPS</parameter></member>
<member><indexterm>show add printer wizard</indexterm><parameter>show add printer wizard = No</parameter></member>
<member><indexterm>add user script</indexterm><parameter>add user script = /usr/sbin/useradd -m '%u'</parameter></member>
<member><indexterm>delete user script</indexterm><parameter>delete user script = /usr/sbin/userdel -r '%u'</parameter></member>
<member><indexterm>add group script</indexterm><parameter>add group script = /usr/sbin/groupadd '%g'</parameter></member>
<member><indexterm>delete group script</indexterm><parameter>delete group script = /usr/sbin/groupdel '%g'</parameter></member>
<member><indexterm>add user to group script</indexterm><parameter>add user to group script = /usr/sbin/usermod -G '%g' '%u'</parameter></member>
<member><indexterm>add machine script</indexterm><parameter>add machine script = /usr/sbin/useradd -s /bin/false -d /var/lib/nobody '%u'</parameter></member>
<member><indexterm>logon script</indexterm><parameter>logon script = scripts\login.bat</parameter></member>
<member><indexterm>logon path</indexterm><parameter>logon path =  </parameter></member>
<member><indexterm>logon drive</indexterm><parameter>logon drive = X:</parameter></member>
<member><indexterm>domain logons</indexterm><parameter>domain logons = Yes</parameter></member>
<member><indexterm>preferred master</indexterm><parameter>preferred master = Yes</parameter></member>
<member><indexterm>wins support</indexterm><parameter>wins support = Yes</parameter></member>
<member><indexterm>printing</indexterm><parameter>printing = CUPS</parameter></member>
<member><indexterm>include</indexterm><parameter>include = /etc/samba/smb-%L.conf</parameter></member>
</simplelist>
</example>

<example id="merlinsmbc">
<title>MERLIN smb-merlin.conf File Share Section</title>
<simplelist>
<member># Global parameters</member>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>workgroup</indexterm><parameter>workgroup = MIDEARTH</parameter></member>
<member><indexterm>netbios name</indexterm><parameter>netbios name = MERLIN</parameter></member>

<member> </member><member><parameter>[homes]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = Home Directories</parameter></member>
<member><indexterm>valid users</indexterm><parameter>valid users = %S</parameter></member>
<member><indexterm>read only</indexterm><parameter>read only = No</parameter></member>
<member><indexterm>browseable</indexterm><parameter>browseable = No</parameter></member>

<member> </member><member><parameter>[office]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = Data</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /data</parameter></member>
<member><indexterm>read only</indexterm><parameter>read only = No</parameter></member>

<member> </member><member><parameter>[netlogon]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = NETLOGON</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /var/lib/samba/netlogon</parameter></member>
<member><indexterm>read only</indexterm><parameter>read only = Yes</parameter></member>
<member><indexterm>browseable</indexterm><parameter>browseable = No</parameter></member>

<member> </member><member><parameter>[printers]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = All Printers</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /var/spool/samba</parameter></member>
<member><indexterm>printable</indexterm><parameter>printable = Yes</parameter></member>
<member><indexterm>use client driver</indexterm><parameter>use client driver = Yes</parameter></member>
<member><indexterm>browseable</indexterm><parameter>browseable = No</parameter></member>
</simplelist>
</example>

<example id="sauronsmbc">
<title>SAURON smb-sauron.conf File Share Section</title>
<simplelist>
<member># Global parameters</member>
<member> </member><member><parameter>[global]</parameter></member>
<member><indexterm>workgroup</indexterm><parameter>workgroup = MIDEARTH</parameter></member>
<member><indexterm>netbios name</indexterm><parameter>netbios name = SAURON</parameter></member>

<member> </member><member><parameter>[www]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = Web Pages</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /srv/www/htdocs</parameter></member>
<member><indexterm>read only</indexterm><parameter>read only = No</parameter></member>
</simplelist>
</example>

</sect2>

</sect1>

</chapter>
	<?latex \cleardoublepage ?>

</part>

<part id="migration">
<title>Migration and Updating</title>

	<?latex \cleardoublepage ?>
	<chapter id="upgrading-to-3.0">
<chapterinfo>
	<author>
<firstname>Jelmer</firstname><othername>R.</othername><surname>Vernooij</surname><othername>R.</othername>
<affiliation>
	<orgname>The Samba Team</orgname>
	<address format="linespecific"><email>jelmer@samba.org</email></address>
</affiliation></author>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
	<author>
<firstname>Gerald</firstname><surname>Carter</surname><othername>(Jerry)</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jerry@samba.org</email></address>
</affiliation></author>
	<pubdate>August 16, 2007</pubdate>
</chapterinfo>

<title>Updating and Upgrading Samba</title>
<para>
This chapter provides a detailed record of changes made during the 3.x series releases. At this time this
series consists of the 3.0.x series that is under the GNU GPL version 2 license, and the Samba 3.2.x series
that is being released under the terms of the GNU GPL version 3 license.
</para>

<sect1>
<title>Key Update Requirements</title>
<para>
Samba is a fluid product in which there may be significant changes between releases. Some of these changes are
brought about as a result of changes in the protocols that are used by Microsoft Windows network clients as a
result of security or functionality updates through official Microsoft patches and updates. Samba must track
such changes, particularly where they affect the internal operation of Samba itself.
</para>

<para>
Please refer to any notes below that make explicit mention of the version of Samba you are using. In general,
all changes that apply to a new release will apply to follow-on releases also. For example, changes to Samba
3.0.23 affect all releases up to an including 3.0.25 and later.  Samba 3.2.x was originaly cut from Samba
3.0.25 before 3.2.0-specific changes were applied. Unless a 3.0.x series feature is specifically revoked, the
behavior of the 3.2.x series can be expected to follow the earlier pattern.
</para>

<sect2>
<title>Upgrading from Samba-3.0.x to Samba-3.2.0</title>
<para>
</para>
</sect2>

<sect2 id="oldupdatenotes">
<title>Upgrading from Samba-2.x to Samba-3.0.25</title>
<para>
<indexterm significance="normal"><primary>Samba differences</primary></indexterm>
<indexterm significance="normal"><primary>changed parameters</primary></indexterm>
<indexterm significance="normal"><primary>simple guide</primary></indexterm>
This chapter deals exclusively with the differences between Samba-3.0.25 and Samba-2.2.8a.
It points out where configuration parameters have changed, and provides a simple guide for
the move from 2.2.x to 3.0.25.
</para>
</sect2>

<sect2>
<title>Quick Migration Guide</title>

<para>
Samba-3.0.25 default behavior should be approximately the same as Samba-2.2.x.
The default behavior when the new parameter <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PASSDBBACKEND" xlink:href="smb.conf.5.html#PASSDBBACKEND">passdb backend</link>
is not defined in the <filename moreinfo="none">smb.conf</filename> file provides the same default behavior as Samba-2.2.x
with <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="ENCRYPTPASSWORDS" xlink:href="smb.conf.5.html#ENCRYPTPASSWORDS">encrypt passwords = Yes</link> and
will use the <filename moreinfo="none">smbpasswd</filename> database.
</para>

<para>
<indexterm significance="normal"><primary>behavior approximately same</primary></indexterm>
<indexterm significance="normal"><primary>differing protocol</primary></indexterm>
So why say that <emphasis>behavior should be approximately the same as Samba-2.2.x</emphasis>? Because
Samba-3.0.25 can negotiate new protocols, such as support for native Unicode, that may result in
differing protocol code paths being taken. The new behavior under such circumstances is not
exactly the same as the old one. The good news is that the domain and machine SIDs will be
preserved across the upgrade.
</para>

<para>
<indexterm significance="normal"><primary>LDAP backend</primary></indexterm>
<indexterm significance="normal"><primary>database</primary></indexterm>
<indexterm significance="normal"><primary>pdbedit</primary></indexterm>
<indexterm significance="normal"><primary>Samba-3-compatible LDAP backend</primary></indexterm>
If the Samba-2.2.x system is using an LDAP backend, and there is no time to update the LDAP
database, then make sure that <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PASSDBBACKEND" xlink:href="smb.conf.5.html#PASSDBBACKEND">passdb backend = ldapsam_compat</link>
is specified in the <filename moreinfo="none">smb.conf</filename> file. For the rest, behavior should remain more or less the same.
At a later date, when there is time to implement a new Samba-3-compatible LDAP backend, it is possible
to migrate the old LDAP database to the new one through use of the <literal>pdbedit</literal>.
See <link linkend="pdbeditthing">The <emphasis>pdbedit</emphasis> Command</link>.
</para>

</sect2>
</sect1>

<sect1>
<title>New Featuers in Samba-3.x Series</title>
<para>
</para>

<sect2>
<title>New Features in Samba-3.2.x Series</title>
<para>
</para>
</sect2>

<sect2>
<title>New Features in Samba-3.0.x</title>

<para>
The major new features are:
</para>

<orderedlist numeration="arabic" inheritnum="ignore" continuation="restarts">
	<listitem><para>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
<indexterm significance="normal"><primary>LDAP/Kerberos</primary></indexterm>
	Active Directory support. This release is able to join an ADS realm
	as a member server and authenticate users using LDAP/Kerberos.
	</para></listitem>

	<listitem><para>
<indexterm significance="normal"><primary>Unicode</primary></indexterm>
<indexterm significance="normal"><primary>multibyte character sets</primary></indexterm>
	Unicode support. Samba will now negotiate Unicode on the wire, and
	internally there is a much better infrastructure for multibyte
	and Unicode character sets.
	</para></listitem>

	<listitem><para>
<indexterm significance="normal"><primary>authentication system</primary></indexterm>
	New authentication system. The internal authentication system has
	been almost completely rewritten. Most of the changes are internal,
	but the new authoring system is also very configurable.
	</para></listitem>

	<listitem><para>
<indexterm significance="normal"><primary>filename mangling</primary></indexterm>
	New filename mangling system. The filename mangling system has been
	completely rewritten. An internal database now stores mangling maps
	persistently.
	</para></listitem>

	<listitem><para>
<indexterm significance="normal"><primary>net command</primary></indexterm>
	New <quote>net</quote> command. A new <quote>net</quote> command has been added. It is
	somewhat similar to the <quote>net</quote> command in Windows. Eventually, we
	plan to replace a bunch of other utilities (such as smbpasswd)
	with subcommands in <quote>net</quote>.
	</para></listitem>

	<listitem><para>
<indexterm significance="normal"><primary>status32 codes</primary></indexterm>
	Samba now negotiates NT-style status32 codes on the wire. This
	considerably improves error handling.
	</para></listitem>

	<listitem><para>
<indexterm significance="normal"><primary>printer attributes publishing</primary></indexterm>
	Better Windows 200x/XP printing support, including publishing
	printer attributes in Active Directory.
	</para></listitem>

	<listitem><para>
<indexterm significance="normal"><primary>RPC modules</primary></indexterm>
<indexterm significance="normal"><primary>passdb backends</primary></indexterm>
<indexterm significance="normal"><primary>character sets</primary></indexterm>
	New loadable RPC modules for passdb backends and character sets.
	</para></listitem>

	<listitem><para>
<indexterm significance="normal"><primary>dual-daemon winbindd</primary></indexterm>
	New default dual-daemon winbindd support for better performance.
	</para></listitem>

	<listitem><para>
<indexterm significance="normal"><primary>migrating</primary></indexterm>
<indexterm significance="normal"><primary>maintaining ids</primary></indexterm>
<indexterm significance="normal"><primary>SID</primary></indexterm>
	Support for migrating from a Windows NT 4.0 domain to a Samba
	domain and maintaining user, group, and domain SIDs.
	</para></listitem>

	<listitem><para>
<indexterm significance="normal"><primary>trust relationships</primary></indexterm>
<indexterm significance="normal"><primary>domain controllers</primary></indexterm>
	Support for establishing trust relationships with Windows NT 4.0
	domain controllers.
	</para></listitem>

	<listitem><para>
<indexterm significance="normal"><primary>Winbind architecture</primary></indexterm>
<indexterm significance="normal"><primary>LDAP directory</primary></indexterm>
<indexterm significance="normal"><primary>ID mapping</primary></indexterm>
	Initial support for a distributed Winbind architecture using
	an LDAP directory for storing SID to UID/GID mappings.
	</para></listitem>

	<listitem><para>
	Major updates to the Samba documentation tree.
	</para></listitem>

	<listitem><para>
<indexterm significance="normal"><primary>SMB signing</primary></indexterm>
<indexterm significance="normal"><primary>security settings</primary></indexterm>
	Full support for client and server SMB signing to ensure
	compatibility with default Windows 2003 security settings.
	</para></listitem>
</orderedlist>

<para>
Plus lots of other improvements!
</para>


<sect3>
<title>Configuration Parameter Changes</title>

<para>
This section contains a brief listing of changes to <filename moreinfo="none">smb.conf</filename> options since the Samba-2.2.x series up to and
including Samba-3.0.25.
</para>

<para>
Please refer to the smb.conf(5) man page for complete descriptions of new or modified
parameters.
</para>

<para>
Whenever a Samba update or upgrade is performed it is highly recommended to read the file called
<emphasis>WHATSNEW.txt</emphasis> that is part of the Samba distribution tarball. This file may also
be obtain on-line from the Samba <ulink url="http://www.samba.org/samba/">web site</ulink>, in
the right column, under Current Stable Release, by clicking on <emphasis>Release Notes</emphasis>.
</para>

</sect3>

<sect3>
<title>Removed Parameters</title>

<indexterm significance="normal"><primary>deleted parameters</primary></indexterm>
<para>
In alphabetical order, these are the parameters eliminated from Samba-2.2.x through 3.0.25.
</para>

<itemizedlist>
	<listitem><para>admin log</para></listitem>
	<listitem><para>alternate permissions</para></listitem>
	<listitem><para>character set</para></listitem>
	<listitem><para>client codepage</para></listitem>
	<listitem><para>code page directory</para></listitem>
	<listitem><para>coding system</para></listitem>
	<listitem><para>domain admin group</para></listitem>
	<listitem><para>domain guest group</para></listitem>
	<listitem><para>enable rid algorithm</para></listitem>
	<listitem><para>enable svcctl</para></listitem>
	<listitem><para>force unknown acl user</para></listitem>
	<listitem><para>hosts equiv</para></listitem>
	<listitem><para>ldap filter</para></listitem>
	<listitem><para>min password length</para></listitem>
	<listitem><para>nt smb support</para></listitem>
	<listitem><para>post script</para></listitem>
	<listitem><para>printer admin</para></listitem>
	<listitem><para>printer driver</para></listitem>
	<listitem><para>printer driver file</para></listitem>
	<listitem><para>printer driver location</para></listitem>
	<listitem><para>read size</para></listitem>
	<listitem><para>source environment</para></listitem>
	<listitem><para>status </para></listitem>
	<listitem><para>strip dot </para></listitem>
	<listitem><para>total print jobs</para></listitem>
	<listitem><para>unicode</para></listitem>
	<listitem><para>use rhosts</para></listitem>
	<listitem><para>valid chars</para></listitem>
	<listitem><para>vfs options</para></listitem>
	<listitem><para>winbind enable local accounts</para></listitem>
	<listitem><para>winbind max idle children</para></listitem>
	<listitem><para>wins partners</para></listitem>
</itemizedlist>

</sect3>

<sect3>
<title>New Parameters</title>

<para>The following new parameters have been released up to and including Samba 3.0.25 (grouped by function:)</para>

<para>Remote Management</para>

<indexterm significance="normal"><primary>new parameters</primary></indexterm>

<itemizedlist>
	<listitem><para>abort shutdown script</para></listitem>
	<listitem><para>shutdown script</para></listitem>
</itemizedlist>

<para>User and Group Account Management</para>

<itemizedlist>
	<listitem><para>add group script</para></listitem>
	<listitem><para>add machine script</para></listitem>
	<listitem><para>add user to group script</para></listitem>
	<listitem><para>algorithmic rid base</para></listitem>
	<listitem><para>delete group script</para></listitem>
	<listitem><para>delete user from group script</para></listitem>
	<listitem><para>passdb backend</para></listitem>
	<listitem><para>rename user script</para></listitem>
	<listitem><para>set primary group script</para></listitem>
	<listitem><para>username map script</para></listitem>
</itemizedlist>

<para>Authentication</para>

<itemizedlist>
	<listitem><para>auth methods</para></listitem>
	<listitem><para>ldap password sync</para></listitem>
	<listitem><para>passdb expand explicit</para></listitem>
	<listitem><para>realm</para></listitem>
</itemizedlist>

<para>Protocol Options</para>

<itemizedlist>
	<listitem><para>add port command</para></listitem>
	<listitem><para>afs token lifetime</para></listitem>
	<listitem><para>client lanman auth</para></listitem>
	<listitem><para>client NTLMv2 auth</para></listitem>
	<listitem><para>client schannel</para></listitem>
	<listitem><para>client signing</para></listitem>
	<listitem><para>client use spnego</para></listitem>
	<listitem><para>defer sharing violations</para></listitem>
	<listitem><para>disable netbios</para></listitem>
	<listitem><para>dmapi support</para></listitem>
	<listitem><para>enable privileges</para></listitem>
	<listitem><para>use kerberos keytab</para></listitem>
	<listitem><para>log nt token command</para></listitem>
	<listitem><para>ntlm auth</para></listitem>
	<listitem><para>paranoid server security </para></listitem>
	<listitem><para>sendfile</para></listitem>
	<listitem><para>server schannel</para></listitem>
	<listitem><para>server signing</para></listitem>
	<listitem><para>smb ports</para></listitem>
	<listitem><para>svcctl list</para></listitem>
	<listitem><para>use spnego</para></listitem>
</itemizedlist>

<para>File Service</para>

<itemizedlist>
	<listitem><para>allocation roundup size</para></listitem>
	<listitem><para>acl check permissions</para></listitem>
	<listitem><para>acl group control</para></listitem>
	<listitem><para>acl map full control</para></listitem>
	<listitem><para>aio read size</para></listitem>
	<listitem><para>aio write size</para></listitem>
	<listitem><para>dfree cache time</para></listitem>
	<listitem><para>dfree command</para></listitem>
	<listitem><para>ea support</para></listitem>
	<listitem><para>enable asu support</para></listitem>
	<listitem><para>fam change notify</para></listitem>
	<listitem><para>force unknown acl user</para></listitem>
	<listitem><para>get quota command</para></listitem>
	<listitem><para>hide special files</para></listitem>
	<listitem><para>hide unwriteable files</para></listitem>
	<listitem><para>inherit owner</para></listitem>
	<listitem><para>hostname lookups</para></listitem>
	<listitem><para>kernel change notify</para></listitem>
	<listitem><para>mangle prefix</para></listitem>
	<listitem><para>map acl inherit</para></listitem>
	<listitem><para>map read only</para></listitem>
	<listitem><para>max stat cache size</para></listitem>
	<listitem><para>msdfs proxy</para></listitem>
	<listitem><para>open files database hash size</para></listitem>
	<listitem><para>set quota command</para></listitem>
	<listitem><para>store dos attributes</para></listitem>
	<listitem><para>use sendfile</para></listitem>
	<listitem><para>usershare allow guests</para></listitem>
	<listitem><para>usershare max shares</para></listitem>
	<listitem><para>usershare owner only</para></listitem>
	<listitem><para>usershare path</para></listitem>
	<listitem><para>usershare prefix allow list</para></listitem>
	<listitem><para>usershare prefix deny list</para></listitem>
	<listitem><para>usershare template share</para></listitem>
	<listitem><para>vfs objects</para></listitem>
</itemizedlist>

<para>Printing</para>

<itemizedlist>
	<listitem><para>cups options</para></listitem>
	<listitem><para>cups server</para></listitem>
	<listitem><para>force printername</para></listitem>
	<listitem><para>iprint server</para></listitem>
	<listitem><para>max reported print jobs</para></listitem>
	<listitem><para>printcap cache time</para></listitem>
</itemizedlist> 


<para>Unicode and Character Sets</para>

<itemizedlist>
	<listitem><para>display charset</para></listitem>
	<listitem><para>dos charset</para></listitem>
	<listitem><para>UNIX charset</para></listitem>
</itemizedlist>

<para>SID to UID/GID Mappings</para>

<itemizedlist>
	<listitem><para>idmap backend</para></listitem>
	<listitem><para>idmap gid</para></listitem>
	<listitem><para>idmap uid</para></listitem>
	<listitem><para>username map script</para></listitem>
	<listitem><para>winbind nss info</para></listitem>
	<listitem><para>winbind offline logon</para></listitem>
	<listitem><para>winbind refresh tickets</para></listitem>
	<listitem><para>winbind trusted domains only</para></listitem>
	<listitem><para>template primary group</para></listitem>
</itemizedlist>

<para>LDAP</para>

<itemizedlist>
	<listitem><para>ldap delete dn</para></listitem>
	<listitem><para>ldap group suffix</para></listitem>
	<listitem><para>ldap idmap suffix</para></listitem>
	<listitem><para>ldap machine suffix</para></listitem>
	<listitem><para>ldap passwd sync</para></listitem>
	<listitem><para>ldap replication sleep</para></listitem>
	<listitem><para>ldap timeout</para></listitem>
	<listitem><para>ldap user suffix</para></listitem>
</itemizedlist>

<para>General Configuration</para>

<itemizedlist>
	<listitem><para>eventlog list</para></listitem>
	<listitem><para>preload modules</para></listitem>
	<listitem><para>reset on zero vc</para></listitem>
	<listitem><para>privatedir</para></listitem>
</itemizedlist>

</sect3>

<sect3>
<title>Modified Parameters (Changes in Behavior)</title>

<itemizedlist>
	<listitem><para>acl group control (new default is No, deprecated parameter)</para></listitem>
	<listitem><para>change notify timeout (scope changed)</para></listitem>
	<listitem><para>dos filemode (disabled by default)</para></listitem>
	<listitem><para>dos filetimes (enabled by default)</para></listitem>
	<listitem><para>enable asu support (disabled by default)</para></listitem>
	<listitem><para>enable privileges (enabled by default)</para></listitem>
	<listitem><para>encrypt passwords (enabled by default) </para></listitem>
	<listitem><para>host msdfs (enabled by default)</para></listitem>
	<listitem><para>mangling method (set to hash2 by default) </para></listitem>
	<listitem><para>map to guest</para></listitem>
	<listitem><para>only user (deprecated)</para></listitem>
	<listitem><para>passwd chat</para></listitem>
	<listitem><para>passwd program</para></listitem>
	<listitem><para>password server</para></listitem>
	<listitem><para>restrict anonymous (integer value)</para></listitem>
	<listitem><para>security (new ads value)</para></listitem>
	<listitem><para>strict locking (auto by default)</para></listitem>
	<listitem><para>winbind cache time (increased to 5 minutes)</para></listitem>
	<listitem><para>winbind enum groups (disabled by default)</para></listitem>
	<listitem><para>winbind enum users (disabled by default)</para></listitem>
	<listitem><para>winbind nested groups (enabled by default)</para></listitem>
	<listitem><para>winbind uid (deprecated in favor of idmap uid)</para></listitem>
	<listitem><para>winbind gid (deprecated in favor of idmap gid)</para></listitem>
	<listitem><para>winbindd nss info</para></listitem>
	<listitem><para>write cache (deprecated)</para></listitem>
</itemizedlist>

</sect3>

</sect2>

<sect2>
<title>New Functionality</title>

	<para>
<indexterm significance="normal"><primary>major changes</primary></indexterm>
	The major changes in behavior since that Samba-2.2.x series are documented in this section.
	Please refer to the <filename moreinfo="none">WHATSNEW.txt</filename> file that ships with every release of
	Samba to obtain detailed information regarding the changes that have been made during the
	life of the current Samba release.
	</para>

	<sect3>
	<title>TDB Data Files</title>

<indexterm significance="normal"><primary>tdb data files</primary></indexterm>
	<para>
	Refer to <link linkend="install">Installation, Chapter 1</link>, <link linkend="tdbdocs">Chapter 1</link>
	for information pertaining to the Samba-3 data files, their location and the information that must be
	preserved across server migrations, updates and upgrades.
	</para>

	<para>
<indexterm significance="normal"><primary>tdb file backup</primary></indexterm>
	Please remember to back up your existing ${lock directory}/*tdb before upgrading to Samba-3. If necessary,
	Samba will upgrade databases as they are opened. Downgrading from Samba-3 to 2.2, or reversion to an earlier
	version of Samba-3 from a later release, is an unsupported path.
	</para>

	<para>
<indexterm significance="normal"><primary>tdb file descriptions</primary></indexterm>
	The old Samba-2.2.x tdb files are described in <link linkend="oldtdbfiledesc">the next table</link>.
	</para>


        <table frame="all" id="oldtdbfiledesc"><title>Samba-2.2.x TDB File Descriptions</title>
        <tgroup cols="3">
			<colspec align="left"/>
			<colspec align="justify" colwidth="1*"/>
			<colspec align="left"/>
                <thead>
                <row>
                        <entry align="left">Name</entry>
                        <entry align="justify">Description</entry>
                        <entry align="center">Backup?</entry>
                </row>
                </thead>
                <tbody>
                <row>
                        <entry>account_policy</entry>
			<entry>User policy settings</entry>
			<entry>yes</entry>
		</row>
		<row>
			<entry>brlock</entry>
			<entry>Byte-range file locking information.</entry>
			<entry>no</entry>
		</row>
		<row>
			<entry>connections</entry>
			<entry><para>Client connection information</para></entry>
			<entry>no</entry>
		</row>
		<row>
			<entry>locking</entry>
			<entry>Temporary file locking data.</entry>
			<entry>no</entry>
		</row>
		<row>
			<entry>messages</entry>
			<entry><para>Temporary storage of messages being processed by smbd.</para></entry>
			<entry>no</entry>
		</row>
       <row>
            <entry>ntdrivers</entry>
            <entry><para>Stores per-printer driver information.</para></entry>
			<entry>yes</entry>
        </row>
        <row>
            <entry>ntforms</entry>
            <entry><para>Stores per-printer forms information.</para></entry>
			<entry>yes</entry>
        </row>
        <row>
            <entry>ntprinters</entry>
            <entry><para>Stores the per-printer devmode configuration settings.</para></entry>
			<entry>yes</entry>
        </row>
		<row>
			<entry>printing/*.tdb</entry>
			<entry><para>Cached output from lpq command created on a per-print-service basis.</para></entry>
			<entry>no</entry>
		</row>
		<row>

			<entry>registry</entry>
			<entry><para>Read-only Samba registry skeleton that provides support for
				exporting various database tables via the winreg RPCs.</para></entry>
			<entry>no</entry>
		</row>
		<row>
			<entry>sessionid</entry>
			<entry><para>Temporary cache for miscellaneous session information.</para></entry>
			<entry>no</entry>
		</row>
		<row>
			<entry>share_info</entry>
			<entry>Share ACL settings.</entry>
			<entry>yes</entry>
		</row>
		<row>

			<entry>unexpected</entry>
			<entry><para>Packets received for which no process was listening.</para></entry>
			<entry>no</entry>
		</row>
		<row>
			<entry>winbindd_cache</entry>
			<entry><para>Cache of identity information received from an NT4 or an ADS domain.</para></entry>
			<entry>yes</entry>
		</row>
		<row>
			<entry>winbindd_idmap</entry>
			<entry><para>New ID map table from SIDS to UNIX UIDs/GIDs.</para></entry>
			<entry>yes</entry>
		</row>
		</tbody>
	</tgroup>
	</table>

	</sect3>

	<sect3>
	<title>Changes in Behavior</title>

	<para>
	The following issues are known changes in behavior between Samba-2.2 and
	Samba-3 that may affect certain installations of Samba.
	</para>

	<orderedlist inheritnum="ignore" continuation="restarts">
		<listitem><para>
<indexterm significance="normal"><primary>Windows domain</primary></indexterm>
<indexterm significance="normal"><primary>getpwnam() call</primary></indexterm>
<indexterm significance="normal"><primary>NT_STATUS_LOGON_FAILURE</primary></indexterm>
		When operating as a member of a Windows domain, Samba-2.2 would map any users authenticated by the remote DC
		to the <quote>guest account</quote> if a UID could not be obtained via the getpwnam() call. Samba-3 rejects
		the connection with the error message <quote>NT_STATUS_LOGON_FAILURE.</quote> There is no current workaround
		to re-establish the Samba-2.2 behavior.
		</para></listitem>

		<listitem><para>
<indexterm significance="normal"><primary>add user script</primary></indexterm>
<indexterm significance="normal"><primary>add machine script</primary></indexterm>
		When adding machines to a Samba-2.2 controlled domain, the
		<quote>add user script</quote> was used to create the UNIX identity of the
		machine trust account. Samba-3 introduces a new <quote>add machine
		script</quote> that must be specified for this purpose. Samba-3 will
		not fall back to using the <quote>add user script</quote> in the absence of
		an <quote>add machine script</quote>.
		</para></listitem>
	</orderedlist>

	</sect3>

	<sect3>
	<title>Passdb Backends and Authentication</title>

	<para>
	There have been a few new changes that Samba administrators should be
	aware of when moving to Samba-3.
	</para>

	<orderedlist inheritnum="ignore" continuation="restarts">
		<listitem><para>
<indexterm significance="normal"><primary>encrypted passwords</primary></indexterm>
		Encrypted passwords have been enabled by default in order to
		interoperate better with out-of-the-box Windows client
		installations. This does mean that either (a) a Samba account
		must be created for each user, or (b) <quote>encrypt passwords = no</quote>
		must be explicitly defined in <filename moreinfo="none">smb.conf</filename>.
		</para></listitem>

		<listitem><para>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
<indexterm significance="normal"><primary>Kerberos</primary></indexterm>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
		Inclusion of new <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SECURITY" xlink:href="smb.conf.5.html#SECURITY">security = ads</link> option for integration
		with an Active Directory domain using the native Windows Kerberos 5 and LDAP protocols.
		</para></listitem>
	</orderedlist>

	<para>
<indexterm significance="normal"><primary>account storage backends</primary></indexterm>
	Samba-3 also includes the possibility of setting up chains of authentication methods (<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="AUTHMETHODS" xlink:href="smb.conf.5.html#AUTHMETHODS">auth methods</link>) and account storage backends (<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PASSDBBACKEND" xlink:href="smb.conf.5.html#PASSDBBACKEND">passdb backend</link>).  Please refer to
	the <filename moreinfo="none">smb.conf</filename> man page and <link linkend="passdb">Account Information Databases</link>, for
	details. While both parameters assume sane default values, it is likely that you will need to understand what
	the values actually mean in order to ensure Samba operates correctly.
	</para>

	<para>
<indexterm significance="normal"><primary>pdbedit</primary></indexterm>
<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
<indexterm significance="normal"><primary>net tool</primary></indexterm>
	Certain functions of the <literal>smbpasswd</literal> tool have been split between the
	new <literal>smbpasswd</literal> utility, the <literal>net</literal> tool, and the new <literal>pdbedit</literal>
	utility. See the respective man pages for details.
	</para>

	</sect3>

	<sect3>
	<title>LDAP</title>

	<para>
	This section outlines the new features effecting Samba/LDAP integration.
	</para>

		<sect4>
		<title>New Schema</title>

		<para>
<indexterm significance="normal"><primary>object class</primary></indexterm>
<indexterm significance="normal"><primary>sambaSamAccount</primary></indexterm>
<indexterm significance="normal"><primary>LDIF</primary></indexterm>
<indexterm significance="normal"><primary>attributes</primary></indexterm>
		A new object class (sambaSamAccount) has been introduced to replace
		the old sambaAccount. This change aids in the renaming of attributes
		to prevent clashes with attributes from other vendors. There is a
		conversion script (examples/LDAP/convertSambaAccount) to modify an LDIF
		file to the new schema.
		</para>

		<para>
		Example:
<indexterm significance="normal"><primary>ldapsearch</primary></indexterm>
		</para>
		<para><screen format="linespecific">
		<prompt moreinfo="none">$ </prompt>ldapsearch .... -LLL -b "ou=people,dc=..." &gt; old.ldif
		<prompt moreinfo="none">$ </prompt>convertSambaAccount --sid &lt;DOM SID&gt; --input old.ldif --output new.ldif
		</screen></para>

		<para>
<indexterm significance="normal"><primary>net</primary><secondary>getlocalsid</secondary></indexterm>
		The &lt;DOM SID&gt; can be obtained by running
<screen format="linespecific">
<prompt moreinfo="none">$ </prompt><userinput moreinfo="none">net getlocalsid &lt;DOMAINNAME&gt;</userinput>
</screen>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
		on the Samba PDC as root.
		</para>

		<para>
		Under Samba-2.x the domain SID can be obtained by executing:
<indexterm significance="normal"><primary>smbpasswd</primary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">$ </prompt><userinput moreinfo="none">smbpasswd -S &lt;DOMAINNAME&gt;</userinput>
</screen>
		</para>

		<para>
<indexterm significance="normal"><primary>old sambaAccount</primary></indexterm>
<indexterm significance="normal"><primary>ldapsam_compat</primary></indexterm>
<indexterm significance="normal"><primary>object class declaration</primary></indexterm>
<indexterm significance="normal"><primary>samba.schema</primary></indexterm>
		The old <literal moreinfo="none">sambaAccount</literal> schema may still be used by specifying the
		<parameter moreinfo="none">ldapsam_compat</parameter> passdb backend. However, the sambaAccount and
		associated attributes have been moved to the historical section of
		the schema file and must be uncommented before use if needed.
		The Samba-2.2 object class declaration for a <literal moreinfo="none">sambaAccount</literal> has not changed
		in the Samba-3 <filename moreinfo="none">samba.schema</filename> file.
		</para>

		<para>
		Other new object classes and their uses include:
		</para>

		<itemizedlist>
			<listitem><para>
<indexterm significance="normal"><primary>sambaDomain</primary></indexterm>
<indexterm significance="normal"><primary>domain information</primary></indexterm>
<indexterm significance="normal"><primary>RID</primary></indexterm>
<indexterm significance="normal"><primary>ldap suffix</primary></indexterm>
<indexterm significance="normal"><primary>ldapsam</primary></indexterm>
<indexterm significance="normal"><primary>idmap</primary></indexterm>
			<literal moreinfo="none">sambaDomain</literal> <?latex --- ?> domain information used to allocate RIDs
			for users and groups as necessary. The attributes are added
			in <quote>ldap suffix</quote> directory entry automatically if
			an idmap UID/GID range has been set and the <quote>ldapsam</quote>
			passdb backend has been selected.
			</para></listitem>

			<listitem><para>
<indexterm significance="normal"><primary>sambaGroupMapping</primary></indexterm>
<indexterm significance="normal"><primary>ldap group suffix</primary></indexterm>
<indexterm significance="normal"><primary>net groupmap</primary></indexterm>
			sambaGroupMapping <?latex --- ?> an object representing the
			relationship between a posixGroup and a Windows
			group/SID. These entries are stored in the <quote>ldap
			group suffix</quote> and managed by the <quote>net groupmap</quote> command.
			</para></listitem>

			<listitem><para>
<indexterm significance="normal"><primary>sambaUNIXIdPool</primary></indexterm>
<indexterm significance="normal"><primary>ldap idmap suffix</primary></indexterm>
<indexterm significance="normal"><primary>idmap UID</primary></indexterm>
<indexterm significance="normal"><primary>idmap GID</primary></indexterm>
			<literal moreinfo="none">sambaUNIXIdPool</literal> <?latex --- ?> created in the <quote>ldap idmap suffix</quote> entry
			automatically and contains the next available <quote>idmap UID</quote> and
			<quote>idmap GID</quote>.
			</para></listitem>

			<listitem><para>
<indexterm significance="normal"><primary>sambaIdmapEntry</primary></indexterm>
<indexterm significance="normal"><primary>idmap_ldap module</primary></indexterm>
			<literal moreinfo="none">sambaIdmapEntry</literal> <?latex --- ?> object storing a mapping between a
			SID and a UNIX UID/GID. These objects are created by the
			idmap_ldap module as needed.
			</para></listitem>
		</itemizedlist>

		</sect4>

		<sect4>
		<title>New Suffix for Searching</title>

		<para>
<indexterm significance="normal"><primary>LDAP queries</primary></indexterm>
<indexterm significance="normal"><primary>passdb backend</primary></indexterm>
<indexterm significance="normal"><primary>ldap suffix</primary></indexterm>
<indexterm significance="normal"><primary>ldap user suffix</primary></indexterm>
<indexterm significance="normal"><primary>ldap machine suffix</primary></indexterm>
<indexterm significance="normal"><primary>ldap group suffix</primary></indexterm>
<indexterm significance="normal"><primary>ldap idmap suffix</primary></indexterm>
		The following new <filename moreinfo="none">smb.conf</filename> parameters have been added to aid in directing
		certain LDAP queries when <parameter moreinfo="none">passdb backend = ldapsam://...</parameter> has been
		specified.
		</para>

		<itemizedlist>
			<listitem><para>ldap suffix         <?latex --- ?> used to search for user and computer accounts.</para></listitem>
			<listitem><para>ldap user suffix    <?latex --- ?> used to store user accounts.</para></listitem>
			<listitem><para>ldap machine suffix <?latex --- ?> used to store machine trust accounts.</para></listitem>
			<listitem><para>ldap group suffix   <?latex --- ?> location of posixGroup/sambaGroupMapping entries.</para></listitem>
			<listitem><para>ldap idmap suffix   <?latex --- ?> location of sambaIdmapEntry objects.</para></listitem>
		</itemizedlist>

		<para>
<indexterm significance="normal"><primary>ldap suffix</primary></indexterm>
<indexterm significance="normal"><primary>subsuffix parameters</primary></indexterm>
		If an <parameter moreinfo="none">ldap suffix</parameter> is defined, it will be appended to all of the
		remaining subsuffix parameters. In this case, the order of the suffix
		listings in <filename moreinfo="none">smb.conf</filename> is important. Always place the <parameter moreinfo="none">ldap suffix</parameter> first
		in the list.
		</para>

		<para>
		Due to a limitation in Samba's <filename moreinfo="none">smb.conf</filename> parsing, you should not surround
		the domain names with quotation marks.
		</para>

		</sect4>

		<sect4>
		<title>IdMap LDAP Support</title>

		<para>
<indexterm significance="normal"><primary>idmap backend</primary></indexterm>
		Samba-3 supports an LDAP backend for the idmap subsystem. The
		following options inform Samba that the idmap table should be
		stored on the directory server <emphasis>onterose</emphasis> in the ou=Idmap,dc=quenya,dc=org partition.
		</para>

		<simplelist>
		<member> </member><member><parameter>[global]</parameter></member>
		<member>...</member>
		<member><indexterm>idmap backend</indexterm><parameter>idmap backend = ldap:ldap://onterose/</parameter></member>
		<member><indexterm>ldap idmap suffix</indexterm><parameter>ldap idmap suffix = ou=Idmap</parameter></member>
		<member><indexterm>idmap uid</indexterm><parameter>idmap uid = 40000-50000</parameter></member>
		<member><indexterm>idmap gid</indexterm><parameter>idmap gid = 40000-50000</parameter></member>
		</simplelist>

		<para>
<indexterm significance="normal"><primary>NFS</primary></indexterm>
		This configuration allows Winbind installations on multiple servers to
		share a UID/GID number space, thus avoiding the interoperability problems
		with NFS that were present in Samba-2.2.
		</para>

		</sect4>

		</sect3>

	</sect2>

</sect1>

</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="NT4Migration">
<chapterinfo>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
	<pubdate>April 3, 2003</pubdate>
</chapterinfo>

<title>Migration from NT4 PDC to Samba-3 PDC</title>

<para>
<indexterm significance="normal"><primary>migrate</primary></indexterm>
<indexterm significance="normal"><primary>domain control</primary></indexterm>
This is a rough guide to assist those wishing to migrate from NT4 domain control to
Samba-3-based domain control.
</para>

<sect1>
<title>Planning and Getting Started</title>

<para>
<indexterm significance="normal"><primary>show-stopper-type</primary></indexterm>
In the IT world there is often a saying that all problems are encountered because of
poor planning. The corollary to this saying is that not all problems can be anticipated
and planned for. Then again, good planning will anticipate most show-stopper-type situations.
</para>

<para>
<indexterm significance="normal"><primary>migration plan</primary></indexterm>
Those wishing to migrate from MS Windows NT4 domain control to a Samba-3 domain control
environment would do well to develop a detailed migration plan. So here are a few pointers to
help migration get underway.
</para>

<sect2>
<title>Objectives</title>

<para>
<indexterm significance="normal"><primary>migration process</primary></indexterm>
The key objective for most organizations is to make the migration from MS Windows NT4 
to Samba-3 domain control as painless as possible. One of the challenges you may experience
in your migration process may well be convincing management that the new environment
should remain in place. Many who have introduced open source technologies have experienced
pressure to return to a Microsoft-based platform solution at the first sign of trouble. 
</para>

<para>
<indexterm significance="normal"><primary>change motivations</primary></indexterm>
Before attempting a migration to a Samba-3-controlled network, make every possible effort to
gain all-round commitment to the change. Know precisely <emphasis>why</emphasis> the change
is important for the organization. Possible motivations to make a change include:
</para>

<indexterm significance="normal"><primary>manageability</primary></indexterm>
<indexterm significance="normal"><primary>functionality</primary></indexterm>
<indexterm significance="normal"><primary>operating costs</primary></indexterm>
<indexterm significance="normal"><primary>support exposure</primary></indexterm>
<indexterm significance="normal"><primary>licensing</primary></indexterm>

<itemizedlist>
    <listitem><para>Improve network manageability.</para></listitem>
    <listitem><para>Obtain better user-level functionality.</para></listitem>
    <listitem><para>Reduce network operating costs.</para></listitem>
    <listitem><para>Reduce exposure caused by Microsoft withdrawal of NT4 support.</para></listitem>
    <listitem><para>Avoid MS License 6 implications.</para></listitem>
    <listitem><para>Reduce organization's dependency on Microsoft.</para></listitem>
</itemizedlist>

<para>
<indexterm significance="normal"><primary>alternative solution</primary></indexterm>
<indexterm significance="normal"><primary>advantages</primary></indexterm>
<indexterm significance="normal"><primary>core values</primary></indexterm>
<indexterm significance="normal"><primary>migration</primary></indexterm>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
<indexterm significance="normal"><primary>without ADS</primary></indexterm>
Make sure everyone knows that Samba-3 is not MS Windows NT4. Samba-3 offers
an alternative solution that is both different from MS Windows NT4 and offers 
advantages compared with it. Gain recognition that Samba-3 lacks many of the
features that Microsoft has promoted as core values in migration from MS Windows NT4 to 
MS Windows 2000 and beyond (with or without Active Directory services).
</para>

<para>
What are the features that Samba-3 cannot provide?
</para>

<indexterm significance="normal"><primary>Active Directory Server</primary></indexterm>
<indexterm significance="normal"><primary>Group Policy Objects</primary></indexterm>
<indexterm significance="normal"><primary>Machine Policy Objects</primary></indexterm>
<indexterm significance="normal"><primary>Logon Scripts</primary></indexterm>
<indexterm significance="normal"><primary>Access Controls</primary></indexterm>

<itemizedlist>
	<listitem><para>Active Directory Server.</para></listitem>
	<listitem><para>Group Policy Objects (in Active Directory).</para></listitem>
	<listitem><para>Machine Policy Objects.</para></listitem>
	<listitem><para>Logon Scripts in Active Directory.</para></listitem>
	<listitem><para>Software Application and Access Controls in Active Directory.</para></listitem>
</itemizedlist>

<para>
The features that Samba-3 does provide and that may be of compelling interest to your site
include:
</para>

<indexterm significance="normal"><primary>ownership cost</primary></indexterm>
<indexterm significance="normal"><primary>Global support</primary></indexterm>
<indexterm significance="normal"><primary>Dynamic SMB servers</primary></indexterm>
<indexterm significance="normal"><primary>on-the-fly logon scripts</primary></indexterm>
<indexterm significance="normal"><primary>on-the-fly policy files</primary></indexterm>
<indexterm significance="normal"><primary>stability</primary></indexterm>
<indexterm significance="normal"><primary>reliability</primary></indexterm>
<indexterm significance="normal"><primary>performance</primary></indexterm>
<indexterm significance="normal"><primary>availability</primary></indexterm>
<indexterm significance="normal"><primary>Manageability</primary></indexterm>
<indexterm significance="normal"><primary>backend authentication</primary></indexterm>
<indexterm significance="normal"><primary>tdbsam</primary></indexterm>
<indexterm significance="normal"><primary>ldapsam</primary></indexterm>
<indexterm significance="normal"><primary>single-sign-on</primary></indexterm>
<indexterm significance="normal"><primary>distribute authentication systems</primary></indexterm>

<itemizedlist>
	<listitem><para>Lower cost of ownership.</para></listitem>
	<listitem><para>Global availability of support with no strings attached.</para></listitem>
	<listitem><para>Dynamic SMB servers (can run more than one SMB/CIFS server per UNIX/Linux system).</para></listitem>
	<listitem><para>Creation of on-the-fly logon scripts.</para></listitem>
	<listitem><para>Creation of on-the-fly policy files.</para></listitem>
	<listitem><para>Greater stability, reliability, performance, and availability.</para></listitem>
	<listitem><para>Manageability via an SSH connection.</para></listitem>
	<listitem><para>Flexible choices of backend authentication technologies (tdbsam, ldapsam).</para></listitem>
	<listitem><para>Ability to implement a full single-sign-on architecture.</para></listitem>
	<listitem><para>Ability to distribute authentication systems for absolute minimum wide-area network bandwidth demand.</para></listitem>
</itemizedlist>

<para>
<indexterm significance="normal"><primary>successful migration</primary></indexterm>
Before migrating a network from MS Windows NT4 to Samba-3, consider all necessary factors. Users
should be educated about changes they may experience so the change will be a welcome one
and not become an obstacle to the work they need to do. The following sections explain factors that will 
help ensure a successful migration.
</para>

<sect3>
<title>Domain Layout</title>

<para>
<indexterm significance="normal"><primary>domain controller</primary></indexterm>
<indexterm significance="normal"><primary>backup domain controller</primary></indexterm>
<indexterm significance="normal"><primary>secondary controller</primary></indexterm>
<indexterm significance="normal"><primary>domain member</primary></indexterm>
<indexterm significance="normal"><primary>standalone server</primary></indexterm>
<indexterm significance="normal"><primary>network security</primary></indexterm>
<indexterm significance="normal"><primary>domain context</primary></indexterm>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>BDCs</primary></indexterm>
<indexterm significance="normal"><primary>LDAP</primary></indexterm>
<indexterm significance="normal"><primary>authentication backend</primary></indexterm>
<indexterm significance="normal"><primary>complex organization</primary></indexterm>
<indexterm significance="normal"><primary>LDAP database</primary></indexterm>
<indexterm significance="normal"><primary>master server</primary></indexterm>
<indexterm significance="normal"><primary>slave servers</primary></indexterm>
<indexterm significance="normal"><primary>multiple domains</primary></indexterm>
Samba-3 can be configured as a domain controller, a backup domain controller (probably best called
a secondary controller), a domain member, or a standalone server. The Windows network security
domain context should be sized and scoped before implementation. Particular attention needs to be
paid to the location of the Primary Domain Controller (PDC) as well as backup controllers (BDCs).
One way in which Samba-3 differs from Microsoft technology is that if one chooses to use an LDAP
authentication backend, then the same database can be used by several different domains. In a
complex organization, there can be a single LDAP database, which itself can be distributed (have
a master server and multiple slave servers) that can simultaneously serve multiple domains.
</para>

<para>
<indexterm significance="normal"><primary>network bandwidth</primary></indexterm>
From a design perspective, the number of users per server as well as the number of servers per
domain should be scaled taking into consideration server capacity and network bandwidth.
</para>

<para>
<indexterm significance="normal"><primary>network segment</primary></indexterm>
<indexterm significance="normal"><primary>multiple network segments</primary></indexterm>
<indexterm significance="normal"><primary>domain controller</primary></indexterm>
<indexterm significance="normal"><primary>ping</primary></indexterm>
<indexterm significance="normal"><primary>BDC</primary></indexterm>
<indexterm significance="normal"><primary>remote segment</primary></indexterm>
A physical network segment may house several domains. Each may span multiple network segments.
Where domains span routed network segments, consider and test the performance implications of
the design and layout of a network. A centrally located domain controller that is designed to
serve multiple routed network segments may result in severe performance problems. Check the
response time (ping timing) between the remote segment and the PDC. If it's long (more than 100 ms),
locate a BDC on the remote segment to serve as the local authentication and access control server.
</para>
</sect3>

<sect3>
<title>Server Share and Directory Layout</title>

<para>
<indexterm significance="normal"><primary>Simplicity is king</primary></indexterm>
<indexterm significance="normal"><primary>well-controlled network</primary></indexterm>
There are cardinal rules to effective network design that cannot be broken with impunity.
The most important rule: Simplicity is king in every well-controlled network. Every part of
the infrastructure must be managed; the more complex it is, the greater will be the demand
of keeping systems secure and functional.
</para>

<para>
<indexterm significance="normal"><primary>disk space</primary></indexterm>
<indexterm significance="normal"><primary>backed up</primary></indexterm>
<indexterm significance="normal"><primary>tape</primary></indexterm>
<indexterm significance="normal"><primary>backup</primary></indexterm>
<indexterm significance="normal"><primary>validate every backup</primary></indexterm>
<indexterm significance="normal"><primary>disaster recovery</primary></indexterm>
Keep in mind the nature of how data must be shared. Physical disk space layout should be considered
carefully. Some data must be backed up. The simpler the disk layout, the easier it will be to
keep track of backup needs. Identify what backup media will meet your needs; consider backup to tape,
CD-ROM or DVD-ROM, or other offline storage medium. Plan and implement for minimum
maintenance. Leave nothing to chance in your design; above all, do not leave backups to chance:
backup, test, and validate every backup; create a disaster recovery plan and prove that it works.
</para>

<para>
<indexterm significance="normal"><primary>access control needs</primary></indexterm>
<indexterm significance="normal"><primary>group permissions</primary></indexterm>
<indexterm significance="normal"><primary>sticky bit</primary></indexterm>
Users should be grouped according to data access control needs. File and directory access 
is best controlled via group permissions, and the use of the <quote>sticky bit</quote> on group-controlled
directories may substantially avoid file access complaints from Samba share users.
</para>

<para>
<indexterm significance="normal"><primary>network administrators</primary></indexterm>
<indexterm significance="normal"><primary>document design</primary></indexterm>
<indexterm significance="normal"><primary>simple access controls</primary></indexterm>
<indexterm significance="normal"><primary>obtuse complexity</primary></indexterm>
<indexterm significance="normal"><primary>document design</primary></indexterm>
Inexperienced  network administrators often attempt elaborate techniques to set access
controls on files, directories, shares, as well as in share definitions.
Keep your design and implementation simple and document your design extensively. Have others
audit your documentation. Do not create a complex mess that your successor will not understand.
Remember, job security through complex design and implementation may cause loss of operations
and downtime to users as the new administrator learns to untangle your knots. Keep access
controls simple and effective, and make sure that users will never be interrupted by obtuse
complexity.
</para>
</sect3>

<sect3>
<title>Logon Scripts</title>

<para>
<indexterm significance="normal"><primary>Logon scripts</primary></indexterm>
Logon scripts can help to ensure that all users gain the share and printer connections they need.
</para>

<para>
Logon scripts can be created on the fly so all commands executed are specific to the
rights and privileges granted to the user. The preferred controls should be effected through
group membership so group information can be used to create a custom logon script using
the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="ROOTPREEXEC" xlink:href="smb.conf.5.html#ROOTPREEXEC">root preexec</link> parameters to the <parameter>NETLOGON</parameter> share.
</para>

<para>
<indexterm significance="normal"><primary>kixstart</primary></indexterm>
Some sites prefer to use a tool such as <literal>kixstart</literal> to establish a controlled
user environment. In any case, you may wish to do a Google search for logon script process controls.
In particular, you may wish to explore the use of the Microsoft Knowledge Base article KB189105 that
deals with how to add printers without user intervention via the logon script process.
</para>
</sect3>

<sect3>
<title>Profile Migration/Creation</title>

<para>
User and group profiles may be migrated using the tools described in the section titled Desktop Profile
Management.
</para>


<para>
<indexterm significance="normal"><primary>SID</primary></indexterm>
<indexterm significance="normal"><primary>NTuser.DAT</primary></indexterm>
Profiles may also be managed using the Samba-3 tool <literal>profiles</literal>. This tool allows the MS
Windows NT-style security identifiers (SIDs) that are stored inside the profile
<filename moreinfo="none">NTuser.DAT</filename> file to be changed to the SID of the Samba-3 domain.
</para>
</sect3>

<sect3>
<title>User and Group Accounts</title>

<para>
<indexterm significance="normal"><primary>migrate account settings</primary></indexterm>
<indexterm significance="normal"><primary>migrate user</primary></indexterm>
<indexterm significance="normal"><primary>migrate group</primary></indexterm>
<indexterm significance="normal"><primary>map</primary></indexterm>
It is possible to migrate all account settings from an MS Windows NT4 domain to Samba-3. Before
attempting to migrate user and group accounts, you are STRONGLY advised to create in Samba-3 the
groups that are present on the MS Windows NT4 domain <emphasis>AND</emphasis> to map them to
suitable UNIX/Linux groups. By following this simple advice, all user and group attributes
should migrate painlessly.
</para>
</sect3>

</sect2>

<sect2>
<title>Steps in Migration Process</title>

<para>
The approximate migration process is described below.
</para>

<itemizedlist>
	<listitem><para>
	You have an NT4 PDC that has the users, groups, policies, and profiles to be migrated.
	</para></listitem>

	<listitem><para>
<indexterm significance="normal"><primary>domain controller</primary></indexterm>
<indexterm significance="normal"><primary>netlogon share</primary></indexterm>
<indexterm significance="normal"><primary>BDC</primary></indexterm>
	Samba-3 is set up as a domain controller with netlogon share, profile share, and so on. Configure the <filename moreinfo="none">smb.conf</filename> file
	to function as a BDC: <parameter moreinfo="none">domain master = No</parameter>.
	</para></listitem>
</itemizedlist>

<procedure>
<title>The Account Migration Process</title>

	<step performance="required"><para>
	<indexterm significance="normal"><primary>pdbedit</primary></indexterm>
	Create a BDC account in the old NT4 domain for the Samba server using NT Server Manager.
	<emphasis>Samba must not be running.</emphasis>
	</para></step>

	<step performance="required"><para>
	<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>join</tertiary></indexterm>
	<userinput moreinfo="none">net rpc join -S <replaceable>NT4PDC</replaceable> -w <replaceable>DOMNAME</replaceable> -U
	Administrator%<replaceable>passwd</replaceable></userinput>
	</para></step>

	<step performance="required"><para>
<indexterm significance="normal"><primary>net</primary><secondary>rpc</secondary><tertiary>vampire</tertiary></indexterm>
	<userinput moreinfo="none">net rpc vampire -S <replaceable>NT4PDC</replaceable> -U 
	administrator%<replaceable>passwd</replaceable></userinput>
	</para></step>

<indexterm significance="normal"><primary>pdbedit</primary></indexterm>
	<step performance="required"><para><userinput moreinfo="none">pdbedit -L</userinput></para>
		<para>Note: Did the users migrate?</para>
	</step>

	<step performance="required"><para>
	<indexterm significance="normal"><primary>net</primary><secondary>groupmap</secondary></indexterm>
	<indexterm significance="normal"><primary>initGroups.sh</primary></indexterm>
	Now assign each of the UNIX groups to NT groups:
	(It may be useful to copy this text to a script called <filename moreinfo="none">initGroups.sh</filename>)
	<programlisting format="linespecific">
#!/bin/bash
#### Keep this as a shell script for future re-use
			
# First assign well known domain global groups
net groupmap add ntgroup="Domain Admins" unixgroup=root rid=512 type=d
net groupmap add ntgroup="Domain Users"  unixgroup=users rid=513 type=d
net groupmap add ntgroup="Domain Guests" unixgroup=nobody rid=514 type=d

# Now for our added domain global groups
net groupmap add ntgroup="Designers" unixgroup=designers type=d
net groupmap add ntgroup="Engineers" unixgroup=engineers type=d
net groupmap add ntgroup="QA Team"   unixgroup=qateam    type=d
</programlisting>
	</para></step>

	<step performance="required"><para><userinput moreinfo="none">net groupmap list</userinput></para>
		<para>Check that all groups are recognized.
	</para></step>
</procedure>

<para>
Migrate all the profiles, then migrate all policy files.
</para>

</sect2>
</sect1>

<sect1>
<title>Migration Options</title>

<para>
Sites that wish to migrate from MS Windows NT4 domain control to a Samba-based solution
generally fit into three basic categories. <link linkend="majtypes">Following table</link> shows the possibilities.
</para>

<table frame="all" id="majtypes"><title>The Three Major Site Types</title>
<tgroup cols="2">
	<colspec align="left"/>
	<colspec align="justify"/> 
	<thead>
	<row><entry>Number of Users</entry><entry>Description</entry></row>
	</thead>
	<tbody>
	<row><entry>&lt; 50</entry><entry><para>Want simple conversion with no pain.</para></entry></row>
	<row><entry>50 - 250</entry><entry><para>Want new features; can manage some inhouse complexity.</para></entry></row>
	<row><entry>&gt; 250</entry><entry><para>Solution/implementation must scale well; complex needs.
		Cross-departmental decision process. Local expertise in most areas.</para></entry></row>
	</tbody>
</tgroup>
</table>

<sect2>
<title>Planning for Success</title>

<para>
There are three basic choices for sites that intend to migrate from MS Windows NT4
to Samba-3:
</para>

<itemizedlist>
	<listitem><para>
	Simple conversion (total replacement).
	</para></listitem>

	<listitem><para>
	Upgraded conversion (could be one of integration).
	</para></listitem>

	<listitem><para>
	Complete redesign (completely new solution).
	</para></listitem>
</itemizedlist>

<para>
Minimize downstream problems by:
</para>

<itemizedlist>
	<listitem><para>
	Taking sufficient time.
	</para></listitem>

	<listitem><para>
	Avoiding panic.
	</para></listitem>

	<listitem><para>
	Testing all assumptions.
	</para></listitem>

	<listitem><para>
	Testing the full roll-out program, including workstation deployment.
	</para></listitem>
</itemizedlist>

<para><link linkend="natconchoices">Following table</link> lists the conversion choices given the type of migration
being contemplated.
</para>

<table frame="all" id="natconchoices"><title>Nature of the Conversion Choices</title>
<tgroup cols="3">
	<colspec align="justify" colwidth="1*"/>
	<colspec align="justify" colwidth="1*"/>
	<colspec align="justify" colwidth="1*"/>
	<thead>
	<row><entry>Simple Install</entry><entry>Upgrade Decisions</entry><entry>Redesign Decisions</entry></row>
	</thead>
	<tbody>
	<row>
	<entry><para>Make use of minimal OS-specific features</para></entry>
	<entry><para>Translate NT4 features to new host OS features</para></entry>
	<entry><para>Improve on NT4 functionality, enhance management capabilities</para></entry>
	</row>
	<row>
	<entry><para>Move all accounts from NT4 into Samba-3</para></entry>
	<entry><para>Copy and improve</para></entry>
	<entry><para>Authentication regime (database location and access)</para></entry>
	</row>
	<row>
	<entry><para>Make least number of operational changes</para></entry>
	<entry><para>Make progressive improvements</para></entry>
	<entry><para>Desktop management methods</para></entry>
	</row>
	<row>
	<entry><para>Take least amount of time to migrate</para></entry>
	<entry><para>Minimize user impact</para></entry>
	<entry><para>Better control of Desktops/Users</para></entry>
	</row>
	<row>
	<entry><para>Live versus isolated conversion</para></entry>
	<entry><para>Maximize functionality</para></entry>
	<entry><para>Identify Needs for: <emphasis>Manageability, Scalability, Security, Availability</emphasis></para></entry>
	</row>
	<row>
	<entry><para>Integrate Samba-3, then migrate while users are active, then change of control (swap out)</para></entry>
	<entry><para>Take advantage of lower maintenance opportunity</para></entry>
	<entry><para/></entry>
	</row>
	</tbody>
</tgroup>
</table>
</sect2>

<sect2>
<title>Samba-3 Implementation Choices</title>

<variablelist>
		<varlistentry><term>Authentication Database/Backend</term><listitem>
		<para>
		Samba-3 can use an external authentication backend:
		</para>

		<para>
		<itemizedlist>
			<listitem><para>Winbind (external Samba or NT4/200x server).</para></listitem>
			<listitem><para>External server could use Active Directory or NT4 domain.</para></listitem>
			<listitem><para>Can use pam_mkhomedir.so to autocreate home directories.</para></listitem>
			<listitem><para> Samba-3 can use a local authentication backend: <parameter moreinfo="none">smbpasswd</parameter>,
				<parameter moreinfo="none">tdbsam</parameter>, <parameter moreinfo="none">ldapsam</parameter>
			</para></listitem>
		</itemizedlist></para></listitem>
		</varlistentry>

        <varlistentry><term>Access Control Points</term><listitem>
		<para>
		Samba permits Access Control points to be set:
		</para>

<indexterm significance="normal"><primary>share ACLs</primary></indexterm>
<indexterm significance="normal"><primary>UNIX permissions</primary></indexterm>
<indexterm significance="normal"><primary>POSIX ACLS</primary></indexterm>
<indexterm significance="normal"><primary>share stanza controls</primary></indexterm>

		<itemizedlist>
			<listitem><para>On the share itself <?latex --- ?> using share ACLs.</para></listitem>
			<listitem><para>On the file system <?latex --- ?> using UNIX permissions on files and directories.</para>
			<para>Note: Can enable Posix ACLs in file system also.</para></listitem>
			<listitem><para>Through Samba share parameters <?latex --- ?> not recommended except as last resort.</para></listitem>
		</itemizedlist></listitem>
        </varlistentry>

        <varlistentry><term>Policies (migrate or create new ones)</term><listitem>
		<para>
<indexterm significance="normal"><primary>policies</primary></indexterm>
<indexterm significance="normal"><primary>NTConfig.POL</primary></indexterm>
		Exercise great caution when making registry changes; use the right tool and be aware
		that changes made through NT4-style <filename moreinfo="none">NTConfig.POL</filename> files can leave
		permanent changes.
<indexterm significance="normal"><primary>Group Policy Editor</primary></indexterm>
<indexterm significance="normal"><primary>tattoo effect</primary></indexterm>
<indexterm significance="normal"><primary>permanent changes</primary></indexterm>
		</para>
                <itemizedlist>
                        <listitem><para>Using Group Policy Editor (NT4).</para></listitem>
                        <listitem><para>Watch out for tattoo effect.</para></listitem>
		</itemizedlist>
                </listitem>
        </varlistentry>

        <varlistentry><term>User and Group Profiles</term><listitem>
		<para>
<indexterm significance="normal"><primary>NTUser.DAT</primary></indexterm>
<indexterm significance="normal"><primary>SIDs</primary></indexterm>
		Platform-specific, so use platform tool to change from a local to a roaming profile.
		Can use new profiles tool to change SIDs (<filename moreinfo="none">NTUser.DAT</filename>).
		</para>
                </listitem>
        </varlistentry>

        <varlistentry><term>Logon Scripts</term><listitem>
                <para>
		Know how they work.
		</para>
                </listitem>
        </varlistentry>


        <varlistentry><term>User and Group Mapping to UNIX/Linux</term><listitem>
		<para>
		<indexterm significance="normal"><primary>pdbedit</primary></indexterm>
		User and group mapping code is new. Many problems have been experienced as network administrators
		who are familiar with Samba-2.2.x migrate to Samba-3. Carefully study the chapters that document
		the new password backend behavior and the new group mapping functionality.
		</para>
			<itemizedlist>
				<listitem><para>The <parameter moreinfo="none">username map</parameter> facility may be needed.</para></listitem>
				<listitem><para>Use <literal>net groupmap</literal> to connect NT4 groups to UNIX groups.</para></listitem>
				<listitem><para>
					Use <literal>pdbedit</literal> to set/change user configuration.
					</para>

					<para>
					When migrating to LDAP backend, it may be easier to dump the initial
					LDAP database to LDIF, edit, then reload into LDAP.
					</para></listitem>
			</itemizedlist></listitem>
        </varlistentry>

        <varlistentry><term>OS-Specific Scripts/Programs May be Needed</term><listitem>
		<para>
		Every operating system has its peculiarities. These are the result of engineering decisions
		that were based on the experience of the designer and may have side effects that were not
		anticipated. Limitations that may bite the Windows network administrator include:
		</para>
                <itemizedlist>
                        <listitem><para>Add/Delete Users: Note OS limits on size of name
				(Linux 8 chars, NT4 up to 254 chars).</para></listitem>
                        <listitem><para>Add/Delete Machines: Applied only to domain members
				(Note: machine names may be limited to 16 characters).</para></listitem>
                        <listitem><para>Use <literal>net groupmap</literal> to connect NT4 groups to UNIX groups.</para></listitem>
                        <listitem><para>Add/Delete Groups: Note OS limits on size and nature.
				Linux limit is 16 char, no spaces, and no uppercase chars (<literal>groupadd</literal>).</para></listitem>
		</itemizedlist></listitem>
        </varlistentry>

        <varlistentry><term>Migration Tools</term><listitem>
                <para>
				<indexterm significance="normal"><primary>pdbedit</primary></indexterm>
				Domain Control (NT4-Style) Profiles, Policies, Access Controls, Security
				<itemizedlist>
					 <listitem><para>Samba: <literal>net, rpcclient, smbpasswd, pdbedit, profiles</literal></para></listitem>
					 <listitem><para>Windows: <literal>NT4 Domain User Manager, Server Manager (NEXUS)</literal></para></listitem>
				</itemizedlist></para></listitem>
        </varlistentry>
</variablelist>

</sect2>

</sect1>

</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="SWAT">
<chapterinfo>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
	<pubdate>April 21, 2003</pubdate>
</chapterinfo>

<title>SWAT: The Samba Web Administration Tool</title>

<para>
<indexterm significance="normal"><primary>configuration tool</primary></indexterm>
<indexterm significance="normal"><primary>SWAT</primary></indexterm>
<indexterm significance="normal"><primary>Web-based configuration</primary></indexterm>
There are many and varied opinions regarding the usefulness of SWAT.  No matter how hard one tries to produce
the perfect configuration tool, it remains an object of personal taste. SWAT is a tool that allows Web-based
configuration of Samba. It has a wizard that may help to get Samba configured quickly, it has
context-sensitive help on each <filename moreinfo="none">smb.conf</filename> parameter, it provides for monitoring of current state of connection
information, and it allows networkwide MS Windows network password management.
</para>

<sect1>
<title>Features and Benefits</title>

<para>
<indexterm significance="normal"><primary>internetworking super daemon</primary></indexterm>
SWAT is a facility that is part of the Samba suite. The main executable is called
<literal>swat</literal> and is invoked by the internetworking super daemon.
See <link linkend="xinetd">appropriate section</link> for details.
</para>

<para>
<indexterm significance="normal"><primary>man</primary></indexterm>
SWAT uses integral Samba components to locate parameters supported by the particular
version of Samba. Unlike tools and utilities that are external to Samba, SWAT is always
up to date as known Samba parameters change. SWAT provides context-sensitive help for each
configuration parameter, directly from <literal>man</literal> page entries.
</para>

<para>
<indexterm significance="normal"><primary>documentation</primary></indexterm>
<indexterm significance="normal"><primary>configuration files</primary></indexterm>
<indexterm significance="normal"><primary>internal ordering</primary></indexterm>
Some network administrators believe that it is a good idea to write systems
documentation inside configuration files, and for them SWAT will always be a nasty tool. SWAT
does not store the configuration file in any intermediate form; rather, it stores only the
parameter settings, so when SWAT writes the <filename moreinfo="none">smb.conf</filename> file to disk, it writes only
those parameters that are at other than the default settings. The result is that all comments,
as well as parameters that are no longer supported, will be lost from the <filename moreinfo="none">smb.conf</filename> file.
Additionally, the parameters will be written back in internal ordering.
</para>

<note><para>
<indexterm significance="normal"><primary>stripped of comments</primary></indexterm>
Before using SWAT, please be warned <?latex --- ?> SWAT will completely replace your <filename moreinfo="none">smb.conf</filename> with
a fully optimized file that has been stripped of all comments you might have placed there
and only nondefault settings will be written to the file.
</para></note>

</sect1>

<sect1>
<title>Guidelines and Technical Tips</title>

<para>
<indexterm significance="normal"><primary>internationalization support</primary></indexterm>
This section aims to unlock the dark secrets behind how SWAT may be made to work,
how it can be made more secure, and how to solve internationalization support problems.
</para>

<sect2>
<title>Validate SWAT Installation</title>

<para>
<indexterm significance="normal"><primary>SWAT binary support</primary></indexterm>
The very first step that should be taken before attempting to configure a host
system for SWAT operation is to check that it is installed. This may seem a trivial
point to some, but several Linux distributions do not install SWAT by default,
even though they do ship an installable binary support package containing SWAT
on the distribution media.
</para>

<para>
<indexterm significance="normal"><primary>swat</primary></indexterm>
When you have confirmed that SWAT is installed, it is necessary to validate
that the installation includes the binary <literal>swat</literal> file as well
as all the supporting text and Web files. A number of operating system distributions
in the past have failed to include the necessary support files, even though the
<literal>swat</literal> binary executable file was installed. 
</para>

<para>
<indexterm significance="normal"><primary>inetd</primary></indexterm>
<indexterm significance="normal"><primary>xinetd</primary></indexterm>
Finally, when you are sure that SWAT has been fully installed, please check that SWAT
is enabled in the control file for the internetworking super-daemon (inetd or xinetd)
that is used on your operating system platform. 
</para>

<sect3>
<title>Locating the <literal>SWAT</literal> File</title>

<para>
<indexterm significance="normal"><primary>/usr/local/samba/bin</primary></indexterm>
<indexterm significance="normal"><primary>/usr/sbin</primary></indexterm>
<indexterm significance="normal"><primary>/opt/samba/bin</primary></indexterm>
To validate that SWAT is installed, first locate the <literal>swat</literal> binary
file on the system. It may be found under the following directories:</para>
<para><simplelist type="vert">
	<member><filename moreinfo="none">/usr/local/samba/bin</filename> <?latex --- ?> the default Samba location</member>
	<member><filename moreinfo="none">/usr/sbin</filename> <?latex --- ?> the default location on most Linux systems</member>
	<member><filename moreinfo="none">/opt/samba/bin</filename></member>
</simplelist>
</para>

<para>
The actual location is much dependent on the choice of the operating system vendor or as determined
by the administrator who compiled and installed Samba.
</para>

<para>
There are a number of methods that may be used to locate the <literal>swat</literal> binary file.
The following methods may be helpful.
</para>

<para>
<indexterm significance="normal"><primary>swat</primary></indexterm>
<indexterm significance="normal"><primary>operating system search path</primary></indexterm>
<indexterm significance="normal"><primary>swat command-line options</primary></indexterm>
If <literal>swat</literal> is in your current operating system search path, it will be easy to 
find it. You can ask what are the command-line options for <literal>swat</literal> as shown here:
<screen format="linespecific">
frodo:~ # swat -?
Usage: swat [OPTION...]
  -a, --disable-authentication         Disable authentication (demo mode)

Help options:
  -?, --help                           Show this help message
  --usage                              Display brief usage message

Common samba options:
  -d, --debuglevel=DEBUGLEVEL          Set debug level
  -s, --configfile=CONFIGFILE          Use alternative configuration file
  -l, --log-basename=LOGFILEBASE       Basename for log/debug files
  -V, --version                        Print version
</screen>
</para>

</sect3>

<sect3>
<title>Locating the SWAT Support Files</title>

<para>
Now that you have found that <literal>swat</literal> is in the search path, it is easy
to identify where the file is located. Here is another simple way this may be done:
<screen format="linespecific">
frodo:~ # whereis swat
swat: /usr/sbin/swat /usr/share/man/man8/swat.8.gz
</screen>
</para>

<para>
If the above measures fail to locate the <literal>swat</literal> binary, another approach
is needed. The following may be used:
<screen format="linespecific">
frodo:/ # find / -name swat -print
/etc/xinetd.d/swat
/usr/sbin/swat
/usr/share/samba/swat
frodo:/ #
</screen>
</para>

<para>
This list shows that there is a control file for <literal>xinetd</literal>, the internetwork
super-daemon that is installed on this server. The location of the SWAT binary file is
<filename moreinfo="none">/usr/sbin/swat</filename>, and the support files for it are located under the
directory <filename moreinfo="none">/usr/share/samba/swat</filename>.
</para>

<para>
We must now check where <literal>swat</literal> expects to find its support files. This can
be done as follows:
<screen format="linespecific">
frodo:/ # strings /usr/sbin/swat | grep "/swat"
/swat/
...
/usr/share/samba/swat
frodo:/ #
</screen>
</para>

<para>
The <filename moreinfo="none">/usr/share/samba/swat/</filename> entry shown in this listing is the location of the
support files. You should verify that the support files exist under this directory. A sample
list is as shown:
<screen format="linespecific">
jht@frodo:/&gt; find /usr/share/samba/swat -print
/usr/share/samba/swat
/usr/share/samba/swat/help
/usr/share/samba/swat/lang
/usr/share/samba/swat/lang/ja
/usr/share/samba/swat/lang/ja/help
/usr/share/samba/swat/lang/ja/help/welcome.html
/usr/share/samba/swat/lang/ja/images
/usr/share/samba/swat/lang/ja/images/home.gif
...
/usr/share/samba/swat/lang/ja/include
/usr/share/samba/swat/lang/ja/include/header.nocss.html
...
/usr/share/samba/swat/lang/tr
/usr/share/samba/swat/lang/tr/help
/usr/share/samba/swat/lang/tr/help/welcome.html
/usr/share/samba/swat/lang/tr/images
/usr/share/samba/swat/lang/tr/images/home.gif
...
/usr/share/samba/swat/lang/tr/include
/usr/share/samba/swat/lang/tr/include/header.html
/usr/share/samba/swat/using_samba
...
/usr/share/samba/swat/images
/usr/share/samba/swat/images/home.gif
...
/usr/share/samba/swat/include
/usr/share/samba/swat/include/footer.html
/usr/share/samba/swat/include/header.html
jht@frodo:/&gt;
</screen>
</para>

<para>
If the files needed are not available, it is necessary to obtain and install them
before SWAT can be used.
</para>

</sect3>
</sect2>

<sect2 id="xinetd">
<title>Enabling SWAT for Use</title>

<para>
SWAT should be installed to run via the network super-daemon. Depending on which system
your UNIX/Linux system has, you will have either an <literal>inetd</literal>- or
<literal>xinetd</literal>-based system.
</para>

<para>
The nature and location of the network super-daemon varies with the operating system
implementation. The control file (or files) can be located in the file 
<filename moreinfo="none">/etc/inetd.conf</filename> or in the directory <filename moreinfo="none">/etc/[x]inet[d].d</filename>
or in a similar location.
</para>

<para>
The control entry for the older style file might be:
<indexterm significance="normal"><primary>swat</primary><secondary>enable</secondary></indexterm>
</para>


<para><programlisting format="linespecific">
	# swat is the Samba Web Administration Tool
	swat stream tcp nowait.400 root /usr/sbin/swat swat
</programlisting></para>

<para>
A control file for the newer style xinetd could be:
</para>

<para>
<programlisting format="linespecific">
# default: off
# description: SWAT is the Samba Web Admin Tool. Use swat \
#              to configure your Samba server. To use SWAT, \
#              connect to port 901 with your favorite web browser.
service swat
{
	port    = 901
	socket_type     = stream
	wait    = no
	only_from = localhost
	user    = root
	server  = /usr/sbin/swat
	log_on_failure  += USERID
	disable = no
}
</programlisting>
In the above, the default setting for <parameter moreinfo="none">disable</parameter> is <constant>yes</constant>.
This means that SWAT is disabled. To enable use of SWAT, set this parameter to <constant>no</constant>
as shown.
</para>

<para>
<indexterm significance="normal"><primary>swat</primary></indexterm>
<indexterm significance="normal"><primary>/usr/sbin</primary></indexterm>
<indexterm significance="normal"><primary>/usr/share/samba/swat</primary></indexterm>
<indexterm significance="normal"><primary>/usr/local/samba/swat</primary></indexterm>
Both of the previous examples assume that the <literal>swat</literal> binary has been
located in the <filename moreinfo="none">/usr/sbin</filename> directory. In addition to the above,
SWAT will use a directory access point from which it will load its Help files
as well as other control information. The default location for this on most Linux
systems is in the directory <filename moreinfo="none">/usr/share/samba/swat</filename>. The default
location using Samba defaults will be <filename moreinfo="none">/usr/local/samba/swat</filename>.
</para>

<para>
<indexterm significance="normal"><primary>SWAT permission allowed</primary></indexterm>
<indexterm significance="normal"><primary>password change facility</primary></indexterm>
Access to SWAT will prompt for a logon. If you log onto SWAT as any non-root user,
the only permission allowed is to view certain aspects of configuration as well as
access to the password change facility. The buttons that will be exposed to the non-root
user are <guibutton moreinfo="none">HOME</guibutton>, <guibutton moreinfo="none">STATUS</guibutton>, <guibutton moreinfo="none">VIEW</guibutton>, and 
<guibutton moreinfo="none">PASSWORD</guibutton>. The only page that allows
change capability in this case is <guibutton moreinfo="none">PASSWORD</guibutton>.
</para>

<para>
As long as you log onto SWAT as the user <emphasis>root</emphasis>, you should obtain
full change and commit ability. The buttons that will be exposed include
<guibutton moreinfo="none">HOME</guibutton>, <guibutton moreinfo="none">GLOBALS</guibutton>, <guibutton moreinfo="none">SHARES</guibutton>, <guibutton moreinfo="none">PRINTERS</guibutton>, 
<guibutton moreinfo="none">WIZARD</guibutton>, <guibutton moreinfo="none">STATUS</guibutton>, <guibutton moreinfo="none">VIEW</guibutton>, and <guibutton moreinfo="none">PASSWORD</guibutton>.
</para>

</sect2>

<sect2>
<title>Securing SWAT through SSL</title>


<para>
<indexterm significance="normal"><primary>SSL</primary></indexterm>
<indexterm significance="normal"><primary>swat</primary><secondary>security</secondary></indexterm>
Many people have asked about how to set up SWAT with SSL to allow for secure remote
administration of Samba. Here is a method that works, courtesy of Markus Krieger.
</para>

<para>
Modifications to the SWAT setup are as follows: 
</para>

<procedure>
	<step performance="required"><para>
<indexterm significance="normal"><primary>OpenSSL</primary></indexterm>
	Install OpenSSL.
	</para></step>

	<step performance="required"><para>
<indexterm significance="normal"><primary>certificate</primary></indexterm>
<indexterm significance="normal"><primary>private key</primary></indexterm>
	Generate certificate and private key.
<indexterm significance="normal"><primary>/usr/bin/openssl</primary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">/usr/bin/openssl req -new -x509 -days 365 -nodes -config \
	/usr/share/doc/packages/stunnel/stunnel.cnf \
	-out /etc/stunnel/stunnel.pem -keyout /etc/stunnel/stunnel.pem</userinput>
</screen></para></step>

	<step performance="required"><para>
	Remove SWAT entry from [x]inetd.
	</para></step>

	<step performance="required"><para>
<indexterm significance="normal"><primary>stunnel</primary></indexterm>
	Start <literal>stunnel</literal>.

<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">stunnel -p /etc/stunnel/stunnel.pem -d 901 \
	 -l /usr/local/samba/bin/swat swat </userinput>
</screen></para></step>
</procedure>

<para>
Afterward, simply connect to SWAT by using the URL <ulink noescape="1" url="https://myhost:901">https://myhost:901</ulink>, accept the certificate, and the SSL connection is up.
</para>

</sect2>

<sect2>
<title>Enabling SWAT Internationalization Support</title>

<para>
SWAT can be configured to display its messages to match the settings of
the language configurations of your Web browser. It will be passed to SWAT 
in the Accept-Language header of the HTTP request.
</para>

<para>
To enable this feature:
</para>

<itemizedlist>
	<listitem><para>
	Install  the proper <literal>msg</literal> files from the Samba
	<filename moreinfo="none">source/po</filename> directory into $LIBDIR.
	</para></listitem>

	<listitem><para>
	Set your browsers language setting.
	</para></listitem>
</itemizedlist>

<para>
<indexterm significance="normal"><primary>msg file</primary></indexterm>
<indexterm significance="normal"><primary>Japanese</primary></indexterm>
<indexterm significance="normal"><primary>French</primary></indexterm>
<indexterm significance="normal"><primary>English</primary></indexterm>
The name of the <literal>msg</literal> file is the same as the language ID sent by the browser. For
example, <emphasis>en</emphasis> means English, <emphasis>ja</emphasis> means Japanese, <emphasis>fr</emphasis> means French.
</para>

<para>
<indexterm significance="normal"><primary>locale</primary></indexterm>
If you do not like some of messages, or there are no <literal>msg</literal> files for
your locale, you can create them simply by copying the <literal>en.msg</literal> files
to the directory for <quote>your language ID.msg</quote> and filling in proper strings
to each <quote>msgstr</quote>. For example, in <filename moreinfo="none">it.msg</filename>, the
<literal>msg</literal> file for the Italian locale, just set:
<screen format="linespecific">
msgid "Set Default"
msgstr "Imposta Default"
</screen>
<indexterm significance="normal"><primary>msg</primary></indexterm>
and so on. If you find a mistake or create a new <literal>msg</literal> file, please email it
to us so we will consider it in the next release of Samba. The <literal>msg</literal> file should be encoded in UTF-8.
</para>

<para>
<indexterm significance="normal"><primary>UTF-8 encoding</primary></indexterm>
Note that if you enable this feature and the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="DISPLAYCHARSET" xlink:href="smb.conf.5.html#DISPLAYCHARSET">display charset</link> is not
matched to your browser's setting, the SWAT display may be corrupted.  In a future version of
Samba, SWAT will always display messages with UTF-8 encoding. You will then not need to set
this <filename moreinfo="none">smb.conf</filename> file parameter.
</para>

</sect2>

</sect1>

<sect1>
<title>Overview and Quick Tour</title>

<para>
SWAT is a tool that may be used to configure Samba or just to obtain useful links
to important reference materials such as the contents of this book as well as other
documents that have been found useful for solving Windows networking problems.
</para>

<sect2>
<title>The SWAT Home Page</title>

<para>
The SWAT title page provides access to the latest Samba documentation. The manual page for
each Samba component is accessible from this page, as are the Samba3-HOWTO (this 
document) as well as the O'Reilly book <quote>Using Samba.</quote>
</para>

<para>
Administrators who wish to validate their Samba configuration may obtain useful information
from the man pages for the diagnostic utilities. These are available from the SWAT home page
also. One diagnostic tool that is not mentioned on this page but that is particularly
useful is <ulink url="http://www.ethereal.com/"><literal>ethereal</literal></ulink>.
</para>

<warning><para>
SWAT can be configured to run in <emphasis>demo</emphasis> mode. This is not recommended
because it runs SWAT without authentication and with full administrative ability. It allows
changes to <filename moreinfo="none">smb.conf</filename> as well as general operation with root privileges. The option that
creates this ability is the <option>-a</option> flag to SWAT. <emphasis>Do not use this in a
production environment.</emphasis>
</para></warning>

</sect2>

<sect2>
<title>Global Settings</title>

<para>
The <guibutton moreinfo="none">GLOBALS</guibutton> button exposes a page that allows configuration of the global parameters
in <filename moreinfo="none">smb.conf</filename>. There are two levels of exposure of the parameters:
</para>

<itemizedlist>
	<listitem><para>
	<guibutton moreinfo="none">Basic</guibutton> <?latex --- ?> exposes common configuration options.
	</para></listitem>

	<listitem><para>
	<guibutton moreinfo="none">Advanced</guibutton> <?latex --- ?> exposes configuration options needed in more 
	complex environments.
	</para></listitem>
</itemizedlist>

<para>
To switch to other than <guibutton moreinfo="none">Basic</guibutton> editing ability, click on <guibutton moreinfo="none">Advanced</guibutton>.
You may also do this by clicking on the radio button, then click on the <guibutton moreinfo="none">Commit Changes</guibutton> button.
</para>

<para>
After making any changes to configuration parameters, make sure that
you click on the 
<guibutton moreinfo="none">Commit Changes</guibutton> button before moving to another area; otherwise,
your changes will be lost.
</para>

<note><para>
SWAT has context-sensitive help. To find out what each parameter is
for, simply click on the
<guibutton moreinfo="none">Help</guibutton> link to the left of the configuration parameter.
</para></note>

</sect2>

<sect2>
<title>Share Settings</title>

<para>
To affect a currently configured share, simply click on the pull-down button between the
<guibutton moreinfo="none">Choose Share</guibutton> and the <guibutton moreinfo="none">Delete Share</guibutton> buttons and
select the share you wish to operate on. To edit the settings,
click on the
<guibutton moreinfo="none">Choose Share</guibutton> button. To delete the share, simply press the
<guibutton moreinfo="none">Delete Share</guibutton> button.
</para>

<para>
To create a new share, next to the button labeled <guibutton moreinfo="none">Create Share</guibutton>, enter
into the text field the name of the share to be created, then click on the 
<guibutton moreinfo="none">Create Share</guibutton> button.
</para>

</sect2>

<sect2>
<title>Printers Settings</title>

<para>
To affect a currently configured printer, simply click on the pull-down button between the
<guibutton moreinfo="none">Choose Printer</guibutton> and the <guibutton moreinfo="none">Delete Printer</guibutton> buttons and
select the printer you wish to operate on. To edit the settings,
click on the
<guibutton moreinfo="none">Choose Printer</guibutton> button. To delete the share, simply press the
<guibutton moreinfo="none">Delete Printer</guibutton> button.
</para>

<para>
To create a new printer, next to the button labeled <guibutton moreinfo="none">Create Printer</guibutton>, enter
into the text field the name of the share to be created, then click on the 
<guibutton moreinfo="none">Create Printer</guibutton> button.
</para>

</sect2>

<sect2>
<title>The SWAT Wizard</title>

<para>
The purpose of the SWAT Wizard is to help the Microsoft-knowledgeable network administrator
to configure Samba with a minimum of effort.
</para>

<para>
The Wizard page provides a tool for rewriting the <filename moreinfo="none">smb.conf</filename> file in fully optimized format.
This will also happen if you press the <guibutton moreinfo="none">Commit</guibutton> button. The two differ
because the <guibutton moreinfo="none">Rewrite</guibutton> button ignores any changes that may have been made,
while the <guibutton moreinfo="none">Commit</guibutton> button causes all changes to be affected.
</para>

<para>
The <guibutton moreinfo="none">Edit</guibutton> button permits the editing (setting) of the minimal set of
options that may be necessary to create a working Samba server.
</para>

<para>
Finally, there are a limited set of options that determine what type of server Samba
will be configured for, whether it will be a WINS server, participate as a WINS client, or
operate with no WINS support. By clicking one button, you can elect to expose (or not) user
home directories.
</para>

</sect2>

<sect2>
<title>The Status Page</title>

<para>
The status page serves a limited purpose. First, it allows control of the Samba daemons.
The key daemons that create the Samba server environment are <application moreinfo="none">smbd</application>, <application moreinfo="none">nmbd</application>, and <application moreinfo="none">winbindd</application>.
</para>

<para>
The daemons may be controlled individually or as a total group. Additionally, you may set
an automatic screen refresh timing. As MS Windows clients interact with Samba, new smbd processes
are continually spawned. The auto-refresh facility allows you to track the changing
conditions with minimal effort.
</para>

<para>
Finally, the status page may be used to terminate specific smbd client connections in order to
free files that may be locked.
</para>

</sect2>

<sect2>
<title>The View Page</title>

<para>
The view page allows you to view the optimized <filename moreinfo="none">smb.conf</filename> file and, if you are
particularly masochistic, permits you also to see all possible global configuration
parameters and their settings.
</para>

</sect2>

<sect2>
<title>The Password Change Page</title>

<para>
The password change page is a popular tool that allows the creation, deletion, deactivation,
and reactivation of MS Windows networking users on the local machine. You can also use
this tool to change a local password for a user account.
</para>

<para>
When logged in as a non-root account, the user must provide the old password as well as
the new password (twice). When logged in as <emphasis>root</emphasis>, only the new password is
required.
</para>

<para>
One popular use for this tool is to change user passwords across a range of remote MS Windows
servers.
</para>

</sect2>
</sect1>

</chapter>
	<?latex \cleardoublepage ?>

</part>

<part id="troubleshooting">
<title>Troubleshooting</title>

	<?latex \cleardoublepage ?>
	<chapter id="diagnosis">
<chapterinfo>
	
<author>
	<firstname>Andrew</firstname><surname>Tridgell</surname>
	<affiliation>
		<orgname>Samba Team</orgname>
		<address format="linespecific"><email>tridge@samba.org</email></address>
	</affiliation>
</author>
	<author>
<firstname>Jelmer</firstname><othername>R.</othername><surname>Vernooij</surname><othername>R.</othername>
<affiliation>
	<orgname>The Samba Team</orgname>
	<address format="linespecific"><email>jelmer@samba.org</email></address>
</affiliation></author>
	<author>
<firstname>Dan</firstname><surname>Shearer</surname>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>dan@samba.org</email></address>
</affiliation>
</author>
	<pubdate>Wed Jan 15</pubdate>
</chapterinfo>

<title>The Samba Checklist</title>

<sect1>
<title>Introduction</title>

<para>
<indexterm significance="normal"><primary>validate</primary></indexterm>
This file contains a list of tests you can perform to validate your
Samba server. It also tells you what the likely cause of the problem
is if it fails any one of these steps. If it passes all these tests,
then it is probably working fine.
</para>

<para>
You should do all the tests in the order shown. We have tried to
carefully choose them so later tests only use capabilities verified in
the earlier tests. However, do not stop at the first error: there
have been some instances when continuing with the tests has helped
to solve a problem.
</para>

<para>
If you send one of the Samba mailing lists  an email saying, <quote>It does not work,</quote>
and you have not followed this test procedure, you should not be surprised
if your email is ignored.
</para>

</sect1>

<sect1>
<title>Assumptions</title>

<para>
In all of the tests, it is assumed you have a Samba server called 
BIGSERVER and a PC called ACLIENT, both in workgroup TESTGROUP.
</para>

<para>
The procedure is similar for other types of clients.
</para>

<para>
It is also assumed you know the name of an available share in your
<filename moreinfo="none">smb.conf</filename>. I for our examples this share is called <parameter>tmp</parameter>.
You can add a <parameter>tmp</parameter> share like this by adding the
lines shown in <link linkend="tmpshare">the next example</link>.
</para>

<example id="tmpshare">
<title>smb.conf with [tmp] Share</title>
<simplelist>
<member> </member><member><parameter>[tmp]</parameter></member>
<member><indexterm>comment</indexterm><parameter>comment = temporary files </parameter></member>
<member><indexterm>path</indexterm><parameter>path = /tmp</parameter></member>
<member><indexterm>read only</indexterm><parameter>read only = yes</parameter></member>
</simplelist>
</example>

<note><para>
These tests assume version 3.0.0 or later of the Samba suite.
Some commands shown did not exist in earlier versions. 
</para></note>

<para>
<indexterm significance="normal"><primary>error messages</primary></indexterm>
<indexterm significance="normal"><primary>name resolution</primary></indexterm>
<indexterm significance="normal"><primary>/etc/resolv.conf</primary></indexterm>
Please pay attention to the error messages you receive. If any error message
reports that your server is being unfriendly, you should first check that your
IP name resolution is correctly set up. Make sure your <filename moreinfo="none">/etc/resolv.conf</filename>
file points to name servers that really do exist.
</para>

<para>
<indexterm significance="normal"><primary>DNS server access</primary></indexterm>
<indexterm significance="normal"><primary>name resolution</primary></indexterm>
<indexterm significance="normal"><primary>dns proxy</primary></indexterm>
<indexterm significance="normal"><primary>testparm</primary></indexterm>
Also, if you do not have DNS server access for name resolution, please check
that the settings for your <filename moreinfo="none">smb.conf</filename> file results in <parameter moreinfo="none">dns proxy = no</parameter>. The
best way to check this is with <literal>testparm smb.conf</literal>.
</para>


<para>
<indexterm significance="normal"><primary>log files</primary></indexterm>
<indexterm significance="normal"><primary>tail</primary></indexterm>
<indexterm significance="normal"><primary>/usr/local/samba/var</primary></indexterm>
<indexterm significance="normal"><primary>/var/log/samba</primary></indexterm>
<indexterm significance="normal"><primary>log files</primary><secondary>monitoring</secondary></indexterm>
It is helpful to monitor the log files during testing by using the
<literal>tail -F log_file_name</literal> in a separate
terminal console (use ctrl-alt-F1 through F6 or multiple terminals in X). 
Relevant log files can be found (for default installations) in
<filename moreinfo="none">/usr/local/samba/var</filename>. Also, connection logs from
machines can be found here or possibly in <filename moreinfo="none">/var/log/samba</filename>,
depending on how or if you specified logging in your <filename moreinfo="none">smb.conf</filename> file.
</para>

<para>
If you make changes to your <filename moreinfo="none">smb.conf</filename> file while going through these test,
remember to restart <application moreinfo="none">smbd</application> and <application moreinfo="none">nmbd</application>.
</para>

</sect1>

<sect1>
<title>The Tests</title>
<procedure>
<title>Diagnosing Your Samba Server</title>


<step performance="required">
<para>
<indexterm significance="normal"><primary>testparm</primary></indexterm>
In the directory in which you store your <filename moreinfo="none">smb.conf</filename> file, run the command
<literal>testparm smb.conf</literal>. If it reports any errors, then your <filename moreinfo="none">smb.conf</filename>
configuration file is faulty.
</para>

<note><para>
<indexterm significance="normal"><primary>/etc/samba</primary></indexterm>
<indexterm significance="normal"><primary>/usr/local/samba/lib</primary></indexterm>
Your <filename moreinfo="none">smb.conf</filename> file may be located in <filename moreinfo="none">/etc/samba</filename>
or in <filename moreinfo="none">/usr/local/samba/lib</filename>.
</para></note>
</step>

<step performance="required">
<para>
<indexterm significance="normal"><primary>ping</primary></indexterm>
Run the command <literal>ping BIGSERVER</literal> from the PC and
<literal>ping ACLIENT</literal> from the UNIX box. If you do not get a valid response,
then your TCP/IP software is not correctly installed. 
</para>

<para>
You will need to start a <quote>DOS prompt</quote> window on the PC to run ping.
</para>

<para>
<indexterm significance="normal"><primary>/etc/hosts</primary></indexterm>
<indexterm significance="normal"><primary>DNS</primary></indexterm>
<indexterm significance="normal"><primary>/etc/resolv.conf</primary></indexterm>
If you get a message saying <quote><errorname>host not found</errorname></quote> or a similar message, then
your DNS software or <filename moreinfo="none">/etc/hosts</filename> file is not correctly set up.  If using DNS, check that
the <filename moreinfo="none">/etc/resolv.conf</filename> has correct, current, entries in it. It is possible to run
Samba without DNS entries for the server and client, but it is assumed you do have correct entries for the
remainder of these tests.
</para>

<para>
<indexterm significance="normal"><primary>firewall</primary></indexterm>
<indexterm significance="normal"><primary>iptables</primary></indexterm>
<indexterm significance="normal"><primary>ipchains</primary></indexterm>
Another reason why ping might fail is if your host is running firewall 
software. You will need to relax the rules to let in the workstation
in question, perhaps by allowing access from another subnet (on Linux
this is done via the appropriate firewall maintenance commands <literal>ipchains</literal>
or <literal>iptables</literal>).
</para>

<note>
<para>
Modern Linux distributions install ipchains/iptables by default. 
This is a common problem that is often overlooked.
</para>
</note>

<para>
<indexterm significance="normal"><primary>iptables</primary></indexterm>
<indexterm significance="normal"><primary>ipchains</primary></indexterm>
If you wish to check what firewall rules may be present in a system under test, simply run
<literal>iptables -L -v</literal>, or if <parameter moreinfo="none">ipchains</parameter>-based firewall rules are in use,
<literal>ipchains -L -v</literal>.
</para>

<para>
Here is a sample listing from a system that has an external Ethernet interface (eth1) on which Samba
is not active and an internal (private network) interface (eth0) on which Samba is active:
<screen format="linespecific">
frodo:~ # iptables -L -v
Chain INPUT (policy DROP 98496 packets, 12M bytes)
 pkts bytes target     prot opt in     out     source     destination
 187K  109M ACCEPT     all  --  lo     any     anywhere   anywhere
 892K  125M ACCEPT     all  --  eth0   any     anywhere   anywhere
1399K 1380M ACCEPT     all  --  eth1   any     anywhere   anywhere  \
					state RELATED,ESTABLISHED

Chain FORWARD (policy DROP 0 packets, 0 bytes)
 pkts bytes target     prot opt in     out     source     destination
 978K 1177M ACCEPT     all  --  eth1   eth0    anywhere   anywhere \
					state RELATED,ESTABLISHED
 658K   40M ACCEPT     all  --  eth0   eth1    anywhere   anywhere
    0     0 LOG        all  --  any    any     anywhere   anywhere \
					LOG level warning

Chain OUTPUT (policy ACCEPT 2875K packets, 1508M bytes)
 pkts bytes target     prot opt in     out     source     destination

Chain reject_func (0 references)
 pkts bytes target     prot opt in     out     source     destination
</screen>
</para>

</step>

<step performance="required">
<para>
Run the command <literal>smbclient -L BIGSERVER</literal>
on the UNIX box. You should get back a list of available shares. 
</para>

<para>
<indexterm significance="normal"><primary>bad password</primary></indexterm>
<indexterm significance="normal"><primary>hosts allow</primary></indexterm>
<indexterm significance="normal"><primary>hosts deny</primary></indexterm>
<indexterm significance="normal"><primary>valid users</primary></indexterm>
<indexterm significance="normal"><primary>guest account</primary></indexterm>
<indexterm significance="normal"><primary>invalid users</primary></indexterm>
If you get an error message containing the string <quote>bad password</quote>, then
you probably have either an incorrect <parameter moreinfo="none">hosts allow</parameter>, 
<parameter moreinfo="none">hosts deny</parameter>, or <parameter moreinfo="none">valid users</parameter> line in your 
<filename moreinfo="none">smb.conf</filename>, or your guest account is not valid. Check what your guest account is using <application moreinfo="none">testparm</application> and
temporarily remove any <parameter moreinfo="none">hosts allow</parameter>, <parameter moreinfo="none">hosts deny</parameter>,
<parameter moreinfo="none">valid users</parameter>, or <parameter moreinfo="none">invalid users</parameter> lines.
</para>

<para>
<indexterm significance="normal"><primary>inetd.conf</primary></indexterm>
If you get a message <literal moreinfo="none">connection refused</literal> response, then the <literal>smbd</literal> server may
not be running. If you installed it in <filename moreinfo="none">inetd.conf</filename>, then you probably edited
that file incorrectly. If you installed it as a daemon, then check that
it is running and check that the netbios-ssn port is in a LISTEN
state using <literal>netstat -a</literal>.
</para>

<note><para>
<indexterm significance="normal"><primary>inetd</primary></indexterm>
<indexterm significance="normal"><primary>xinetd</primary><see>inetd</see></indexterm>
Some UNIX/Linux systems use <literal>xinetd</literal> in place of
<literal>inetd</literal>. Check your system documentation for the location
of the control files for your particular system implementation of
the network super daemon.
</para></note>

<para>
If you get a message saying <literal moreinfo="none">session request failed,</literal> the server refused the
connection. If it says <quote>Your server software is being unfriendly,</quote> then
it's probably because you have invalid command line parameters to <application moreinfo="none">smbd</application>,
or a similar fatal problem with the initial startup of <application moreinfo="none">smbd</application>. Also
check your config file (<filename moreinfo="none">smb.conf</filename>) for syntax errors with <application moreinfo="none">testparm</application>
and that the various directories where Samba keeps its log and lock
files exist.
</para>

<para>
There are a number of reasons for which smbd may refuse or decline
a session request. The most common of these involve one or more of
the <filename moreinfo="none">smb.conf</filename> file entries as shown in <link linkend="modif1">the next example</link>.
</para>


<example id="modif1">
<title>Configuration for Allowing Connections Only from a Certain Subnet</title>
<simplelist>
<member> </member><member><parameter>[globals]</parameter></member>
<member><indexterm>hosts deny</indexterm><parameter>hosts deny = ALL</parameter></member>
<member><indexterm>hosts allow</indexterm><parameter>hosts allow = xxx.xxx.xxx.xxx/yy</parameter></member>
<member><indexterm>interfaces</indexterm><parameter>interfaces = eth0</parameter></member>
<member><indexterm>bind interfaces only</indexterm><parameter>bind interfaces only = Yes</parameter></member>
</simplelist>
</example>

<para>
<indexterm significance="normal"><primary>loopback adapter</primary></indexterm>
In <link linkend="modif1">Configuration for Allowing Connections Only from a Certain Subnet</link>, no
allowance has been made for any session requests that will automatically translate to the loopback adapter
address 127.0.0.1.  To solve this problem, change these lines as shown in <link linkend="modif2">the following
example</link>.
</para>

<example id="modif2">
<title>Configuration for Allowing Connections from a Certain Subnet and localhost</title>
<simplelist>
<member> </member><member><parameter>[globals]</parameter></member>
<member><indexterm>hosts deny</indexterm><parameter>hosts deny = ALL</parameter></member>
<member><indexterm>hosts allow</indexterm><parameter>hosts allow = xxx.xxx.xxx.xxx/yy 127.</parameter></member>
<member><indexterm>interfaces</indexterm><parameter>interfaces = eth0 lo</parameter></member>
</simplelist>
</example>

<para>
<indexterm significance="normal"><primary>inetd</primary></indexterm>
<indexterm significance="normal"><primary>smbclient</primary></indexterm>
Another common cause of these two errors is having something already running on port <constant>139</constant>,
such as Samba (<application moreinfo="none">smbd</application> is running from <application moreinfo="none">inetd</application> already) or Digital's Pathworks. Check
your <filename moreinfo="none">inetd.conf</filename> file before trying to start <application moreinfo="none">smbd</application> as a daemon <?latex --- ?> it can avoid a
lot of frustration!
</para>

<para>
<indexterm significance="normal"><primary>subnet mask</primary></indexterm>
<indexterm significance="normal"><primary>broadcast address</primary></indexterm>
<indexterm significance="normal"><primary>log.nmbd</primary></indexterm>
<indexterm significance="normal"><primary>network interface</primary></indexterm>
<indexterm significance="normal"><primary>IP address</primary></indexterm>
And yet another possible cause for failure of this test is when the subnet mask and/or broadcast address
settings are incorrect. Please check that the network interface IP address/broadcast address/subnet mask
settings are correct and that Samba has correctly noted these in the <filename moreinfo="none">log.nmbd</filename> file.
</para>

</step>

<step performance="required">

<para>
<indexterm significance="normal"><primary>nmblookup</primary></indexterm>
Run the command <literal>nmblookup -B BIGSERVER __SAMBA__</literal>.
You should get back the IP address of your Samba server.
</para>

<para>
<indexterm significance="normal"><primary>inetd.conf</primary></indexterm>
<indexterm significance="normal"><primary>nmbd</primary></indexterm>
<indexterm significance="normal"><primary>port 137</primary></indexterm>
If you do not, then <application moreinfo="none">nmbd</application> is incorrectly installed. Check your <filename moreinfo="none">inetd.conf</filename>
if you run it from there, or that the daemon is running and listening to UDP port 137.
</para>

<para>
One common problem is that many inetd implementations can't take many
parameters on the command line. If this is the case, then create a
one-line script that contains the right parameters and run that from
inetd.
</para>

</step>

<step performance="required">

<para>
<indexterm significance="normal"><primary>nmblookup</primary></indexterm>
Run the command <literal>nmblookup -B ACLIENT `*'</literal>.
</para>

<para>
You should get the PC's IP address back. If you do not, then the client
software on the PC isn't installed correctly, or isn't started, or you
got the name of the PC wrong. 
</para>

<para>
If ACLIENT does not resolve via DNS, then use the IP address of the
client in the above test.
</para>

</step>

<step performance="required">

<para>
Run the command <literal>nmblookup -d 2 `*'</literal>.
</para>

<para>
This time we are trying the same as the previous test but are trying
it via a broadcast to the default broadcast address. A number of
NetBIOS/TCP/IP hosts on the network should respond, although Samba may
not catch all of the responses in the short time it listens. You
should see the <literal moreinfo="none">got a positive name query response</literal>
messages from several hosts.
</para>

<para>
<indexterm significance="normal"><primary>nmblookup</primary></indexterm>
If this does not give a result similar to the previous test, then nmblookup isn't correctly getting your
broadcast address through its automatic mechanism. In this case you should experiment with the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="INTERFACES" xlink:href="smb.conf.5.html#INTERFACES">interfaces</link> option in <filename moreinfo="none">smb.conf</filename> to manually configure your IP address, broadcast, and netmask.
</para>

<para>
If your PC and server aren't on the same subnet, then you will need to use the
<option>-B</option> option to set the broadcast address to that of the PC's subnet.
</para>

<para>
This test will probably fail if your subnet mask and broadcast address are
not correct. (Refer to test 3 notes above).
</para>

</step>

<step performance="required">


<para>
<indexterm significance="normal"><primary>smbclient</primary></indexterm>
Run the command <literal>smbclient //BIGSERVER/TMP</literal>. You should 
then be prompted for a password. You should use the password of the account
with which you are logged into the UNIX box. If you want to test with
another account, then add the <option>-U accountname</option> option to the end of
the command line <?latex --- ?> for example, <literal>smbclient //bigserver/tmp -Ujohndoe</literal>.
</para>

<note><para>
It is possible to specify the password along with the username as follows:
<literal>smbclient //bigserver/tmp -Ujohndoe%secret</literal>.
</para></note>

<para>
Once you enter the password, you should get the <prompt moreinfo="none">smb&gt;</prompt> prompt. If you
do not, then look at the error message. If it says <quote><errorname>invalid network
name,</errorname></quote> then the service <parameter>tmp</parameter> is not correctly set up in your <filename moreinfo="none">smb.conf</filename>.
</para>

<para>
If it says <quote><errorname>bad password,</errorname></quote> then the likely causes are:
</para>

<orderedlist inheritnum="ignore" continuation="restarts">
<listitem>
	<para>
	You have shadow passwords (or some other password system) but didn't
	compile in support for them in <application moreinfo="none">smbd</application>.
	</para>
</listitem>

<listitem>
	<para>
	Your <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="VALIDUSERS" xlink:href="smb.conf.5.html#VALIDUSERS">valid users</link> configuration is incorrect.
	</para>
</listitem>

<listitem>
	<para>
	You have a mixed-case password and you haven't enabled the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PASSWORDLEVEL" xlink:href="smb.conf.5.html#PASSWORDLEVEL">password level</link> option at a high enough level.
	</para>
</listitem>

<listitem>
	<para>
	The <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PATH" xlink:href="smb.conf.5.html#PATH">path</link> line in <filename moreinfo="none">smb.conf</filename> is incorrect. Check it with <application moreinfo="none">testparm</application>.
	</para>
</listitem>

<listitem>
	<para>
	You enabled password encryption but didn't map UNIX to Samba users. Run
	<literal>smbpasswd -a username</literal>
	</para>
</listitem>
</orderedlist>

<para>
<indexterm significance="normal"><primary>dir</primary></indexterm>
<indexterm significance="normal"><primary>get</primary></indexterm>
<indexterm significance="normal"><primary>put</primary></indexterm>
<indexterm significance="normal"><primary>help command</primary></indexterm>
Once connected, you should be able to use the commands <literal>dir</literal>, <literal>get</literal>,
<literal>put</literal>, and so on. Type <literal>help command</literal> for instructions. You should
especially check that the amount of free disk space shown is correct when you type <literal>dir</literal>.
</para>

</step>

<step performance="required">

<para>
<indexterm significance="normal"><primary>net view</primary></indexterm>
On the PC, type the command <literal>net view \\BIGSERVER</literal>. You will 
need to do this from within a DOS prompt window. You should get back a 
list of shares available on the server.
</para>

<para>
<indexterm significance="normal"><primary>nmbd</primary></indexterm>
If you get a message <literal moreinfo="none">network name not found</literal> or similar error, then NetBIOS
name resolution is not working. This is usually caused by a problem in <literal>nmbd</literal>.
To overcome it, you could do one of the following (you only need to choose one of them):
</para>

<orderedlist inheritnum="ignore" continuation="restarts">
<listitem><para>
	Fix the <application moreinfo="none">nmbd</application> installation.
</para></listitem>

<listitem><para>
	Add the IP address of BIGSERVER to the <literal>wins server</literal> box in the
	advanced TCP/IP setup on the PC.
</para></listitem>

<listitem><para>
	Enable Windows name resolution via DNS in the advanced section of the TCP/IP setup.
</para></listitem>

<listitem><para>
	Add BIGSERVER to your lmhosts file on the PC.
</para></listitem>
</orderedlist>

<para>
If you get a message <quote><errorname>invalid network name</errorname></quote> or 
<quote><errorname>bad password error,</errorname></quote> then apply the
same fixes as for the <literal>smbclient -L</literal> test. In
particular, make sure your <literal>hosts allow</literal> line is correct (see the man pages).
</para>

<para>
Also, do not overlook that fact that when the workstation requests the
connection to the Samba server, it will attempt to connect using the 
name with which you logged onto your Windows machine. You need to make
sure that an account exists on your Samba server with that exact same
name and password.
</para>

<para>
If you get a message <quote><errorname>specified computer is not receiving requests</errorname></quote> or similar error,
it probably means that the host is not contactable via TCP services.
Check to see if the host is running TCP wrappers, and if so, add an entry in
the <filename moreinfo="none">hosts.allow</filename> file for your client (or subnet, and so on.)
</para>

</step>

<step performance="required">

<para>
Run the command <literal>net use x: \\BIGSERVER\TMP</literal>. You should 
be prompted for a password, then you should get a <computeroutput moreinfo="none">command completed 
successfully</computeroutput> message. If not, then your PC software is incorrectly 
installed or your <filename moreinfo="none">smb.conf</filename> is incorrect. Make sure your <parameter moreinfo="none">hosts allow</parameter>
and other config lines in <filename moreinfo="none">smb.conf</filename> are correct.
</para>

<para>
It's also possible that the server can't work out what username to connect you as.
To see if this is the problem, add the line
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="USER" xlink:href="smb.conf.5.html#USER">user = username</link> to the
<parameter>[tmp]</parameter> section of 
<filename moreinfo="none">smb.conf</filename> where <parameter moreinfo="none">username</parameter> is the
username corresponding to the password you typed. If you find this
fixes things, you may need the username mapping option. 
</para>

<para>
It might also be the case that your client only sends encrypted passwords 
and you have <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="ENCRYPTPASSWORDS" xlink:href="smb.conf.5.html#ENCRYPTPASSWORDS">encrypt passwords = no</link> in <filename moreinfo="none">smb.conf</filename>.
Change this setting to `yes' to fix this.
</para>

</step>

<step performance="required">

<para>
Run the command <literal>nmblookup -M <parameter moreinfo="none">testgroup</parameter></literal> where 
<parameter moreinfo="none">testgroup</parameter> is the name of the workgroup that your Samba server and 
Windows PCs belong to. You should get back the IP address of the 
master browser for that workgroup.
</para>

<para>
If you do not, then the election process has failed. Wait a minute to
see if it is just being slow, then try again. If it still fails after
that, then look at the browsing options you have set in <filename moreinfo="none">smb.conf</filename>. Make
sure you have <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PREFERREDMASTER" xlink:href="smb.conf.5.html#PREFERREDMASTER">preferred master = yes</link> to ensure that 
an election is held at startup.
</para>

</step>

<step performance="required">

<para>
From file manager, try to browse the server. Your Samba server should
appear in the browse list of your local workgroup (or the one you
specified in <filename moreinfo="none">smb.conf</filename>). You should be able to double-click on the name
of the server and get a list of shares. If you get the error message <quote>invalid password,</quote>
 you are probably running Windows NT and it
is refusing to browse a server that has no encrypted password
capability and is in user-level security mode. In this case, either set
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SECURITY" xlink:href="smb.conf.5.html#SECURITY">security = server</link> and 
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PASSWORDSERVER" xlink:href="smb.conf.5.html#PASSWORDSERVER">password server = Windows_NT_Machine</link> in your
<filename moreinfo="none">smb.conf</filename> file or make sure <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="ENCRYPTPASSWORDS" xlink:href="smb.conf.5.html#ENCRYPTPASSWORDS">encrypt passwords</link> is
set to <quote>yes</quote>.
</para>

</step>
</procedure>
</sect1>

</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="problems">

<chapterinfo>
	<author>
<firstname>Gerald</firstname><surname>Carter</surname><othername>(Jerry)</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jerry@samba.org</email></address>
</affiliation></author>
	<author>
<firstname>Jelmer</firstname><othername>R.</othername><surname>Vernooij</surname><othername>R.</othername>
<affiliation>
	<orgname>The Samba Team</orgname>
	<address format="linespecific"><email>jelmer@samba.org</email></address>
</affiliation></author>
	
<author>
	<firstname>David</firstname><surname>Bannon</surname>
	<affiliation>
		<orgname>Samba Team</orgname>
		<address format="linespecific"><email>dbannon@samba.org</email></address>
	</affiliation>
</author>
	<author>
<firstname>Dan</firstname><surname>Shearer</surname>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>dan@samba.org</email></address>
</affiliation>
</author>
	<pubdate>8 Apr 2003</pubdate>
</chapterinfo>

<title>Analyzing and Solving Samba Problems</title>

<para>
<indexterm significance="normal"><primary>RFCs</primary></indexterm>
<indexterm significance="normal"><primary>SMB</primary></indexterm>
<indexterm significance="normal"><primary>documentation</primary></indexterm>
There are many sources of information available in the form of mailing lists, RFCs, and documentation. The
documentation that comes with the Samba distribution contains good explanations of general SMB topics such as
browsing.
</para> 
	
<sect1>
<title>Diagnostics Tools</title>

<para>
<indexterm significance="normal"><primary>sniffer</primary></indexterm>
<indexterm significance="normal"><primary>LAN</primary></indexterm>
<indexterm significance="normal"><primary>analyzes data</primary></indexterm>
<indexterm significance="normal"><primary>SMB networking</primary></indexterm>
<indexterm significance="normal"><primary>network analyzer</primary></indexterm>
With SMB networking, it is often not immediately clear what the cause is of a certain problem. Samba itself
provides rather useful information, but in some cases you might have to fall back to using a
<emphasis>sniffer</emphasis>. A sniffer is a program that listens on your LAN, analyzes the data sent on it,
and displays it on the screen.
</para>

<sect2>
<title>Debugging with Samba Itself</title>

<para>
<indexterm significance="normal"><primary>diagnostic tools</primary></indexterm>
<indexterm significance="normal"><primary>debugging problems</primary></indexterm>
<indexterm significance="normal"><primary>smbd</primary></indexterm>
<indexterm significance="normal"><primary>nmbd</primary></indexterm>
<indexterm significance="normal"><primary>debugging passwords</primary></indexterm>
<indexterm significance="normal"><primary>debug level</primary></indexterm>
<indexterm significance="normal"><primary>log level</primary></indexterm>
One of the best diagnostic tools for debugging problems is Samba itself.  You can use the <option>-d
option</option> for both <application moreinfo="none">smbd</application> and <application moreinfo="none">nmbd</application> to specify the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="DEBUGLEVEL" xlink:href="smb.conf.5.html#DEBUGLEVEL">debug level</link> at which to run.
See the man pages for <literal>smbd, nmbd</literal>, and <filename moreinfo="none">smb.conf</filename> for more information regarding debugging
options. The debug level (log level) can range from 1 (the default) to 10 (100 for debugging passwords).
</para>
	
<para>
<indexterm significance="normal"><primary>debugging</primary></indexterm>
<indexterm significance="normal"><primary>gcc</primary></indexterm>
<indexterm significance="normal"><primary>gdb</primary></indexterm>
<indexterm significance="normal"><primary>smbd</primary></indexterm>
<indexterm significance="normal"><primary>nmbd</primary></indexterm>
<indexterm significance="normal"><primary>LsaEnumTrustedDomains</primary></indexterm>
<indexterm significance="normal"><primary>attach gdb</primary></indexterm>
Another helpful method of debugging is to compile Samba using the <literal>gcc -g </literal> flag. This will
include debug information in the binaries and allow you to attach <literal>gdb</literal> to the running
<literal>smbd/nmbd</literal> process.  To attach <literal>gdb</literal> to an <literal>smbd</literal> process
for an NT workstation, first get the workstation to make the connection. Pressing ctrl-alt-delete and going
down to the domain box is sufficient (at least, the first time you join the domain) to generate a
<parameter moreinfo="none">LsaEnumTrustedDomains</parameter>. Thereafter, the workstation maintains an open connection and
there will be an smbd process running (assuming that you haven't set a really short smbd idle timeout). So, in
between pressing <literal>ctrl-alt-delete</literal> and actually typing in your password, you can attach
<literal>gdb</literal> and continue.
</para>

<para>
Some useful Samba commands worth investigating are:
<indexterm significance="normal"><primary>testparm</primary></indexterm>
<indexterm significance="normal"><primary>smbclient</primary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">$ </prompt><userinput moreinfo="none">testparm | more</userinput>
<prompt moreinfo="none">$ </prompt><userinput moreinfo="none">smbclient -L //{netbios name of server}</userinput>
</screen>
</para>

</sect2>

<sect2>
	<title>Tcpdump</title>
  
<para>
<indexterm significance="normal"><primary>tcpdump</primary></indexterm>
<indexterm significance="normal"><primary>tethereal</primary></indexterm>
<indexterm significance="normal"><primary>ethereal</primary></indexterm>
<ulink url="http://www.tcpdump.org/">Tcpdump</ulink> was the first 
UNIX sniffer with SMB support. It is a command-line utility and 
now, its SMB support is somewhat lagging that of <literal>ethereal</literal> 
and <literal>tethereal</literal>.
</para>

</sect2>

<sect2>
	<title>Ethereal</title>

<para>
<indexterm significance="normal"><primary>ethereal</primary></indexterm>
<ulink url="http://www.ethereal.com/">Ethereal</ulink> is a graphical sniffer, available for both UNIX (Gtk)
and Windows. Ethereal's SMB support is quite good. For details on the use of <literal>ethereal</literal>, read
the well-written Ethereal User Guide.
</para>

<figure id="ethereal1" float="0"><title>Starting a Capture.</title><mediaobject><imageobject role="html"><imagedata fileref="images/ethereal1.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/ethereal1.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/ethereal1"/></imageobject></mediaobject></figure>

<para>
<indexterm significance="normal"><primary>ports</primary></indexterm>
Listen for data on ports 137, 138, 139, and 445. For example, use the filter <userinput moreinfo="none">port 137, port 138,
port 139, or port 445</userinput> as seen in <link linkend="ethereal1">Starting a Capture</link> snapshot.
</para>

<para>
A console version of ethereal is available as well and is called <literal>tethereal</literal>.
</para>

<figure id="ethereal2" float="0"><title>Main Ethereal Data Window.</title><mediaobject><imageobject role="html"><imagedata fileref="images/ethereal2.png" scale="100" scalefit="1"/></imageobject><imageobject><imagedata fileref="images/ethereal2.png" scale="50" scalefit="1"/></imageobject><imageobject role="latex"><imagedata fileref="Samba3-HOWTO/images/ethereal2"/></imageobject></mediaobject></figure>

</sect2>

<sect2>
<title>The Windows Network Monitor</title>
	
<para>
<indexterm significance="normal"><primary>Network Monitor</primary></indexterm>
<indexterm significance="normal"><primary>Netmon</primary></indexterm>
<indexterm significance="normal"><primary>Microsoft Developer Network CDs</primary></indexterm>
<indexterm significance="normal"><primary>SMS</primary></indexterm>
<indexterm significance="normal"><primary>promiscuous mode</primary></indexterm>
<indexterm significance="normal"><primary>ethereal</primary></indexterm>
For tracing things on Microsoft Windows NT, Network Monitor (aka Netmon) is available on Microsoft Developer
Network CDs, the Windows NT Server install CD, and the SMS CDs. The version of Netmon that ships with SMS
allows for dumping packets between any two computers (i.e., placing the network interface in promiscuous
mode).  The version on the NT Server install CD will only allow monitoring of network traffic directed to the
local NT box and broadcasts on the local subnet. Be aware that Ethereal can read and write Netmon formatted
files.
</para>

<sect3>
<title>Installing Network Monitor on an NT Workstation</title>

<para>
<indexterm significance="normal"><primary>Netmon.</primary></indexterm>
Installing Netmon on an NT workstation requires a couple of steps. The following are instructions for
installing Netmon V4.00.349, which comes with Microsoft Windows NT Server 4.0, on Microsoft Windows NT
Workstation 4.0. The process should be similar for other versions of Windows NT version of Netmon. You will
need both the Microsoft Windows NT Server 4.0 Install CD and the Workstation 4.0 Install CD.
</para> 

<para>
<indexterm significance="normal"><primary>Network Monitor Tools and Agent</primary></indexterm>
Initially you will need to install <application moreinfo="none">Network Monitor Tools and Agent</application>
on the NT Server to do this: 
</para>

<itemizedlist>
	<listitem><para>Go to <guibutton moreinfo="none">Start</guibutton> -&gt; <guibutton moreinfo="none">Settings</guibutton> -&gt; <guibutton moreinfo="none">Control Panel</guibutton> -&gt; 
	<guibutton moreinfo="none">Network</guibutton> -&gt; <guibutton moreinfo="none">Services</guibutton> -&gt; <guibutton moreinfo="none">Add</guibutton>.</para></listitem>
	
	<listitem><para>Select the <guilabel moreinfo="none">Network Monitor Tools and Agent</guilabel> and click on <guibutton moreinfo="none">OK</guibutton>.</para></listitem> 
	
	<listitem><para>Click on <guibutton moreinfo="none">OK</guibutton> on the Network Control Panel.</para></listitem> 
	
	<listitem><para>Insert the Windows NT Server 4.0 install CD when prompted.</para></listitem> 
</itemizedlist>

<para>
At this point, the Netmon files should exist in <filename moreinfo="none">%SYSTEMROOT%\System32\netmon\*.*</filename>.   
Two subdirectories exist as well: <filename moreinfo="none">parsers\</filename>, which contains the necessary DLLs
for parsing the Netmon packet dump, and <filename moreinfo="none">captures\</filename>.
</para>

<para>
To install the Netmon tools on an NT Workstation, you will first need to install the 
Network  Monitor Agent from the Workstation install CD.
</para>

<itemizedlist>
	<listitem><para>Go to <guibutton moreinfo="none">Start</guibutton> -&gt; <guibutton moreinfo="none">Settings</guibutton> -&gt;
		<guibutton moreinfo="none">Control Panel</guibutton> -&gt; <guibutton moreinfo="none">Network</guibutton> -&gt;
		<guibutton moreinfo="none">Services</guibutton> -&gt; <guibutton moreinfo="none">Add</guibutton>.</para></listitem> 

	<listitem><para>Select the <guilabel moreinfo="none">Network Monitor Agent</guilabel>, click on
		<guibutton moreinfo="none">OK</guibutton>.</para></listitem> 
	
	<listitem><para>Click on <guibutton moreinfo="none">OK</guibutton> in the Network Control Panel.
	</para></listitem> 
	
	<listitem><para>Insert the Windows NT Workstation 4.0 install CD when prompted.</para></listitem> 
</itemizedlist>

<para>
Now copy the files from the NT Server in <filename moreinfo="none">%SYSTEMROOT%\System32\netmon</filename>
to <filename moreinfo="none">%SYSTEMROOT%\System32\netmon</filename> on the workstation and set permissions
as you deem appropriate for your site. You will need administrative rights on the NT box to run Netmon.
</para>

</sect3>
<sect3>
<title>Installing Network Monitor on Windows 9x/Me</title>
<para>
To install Netmon on Windows 9x/Me, install the Network Monitor Agent 
from the Windows 9x/Me CD (<filename moreinfo="none">\admin\nettools\netmon</filename>). 
There is a readme file included with the Netmon driver files on the CD if you need 
information on how to do this. Copy the files from a working Netmon installation.
</para>
</sect3>
</sect2>
</sect1>

<sect1>
<title>Useful URLs</title>
<itemizedlist>

<listitem><para>See how Scott Merrill simulates a BDC behavior at 
       <ulink noescape="1" url="http://www.skippy.net/linux/smb-howto.html">
       http://www.skippy.net/linux/smb-howto.html</ulink>. </para></listitem>

<listitem><para>FTP site for older SMB specs, 
       <ulink noescape="1" url="ftp://ftp.microsoft.com/developr/drg/CIFS/">
       ftp://ftp.microsoft.com/developr/drg/CIFS/</ulink></para></listitem>.

</itemizedlist>

</sect1>

<sect1>
<title>Getting Mailing List Help</title>

<para>
There are a number of Samba-related mailing lists. Go to <ulink noescape="1" url="http://samba.org">http://samba.org</ulink>, click on your nearest mirror,
and then click on <literal>Support</literal>. Next, click on <literal>
Samba-related mailing lists</literal>.
</para>

<para>
For questions relating to Samba TNG, go to
<ulink noescape="1" url="http://www.samba-tng.org/">http://www.samba-tng.org/</ulink>. 
It has been requested that you do not post questions about Samba-TNG to the
mainstream Samba lists.</para>

<para>
If you do post a message to one of the lists, please observe the following guidelines:
</para>

<itemizedlist>

	<listitem><para>
<indexterm significance="normal"><primary>volunteers</primary></indexterm>
	Always remember that the developers are volunteers; they are
	not paid and they never guarantee to produce a particular feature at 
	a particular time. Any timelines are <quote>best guess,</quote> and nothing more.
	</para></listitem>

	<listitem><para>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
	Always mention what version of Samba you are using and what 
	operating system it's running under. You should list the relevant sections of
	your <filename moreinfo="none">smb.conf</filename> file, at least the options in <parameter>[global]</parameter>
	that affect PDC support.
	</para></listitem>

	<listitem><para>In addition to the version, if you obtained Samba via
	CVS, mention the date when you last checked it out.</para></listitem>

	<listitem><para> Try to make your questions clear and brief. Lots of long, 
	convoluted questions get deleted before	they are completely read!
	Do not post HTML-encoded messages. Most people on mailing lists simply delete
	them.
	</para></listitem>

	<listitem><para> If you run one of those nifty <quote>I'm on holiday</quote> things when
	you are away, make sure its configured to not answer mailing list traffic. Autoresponses
	to mailing lists really irritate the thousands of people who end up having to deal
	with such bad netiquet bahavior.
	</para></listitem> 

	<listitem><para>
<indexterm significance="normal"><primary>cross post</primary></indexterm>
	Don't cross post. Work out which is the best list to post to 
	and see what happens. Do not post to both samba-ntdom and samba-technical.
	Many people active on the lists subscribe to more 
	than one list and get annoyed to see the same message two or more times.
	Often someone who thinks a message would be better dealt 
	with on another list will forward it on for you.</para></listitem>

	<listitem><para>You might include <emphasis>partial</emphasis>
	log files written at a log level set to as much as 20. 
	Please do not send the entire log but just enough to give the context of the
	error messages.</para></listitem>

	<listitem><para>If you have a complete Netmon trace (from the opening of
	the pipe to the error), you can send the *.CAP file as well.</para></listitem> 
	    
	<listitem><para>Please think carefully before attaching a document to an email.
	Consider pasting the relevant parts into the body of the message. The Samba
	mailing lists go to a huge number of people. Do they all need a copy of your 
	<filename moreinfo="none">smb.conf</filename> in their attach directory?</para></listitem>

</itemizedlist>

</sect1>

<sect1>
<title>How to Get Off the Mailing Lists</title>

<para>To have your name removed from a Samba mailing list, go to the same
place where you went to
subscribe to it, go to <ulink noescape="1" url="http://lists.samba.org/">http://lists.samba.org</ulink>, 
click on your nearest mirror, click on <literal>Support</literal>, and 
then click on <literal>Samba-related mailing lists</literal>. 
</para>

<para>
Please do not post messages to the list asking to be removed. You will only
be referred to the above address (unless that process failed in some way).
</para>

</sect1>

</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="bugreport">

<chapterinfo>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
	<author>
<firstname>Jelmer</firstname><othername>R.</othername><surname>Vernooij</surname><othername>R.</othername>
<affiliation>
	<orgname>The Samba Team</orgname>
	<address format="linespecific"><email>jelmer@samba.org</email></address>
</affiliation></author>
	
<author>
	<firstname>Andrew</firstname><surname>Tridgell</surname>
	<affiliation>
		<orgname>Samba Team</orgname>
		<address format="linespecific"><email>tridge@samba.org</email></address>
	</affiliation>
</author>
	<pubdate> 27 June 1997 </pubdate>
</chapterinfo>

<title>Reporting Bugs</title>

<sect1>
<title>Introduction</title>

<para>
<indexterm significance="normal"><primary>Bugzilla</primary></indexterm>
<indexterm significance="normal"><primary>bug reports</primary></indexterm>
Please report bugs using Samba's <ulink url="https://bugzilla.samba.org/">Bugzilla</ulink> facilities and take
the time to read this file before you submit a bug report. Also, check to see if it has changed between
releases, as we may be changing the bug reporting mechanism at some point.
</para>

<para>
Please do as much as you can yourself to help track down the
bug. Samba is maintained by a dedicated group of people who volunteer
their time, skills, and efforts. We receive far more mail than
we can possibly answer, so you have a much higher chance of a response
and a fix if you send us a <quote>developer-friendly</quote> bug report that lets
us fix it fast. 
</para>

<para>
<indexterm significance="normal"><primary>comp.protocols.smb</primary></indexterm>
<indexterm significance="normal"><primary>newsgroup</primary></indexterm>
<indexterm significance="normal"><primary>configuration problem</primary></indexterm>
If you post the bug to the comp.protocols.smb
newsgroup or the mailing list, do not assume that we will read it. If you suspect that your 
problem is not a bug but a configuration problem, it is better to send 
it to the Samba mailing list, as there are thousands of other users on
that list who may be able to help you.
</para>

<para>
You may also like to look though the recent mailing list archives,
which are conveniently accessible on the Samba Web pages
at <ulink noescape="1" url="http://samba.org/samba/">http://samba.org/samba/</ulink>.
</para>

</sect1>

<sect1>
<title>General Information</title>

<para>
Before submitting a bug report, check your config for silly
errors. Look in your log files for obvious messages that tell
you've misconfigured something. Run testparm to check your config
file for correct syntax.
</para>

<para>
Have you looked through <link linkend="diagnosis">The Samba Checklist</link>? This is extremely important.
</para>

<para>
If you include part of a log file with your bug report, then be sure to
annotate it with exactly what you were doing on the client at the
time and exactly what the results were.
</para>

</sect1>

<sect1 id="dbglvl">
<title>Debug Levels</title>

<para>
If the bug has anything to do with Samba behaving incorrectly as a
server (like refusing to open a file), then the log files will probably
be quite useful. Depending on the problem, a log level of between 3 and
10 showing the problem may be appropriate. A higher level gives more
detail but may use too much disk space.
</para>

<para>
<indexterm significance="normal"><primary>debug level</primary></indexterm>
<indexterm significance="normal"><primary>log level</primary></indexterm>
To set the debug level, use the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOGLEVEL" xlink:href="smb.conf.5.html#LOGLEVEL">log level</link> in your 
<filename moreinfo="none">smb.conf</filename>. You may also find it useful to set the log 
level higher for just one machine and keep separate logs for each machine. 
To do this, add the following lines to your main <filename moreinfo="none">smb.conf</filename> file:
</para>

<simplelist>
<member><indexterm>log level</indexterm><parameter>log level = 10</parameter></member>
<member><indexterm>log file</indexterm><parameter>log file = /usr/local/samba/lib/log.%m</parameter></member>
<member><indexterm>include</indexterm><parameter>include = /usr/local/samba/lib/smb.conf.%m</parameter></member>
</simplelist>

<para>
and create a file <filename moreinfo="none">/usr/local/samba/lib/smb.conf.<replaceable>machine</replaceable></filename> where
<replaceable>machine</replaceable> is the name of the client you wish to debug. In that file put any
<filename moreinfo="none">smb.conf</filename> commands you want; for example, <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOGLEVEL" xlink:href="smb.conf.5.html#LOGLEVEL">log level</link> may be useful. This also allows
you to experiment with different security systems, protocol levels, and so on, on just one machine.
</para>

<para>
The <filename moreinfo="none">smb.conf</filename> entry <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOGLEVEL" xlink:href="smb.conf.5.html#LOGLEVEL">log level</link> is synonymous with the parameter <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="DEBUGLEVEL" xlink:href="smb.conf.5.html#DEBUGLEVEL">debuglevel</link> that has been used in older versions of Samba and is being retained for backward
compatibility of <filename moreinfo="none">smb.conf</filename> files.
</para>

<para>
As the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="LOGLEVEL" xlink:href="smb.conf.5.html#LOGLEVEL">log level</link> value is increased, you will record a significantly greater level of
debugging information. For most debugging operations, you may not need a setting higher than
<constant>3</constant>. Nearly all bugs can be tracked at a setting of <constant>10</constant>, but be
prepared for a large volume of log data.
</para>

	<sect2>
	<title>Debugging-Specific Operations</title>

	<para>
<indexterm significance="normal"><primary>debugging</primary></indexterm>
<indexterm significance="normal"><primary>logging</primary></indexterm>
<indexterm significance="normal"><primary>functional components</primary></indexterm>
<indexterm significance="normal"><primary>cluttering</primary></indexterm>
	Samba-3.x permits debugging (logging) of specific functional components without unnecessarily
	cluttering the log files with detailed logs for all operations. An example configuration to 
	achieve this is shown in:
	</para>

<para>
<simplelist>
<member><indexterm>log level</indexterm><parameter>log level = 0 tdb:3 passdb:5 auth:4 vfs:2</parameter></member>
<member><indexterm>max log size</indexterm><parameter>max log size = 0</parameter></member>
<member><indexterm>log file</indexterm><parameter>log file = /var/log/samba/%U.%m.log</parameter></member>
</simplelist>
</para>

	<para>
	This will cause the level of detail to be expanded to the debug class (log level) passed to
	each functional area per the value shown above. The first value passed to the <parameter moreinfo="none">log level</parameter>
	of <constant>0</constant> means turn off all unnecessary debugging except the debug classes set for
	the functional areas as specified. The table shown in <link linkend="dbgclass">Debuggable Functions</link>
	may be used to attain very precise analysis of each SMB operation Samba is conducting.
	</para>

	<table frame="all" id="dbgclass">
		<title>Debuggable Functions</title>
	<tgroup cols="2" align="center">
		<thead>
		<row><entry>Function Name</entry><entry>Function Name</entry></row>
		</thead>
		<tbody>
		<row><entry>all</entry><entry>passdb</entry></row>
		<row><entry>tdb</entry><entry>sam</entry></row>
		<row><entry>printdrivers</entry><entry>auth</entry></row>
		<row><entry>lanman</entry><entry>winbind</entry></row>
		<row><entry>smb</entry><entry>vfs</entry></row>
		<row><entry>rpc_parse</entry><entry>idmap</entry></row>
		<row><entry>rpc_srv</entry><entry>quota</entry></row>
		<row><entry>rpc_cli</entry><entry>acls</entry></row>
		</tbody>
	</tgroup>
	</table>

	</sect2>

</sect1>

<sect1>
<title>Internal Errors</title>

<para>
If you get the message <quote><errorname>INTERNAL ERROR</errorname></quote> in your log files, 
it means that Samba got an unexpected signal while running. It is probably a
segmentation fault and almost certainly means a bug in Samba (unless
you have faulty hardware or system software).
</para>

<para>
If the message came from smbd, it will probably be accompanied by
a message that details the last SMB message received by smbd. This
information is often useful in tracking down the problem, so please
include it in your bug report.
</para>

<para>
You should also detail how to reproduce the problem, if
possible. Please make this reasonably detailed.
</para>


<para>
<indexterm significance="normal"><primary>core files</primary></indexterm>
You may also find that a core file appeared in a <filename moreinfo="none">corefiles</filename>
subdirectory of the directory where you keep your Samba log
files. This file is the most useful tool for tracking down the bug. To
use it, you do this:
<indexterm significance="normal"><primary>gdb</primary></indexterm>
<indexterm significance="normal"><primary>debug</primary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">$ </prompt><userinput moreinfo="none">gdb smbd core</userinput>
</screen>
</para>

<para>
<indexterm significance="normal"><primary>dbx</primary></indexterm>
<indexterm significance="normal"><primary>stack trace</primary></indexterm>
adding appropriate paths to smbd and core so gdb can find them. If you
do not have gdb, try <userinput moreinfo="none">dbx</userinput>. Then within the debugger,
use the command <literal>where</literal> to give a stack trace of where the
problem occurred. Include this in your report.
</para>

<para>
<indexterm significance="normal"><primary>disass</primary></indexterm>
If you know any assembly language, do a <literal>disass</literal> of the routine
where the problem occurred (if it's in a library routine, then
disassemble the routine that called it) and try to work out exactly
where the problem is by looking at the surrounding code. Even if you
do not know assembly, including this information in the bug report can be
useful. 
</para>
</sect1>

<sect1>
<title>Attaching to a Running Process</title>

<para>
<indexterm significance="normal"><primary>PID</primary></indexterm>
<indexterm significance="normal"><primary>gdb</primary></indexterm>
<indexterm significance="normal"><primary>smbstatus</primary></indexterm>
Unfortunately, some UNIXes (in particular some recent Linux kernels)
refuse to dump a core file if the task has changed UID (which smbd
does often). To debug with this sort of system, you could try to attach
to the running process using
<userinput moreinfo="none">gdb smbd <replaceable>PID</replaceable></userinput>, where you get
<replaceable>PID</replaceable> from <application moreinfo="none">smbstatus</application>.
Then use <literal>c</literal> to continue and try to cause the core dump
using the client. The debugger should catch the fault and tell you
where it occurred.
</para>

<para>
Sometimes it is necessary to build Samba binary files that have debugging
symbols so as to make it possible to capture enough information from a crashed
operation to permit the Samba Team to fix the problem.
</para>

<para>
Compile with <constant>-g</constant> to ensure you have symbols in place. 
Add the following line to the <filename moreinfo="none">smb.conf</filename> file global section:
<screen format="linespecific">
panic action = "/bin/sleep 90000"
</screen>
to catch any panics. If <literal>smbd</literal> seems to be frozen, look for any sleep
processes. If it is not, and appears to be spinning, find the PID
of the spinning process and type:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> gdb /usr/local/samba/sbin/smbd
</screen>
<indexterm significance="normal"><primary>spinning process</primary></indexterm>
then <quote>attach `pid'</quote> (of the spinning process), then type <quote>bt</quote> to
get a backtrace to see where the smbd is in the call path.
</para>

</sect1>

<sect1>
<title>Patches</title>


<para>
<indexterm significance="normal"><primary>diff</primary></indexterm>
<indexterm significance="normal"><primary>patch</primary></indexterm>
The best sort of bug report is one that includes a fix! If you send us
patches, please use <userinput moreinfo="none">diff -u</userinput> format if your version of 
diff supports it; otherwise, use <userinput moreinfo="none">diff -c4</userinput>. Make sure 
you do the diff against a clean version of the source and let me know 
exactly what version you used. 
</para>

</sect1>
</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="tdb">

<chapterinfo>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
	
	<pubdate>May 28, 2008</pubdate>
</chapterinfo>

<title>Managing TDB Files</title>

<sect1>
<title>Features and Benefits</title>

	<para>
<indexterm significance="normal"><primary>tdb</primary></indexterm>
<indexterm significance="normal"><primary>Trivial Database</primary></indexterm>
	Samba uses a lightweight database called Trivial Database (tdb) in which it stores persistent and transient data.
	Some tdb files can be disposed of before restarting Samba, but others are used to store information that is vital
	to Samba configuration and behavior. The following information is provided to help administrators who are seeking
	to better manage their Samba installations.
	</para>

	<para>
<indexterm significance="normal"><primary>tdb</primary></indexterm>
<indexterm significance="normal"><primary>currupted</primary></indexterm>
<indexterm significance="normal"><primary>backup</primary></indexterm>
<indexterm significance="normal"><primary>restore</primary></indexterm>
	Those who package Samba for commercial distribution with operating systems and appliances would do well to take
	note that tdb files can get corrupted, and for this reason ought to be backed up regularly.  An appropriate time
	is at system shutdown (backup) and startup (restore from backup).
	</para>

<table frame="all" id="TOSH-TDB">
                <title>Samba's Trivial Database Files</title>
                <tgroup align="center" cols="2">
                <thead>
                <row><entry>File name</entry><entry>Preserve</entry><entry>Description</entry></row>
                </thead>

                <tbody>
                        <row><entry>account_policy.tdb</entry><entry>Y</entry>
				<entry><para>NT account policy settings such as pw expiration, etc...</para></entry></row>
                        <row><entry>brlock.tdb</entry><entry>N</entry>
				<entry><para>Byte range locks.</para></entry></row>
                        <row><entry>browse.dat</entry><entry>N</entry>
				<entry><para>Browse lists - gets rebuilt automatically.</para></entry></row>
                        <row><entry>connections.tdb</entry><entry>N</entry>
				<entry><para>Share connections. Used to enforce max connections, etc.</para></entry></row>
                        <row><entry>gencache.tdb</entry><entry>N</entry>
				<entry><para>Generic caching database.</para></entry></row>
                        <row><entry>group_mapping.tdb</entry><entry>Y</entry>
				<entry><para>Stores group mapping information. Not used when using LDAP backend.</para></entry></row>
                        <row><entry>  lang_en.tdb</entry><entry>Y</entry>
				<entry><para>Stores language encoding information.</para></entry></row>
                        <row><entry>locking.tdb</entry><entry>N</entry>
				<entry><para>Stores share mode and oplock information.</para></entry></row>
                        <row><entry>login_cache.tdb</entry><entry>N</entry>
				<entry><para>Keeps a log of bad pw attempts.</para></entry></row>
                        <row><entry>messages.tdb</entry><entry>N</entry>
				<entry><para>Used to keep track of Samba internal messaging.</para></entry></row>
                        <row><entry>netsamlogon_cache.tdb</entry><entry>Y</entry>
				<entry><para>
					Cache of user net_info_3 struct from <emphasis>net_samlogon()</emphasis>
					requests from domain member machines.
				</para></entry></row>
                        <row><entry>ntdrivers.tdb</entry><entry>Y</entry>
				<entry><para>Stores installed printer driver information.</para></entry></row>
                        <row><entry>ntforms.tdb</entry><entry>Y</entry>
				<entry><para>Stores installed printer forms information.</para></entry></row>
                        <row><entry>ntprinters.tdb</entry><entry>Y</entry>
				<entry><para>Stores installed printers information.</para></entry></row>
                        <row><entry>printing directory</entry><entry>Y</entry>
				<entry><para>Directory containing tdb per print queue of cached lpq output.</para></entry></row>
                        <row><entry>registry.tdb</entry><entry>Y</entry>
				<entry><para>Windows registry skeleton (connect via regedit.exe).</para></entry></row>
                        <row><entry>sessionid.tdb</entry><entry>N</entry>
				<entry><para>Session information to support <literal moreinfo="none">utmp = yes</literal> capabilities.</para></entry></row>
                        <row><entry>share_info.tdb</entry><entry>Y</entry>
				<entry><para>Stores share-level ACL configuration settings.
					Default ACL is <emphasis>Everyone - Full Control</emphasis>.
				</para></entry></row>
                        <row><entry>unexpected.tdb</entry><entry>N</entry>
				<entry><para>
					Unexpected packet queue needed to support windows clients that respond on a 
					different port that the originating reques.
				</para></entry></row>
                        <row><entry>winbindd_cache.tdb</entry><entry>N</entry>
				<entry><para>Winbind's cache of user lists.</para></entry></row>
                        <row><entry>winbindd_idmap.tdb</entry><entry>Y</entry>
				<entry><para>Winbind's local IDMAP database.</para></entry></row>
                        <row><entry>wins.dat</entry><entry>N</entry>
				<entry><para>
				WINS database iused only when <parameter moreinfo="none">wins support = yes</parameter>
				has been set. This gets rebuilt or updated at every restart.
				</para></entry></row>
                        <row><entry>wins.tdb</entry><entry>Y</entry>
				<entry><para>
				The working permanent storage for all WINS data. This database is used only
				when <parameter moreinfo="none">wins support = yes</parameter> has been set in the <filename moreinfo="none">smb.conf</filename> file.
				Note: This retains all manually configured WINS entries. Manual setting can be done use the net utility.
				</para></entry></row>
                        <row><entry>secrets.tdb</entry><entry>Y</entry>
				<entry><para>
				This tdb file stores internal settings such as the machine and the domain SID, secret passwords
				that are used with LDAP, the machine secret token, etc. This is an essential file that is stored
				in a secure area. Vendors locate this in various folders. Check <literal>smbd -b</literal> to
				find its location on your system.
				</para></entry></row>
                        <row><entry>schannel_store.tdb</entry><entry>Y</entry>
				<entry><para>
				This stores secure channel access token information used with SMB signing.
				</para></entry></row>
                        <row><entry>passdb.tdb</entry><entry>Y</entry>
				<entry><para>
				This stores the Samba SAM account information when using a tdbsam password backend.
				</para></entry></row>
                </tbody>
        </tgroup>
        </table>

</sect1>

<sect1>
<title>Managing TDB Files</title>

	<para>
	The <literal>tdbbackup</literal> utility is a tool that may be used to backup samba tdb files.
	This tool may also be used to verify the integrity of the tdb files prior to Samba startup or
	during normal operation. If it finds file damage it will search for a prior backup the backup
	file from which the damaged tdb file will be restored. The <literal>tdbbackup</literal>
	utility can safely be run at any time. It was designed so that it can be used at any time to
	validate the integrity of tdb files, even during Samba operation.
	</para>

	<para>
	It is recommended to backup all tdb files as part of the Samba start-up scripts on a Samba
	server. The following command syntax can be used:
	</para>
<screen format="linespecific">
myserver# &gt; cd /var/lib/samba
myserver@ &gt; tdbbackup *.tdb
</screen>
	<para>
	The default extension is <filename moreinfo="none">.bak</filename>. Any alternate extension can be specified
	by executing <literal moreinfo="none">tdbbackup -s 'new_extension' *.tdb</literal> as part of your startup script.
	</para>

</sect1>

</chapter>
	<?latex \cleardoublepage ?>

</part>

<part id="Appendix">
<title>Reference Section</title>

	<?latex \cleardoublepage ?>
	<chapter id="compiling">
<chapterinfo>
	<author>
<firstname>Jelmer</firstname><othername>R.</othername><surname>Vernooij</surname><othername>R.</othername>
<affiliation>
	<orgname>The Samba Team</orgname>
	<address format="linespecific"><email>jelmer@samba.org</email></address>
</affiliation></author>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
	
<author>
	<firstname>Andrew</firstname><surname>Tridgell</surname>
	<affiliation>
		<orgname>Samba Team</orgname>
		<address format="linespecific"><email>tridge@samba.org</email></address>
	</affiliation>
</author>
	
	<pubdate> 22 May 2001 </pubdate>
	<pubdate> 18 March 2003 </pubdate>
	<pubdate> June 2005 </pubdate>
</chapterinfo>

<title>How to Compile Samba</title>

<para>
You can obtain the Samba source file from the
<ulink url="http://samba.org/">Samba Web site</ulink>. To obtain a development version, 
you can download Samba from Subversion or using <literal>rsync</literal>.
</para>

<sect1>
<title>Access Samba Source Code via Subversion</title>


<sect2>
<title>Introduction</title>

<para>
<indexterm significance="normal"><primary>Subversion</primary></indexterm>
Samba is developed in an open environment. Developers use a
Subversion to <quote>checkin</quote> (also known as 
<quote>commit</quote>) new source code. Samba's various Subversion branches can
be accessed via anonymous Subversion using the instructions
detailed in this chapter.
</para>

<para>
This chapter is a modified version of the instructions found at the
<ulink noescape="1" url="http://samba.org/samba/subversion.html">Samba</ulink> Web site.
</para>

</sect2>

<sect2>
<title>Subversion Access to samba.org</title>

<para>
The machine samba.org runs a publicly accessible Subversion
repository for access to the source code of several packages, 
including Samba, rsync, distcc, ccache, and jitterbug. There are two main ways
of accessing the Subversion server on this host.
</para>

<sect3>
<title>Access via ViewCVS</title>


<para>
<indexterm significance="normal"><primary>SVN</primary><secondary>web</secondary></indexterm>
You can access the source code via your favorite WWW browser. This allows you to access
the contents of individual files in the repository and also to look at the revision 
history and commit logs of individual files. You can also ask for a diff 
listing between any two versions on the repository.
</para>

<para>
Use the URL
<ulink noescape="1" url="http://viewcvs.samba.org/">http://viewcvs.samba.org/</ulink>.
</para>
</sect3>

<sect3>
<title>Access via Subversion</title>

<para>
<indexterm significance="normal"><primary>Subversion</primary></indexterm>
You can also access the source code via a normal Subversion client. This gives you much more control over what
you can do with the repository and allows you to check out whole source trees and keep them up to date via
normal Subversion commands. This is the preferred method of access if you are a developer and not just a
casual browser.
</para>

<para>In order to be able to download the Samba sources off Subversion, you need 
a Subversion client. Your distribution might include one, or you can download the 
sources from <ulink noescape="1" url="http://subversion.tigris.org/">http://subversion.tigris.org/</ulink>.
</para>

<para>
To gain access via anonymous Subversion, use the following steps. 
</para>

<procedure>
	<title>Retrieving Samba using Subversion</title>

	<step performance="required">
	<para>
	Install a recent copy of Subversion. All you really need is a 
	copy of the Subversion client binary. 
	</para>
	</step>

	<step performance="required">
	<para>
	Run the command 
	<screen format="linespecific">
	<userinput moreinfo="none">svn co svn://svnanon.samba.org/samba/trunk samba</userinput>.
	</screen>
	</para>
	
	<para>
	This will create a directory called <filename moreinfo="none">samba</filename> containing the 
	latest Samba source code (usually the branch that is going to be the next major release). This 
	currently corresponds to the 3.1 development tree. 
	</para>
	
	<para>
	Subversion branches other then trunk can be obtained by adding branches/BRANCH_NAME to the URL you check
	out. A list of branch names can be found on the <quote>Development</quote> page of the Samba Web site. A
	common request is to obtain the latest 3.0 release code. This could be done by using the following command:
	<screen format="linespecific">
	<userinput moreinfo="none">svn co svn://svnanon.samba.org/samba/branches/SAMBA_3_0 samba_3</userinput>.
	</screen>
	</para>
	</step>

	<step performance="required">
	<para>
	Whenever you want to merge in the latest code changes, use the following command from within the Samba
	directory:
	<screen format="linespecific">
	<userinput moreinfo="none">svn update</userinput>
	</screen>
	</para>
	</step>
</procedure>
	
</sect3>
</sect2>

</sect1>

<sect1>
	<title>Accessing the Samba Sources via rsync and ftp</title>


	<para>
	<indexterm significance="normal"><primary>rsync</primary></indexterm>
	<indexterm significance="normal"><primary>ftp</primary></indexterm>
	<parameter moreinfo="none">pserver.samba.org</parameter> also exports unpacked copies of most parts of the Subversion tree
	at the Samba <ulink noescape="1" url="ftp://pserver.samba.org/pub/unpacked">pserver</ulink> location and also
	via anonymous rsync at the Samba <ulink noescape="1" url="rsync://pserver.samba.org/ftp/unpacked/">rsync</ulink> server location.  I recommend using rsync rather
	than ftp, because rsync is capable of compressing data streams, but it is also more useful than FTP because
	during a partial update it will transfer only the data that is missing plus a small overhead.  See <ulink noescape="1" url="http://rsync.samba.org/">the rsync home page</ulink> for more info on rsync.
	</para>

	<para>
	The disadvantage of the unpacked trees is that they do not support automatic
	merging of local changes as Subversion does. <literal>rsync</literal> access is most convenient 
	for an initial install.                      
	</para>
</sect1>

<sect1>
<title>Verifying Samba's PGP Signature</title>

<para>
<indexterm significance="normal"><primary>GPG</primary></indexterm>
<indexterm significance="normal"><primary>PGP</primary></indexterm>
It is strongly recommended that you verify the PGP signature for any source file before
installing it. Even if you're not downloading from a mirror site, verifying PGP signatures
should be a standard reflex. Many people today use the GNU GPG tool set in place of PGP.
GPG can substitute for PGP.
</para>


<para>
With that said, go ahead and download the following files:
</para>

<para><screen format="linespecific">
<prompt moreinfo="none">$ </prompt><userinput moreinfo="none">wget http://us1.samba.org/samba/ftp/samba-3.0.20.tar.asc</userinput>
<prompt moreinfo="none">$ </prompt><userinput moreinfo="none">wget http://us1.samba.org/samba/ftp/samba-pubkey.asc</userinput>
</screen></para>


<para>
<indexterm significance="normal"><primary>PGP</primary></indexterm>
The first file is the PGP signature for the Samba source file; the other is the Samba public
PGP key itself. Import the public PGP key with:
<screen format="linespecific">
<prompt moreinfo="none">$ </prompt><userinput moreinfo="none">gpg --import samba-pubkey.asc</userinput>
</screen>
and verify the Samba source code integrity with:
<screen format="linespecific">
<prompt moreinfo="none">$ </prompt><userinput moreinfo="none">gzip -d samba-3.0.20.tar.gz</userinput>
<prompt moreinfo="none">$ </prompt><userinput moreinfo="none">gpg --verify samba-3.0.20.tar.asc</userinput>
</screen>
</para>

<para>
If you receive a message like, <quote>Good signature from Samba Distribution Verification Key...,</quote>
then all is well. The warnings about trust relationships can be ignored. An
example of what you would not want to see would be:
<screen format="linespecific">
gpg: BAD signature from <quote>Samba Distribution Verification Key</quote>
</screen>
</para>

</sect1>

<sect1>
	<title>Building the Binaries</title>
	
	<para>
	<indexterm significance="normal"><primary>autogen.sh</primary></indexterm>
<indexterm significance="normal"><primary>configure</primary></indexterm>
	After the source tarball has been unpacked, the next step involves
	configuration to match Samba to your operating system platform.
	If your source directory does not contain the <literal>configure</literal> script,
	it is necessary to build it before you can continue. Building of
	the configure script requires the correct version of the autoconf
	tool kit. Where the necessary version of autoconf is present,
	the configure script can be generated by executing the following:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> cd samba-3.0.20/source
<prompt moreinfo="none">root# </prompt> ./autogen.sh
</screen>
	</para>
	

	<para>
	<indexterm significance="normal"><primary>configure</primary></indexterm>
	To build the binaries, run the program <userinput moreinfo="none">./configure
	</userinput> in the source directory. This should automatically 
	configure Samba for your operating system. If you have unusual 
	needs, then you may wish to first run:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">./configure --help</userinput>
</screen>
</para>
	
	<para>
	This will help you to see what special options can be enabled. Now execute
	<userinput moreinfo="none">./configure</userinput> with any arguments it might need:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">./configure <replaceable>[... arguments ...]</replaceable></userinput>
</screen>
	</para>
	
	<para>
	<indexterm significance="normal"><primary>make</primary></indexterm>
	Execute the following create the binaries:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> <userinput moreinfo="none">make</userinput>
</screen>
	Once it is successfully compiled, you can execute the command shown here to
	install the binaries and manual pages:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> <userinput moreinfo="none">make install</userinput>
</screen>
	</para>
	
	<para>
	Some people prefer to install binary files and man pages separately. If this is
	your wish, the binary files can be installed by executing:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> <userinput moreinfo="none">make installbin</userinput>
</screen>
	The man pages can be installed using this command:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> <userinput moreinfo="none">make installman</userinput>
</screen>
	</para>

	<para>
	Note that if you are upgrading from a previous version of Samba the old
	versions of the binaries will be renamed with an <quote>.old</quote> extension.
	You can go back to the previous version by executing:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> <userinput moreinfo="none">make revert</userinput>
</screen>
	As you can see from this, building and installing Samba does not need to
	result in disaster!
	</para>
	

	<sect2>
	<title>Compiling Samba with Active Directory Support</title>
	
	<para>
	In order to compile Samba with ADS support, you need to have installed
	on your system:
	</para>

	<itemizedlist>
	
	    <listitem><para>
		The MIT or Heimdal Kerberos development libraries
	    (either install from the sources or use a package).
		</para></listitem>
	
	    <listitem><para>
		The OpenLDAP development libraries.
		</para></listitem>
	    
	</itemizedlist>

	<para>
	If your Kerberos libraries are in a nonstandard location, then
	remember to add the configure option
	<option>--with-krb5=<replaceable>DIR</replaceable></option>.
	</para>

	<para>
	After you run configure, make sure that the 
	<filename moreinfo="none">include/config.h</filename> it generates contain lines like this:
<programlisting format="linespecific">
#define HAVE_KRB5 1
#define HAVE_LDAP 1
</programlisting>
	</para>

	<para>
	If it does not, configure did not find your KRB5 libraries or
	your LDAP libraries. Look in <filename moreinfo="none">config.log</filename> to figure
	out why and fix it.
	</para>

	<sect3>
	<title>Installing the Required Packages for Debian</title>

	<para>On Debian, you need to install the following packages:</para>
	<para>
		<itemizedlist>
			<listitem><para>libkrb5-dev</para></listitem>
			<listitem><para>krb5-user</para></listitem>
		</itemizedlist>
	</para>
	</sect3>

	<sect3>
	<title>Installing the Required Packages for Red Hat Linux</title>

	<para>On Red Hat Linux, this means you should have at least: </para>
	<para>
		<itemizedlist>
			<listitem><para>krb5-workstation (for kinit)</para></listitem>
			<listitem><para>krb5-libs (for linking with)</para></listitem>
			<listitem><para>krb5-devel (because you are compiling from source)</para></listitem>
		</itemizedlist>
	</para>

	<para>in addition to the standard development environment.</para>

	<para>If these files are not installed on your system, you should check the installation
	CDs to find which has them and install the files using your tool of choice. If in doubt
	about what tool to use, refer to the Red Hat Linux documentation.</para>

	</sect3>

	<sect3>
	<title>SuSE Linux Package Requirements</title>

	<para>
	SuSE Linux installs Heimdal packages that may be required to allow you to build
	binary packages. You should verify that the development libraries have been installed on
	your system.
	</para>

	<para>
	SuSE Linux Samba RPMs support Kerberos. Please refer to the documentation for
	your SuSE Linux system for information regarding SuSE Linux specific configuration.
	Additionally, SuSE is very active in the maintenance of Samba packages that provide
	the maximum capabilities that are available. You should consider using SuSE-provided
	packages where they are available.
	</para>

	</sect3>
	
	</sect2>
			  
</sect1>

<sect1 id="startingSamba">
	<title>Starting the <application moreinfo="none">smbd</application> <application moreinfo="none">nmbd</application> and <application moreinfo="none">winbindd</application></title>


	<para>
	<indexterm significance="normal"><primary>inetd</primary></indexterm>
	You must choose to start <application moreinfo="none">smbd</application>, <application moreinfo="none">winbindd</application>  and <application moreinfo="none">nmbd</application> either as daemons or from
	<application moreinfo="none">inetd</application>. Don't try to do both!  Either you can put
	them in <filename moreinfo="none"> inetd.conf</filename> and have them started on demand by
	<application moreinfo="none">inetd</application> or <application moreinfo="none">xinetd</application>, or you
	can start them as daemons either from the command-line or in
	<filename moreinfo="none">/etc/rc.local</filename>. See the man pages for details on the
	command line options. Take particular care to read the bit about what user
	you need to have to start Samba. In many cases, you must be root.
	</para>

	<para>
	The main advantage of starting <application moreinfo="none">smbd</application> and <application moreinfo="none">nmbd</application> using the recommended daemon method
	is that they will respond slightly more quickly to an initial connection request.
	</para>

	<sect2>
	<title>Starting from inetd.conf</title>

	<indexterm significance="normal"><primary>inetd</primary></indexterm>
	
	<note>
	<para>The following will be different if 
	you use NIS, NIS+, or LDAP to distribute services maps.</para>
	</note>
	
	<para>Look at your <filename moreinfo="none">/etc/services</filename>. 
	What is defined at port 139/tcp? If nothing is defined, 
	then add a line like this:</para>

	<para><programlisting format="linespecific">netbios-ssn     139/tcp</programlisting></para>

	<para>Similarly for 137/udp, you should have an entry like:</para>

	<para><programlisting format="linespecific">netbios-ns	137/udp</programlisting></para>

	<para>
	Next, edit your <filename moreinfo="none">/etc/inetd.conf</filename> and add two lines like this:
<programlisting format="linespecific">
netbios-ssn stream tcp nowait root /usr/local/samba/sbin/smbd smbd 
netbios-ns dgram udp wait root /usr/local/samba/sbin/nmbd nmbd 
</programlisting>
	</para>

<indexterm significance="normal"><primary>/etc/inetd.conf</primary></indexterm>
	<para>
	The exact syntax of <filename moreinfo="none">/etc/inetd.conf</filename> 
	varies between UNIXes. Look at the other entries in inetd.conf 
	for a guide.
	</para>

	<para>
	<indexterm significance="normal"><primary>xinetd</primary></indexterm>
	Some distributions use xinetd instead of inetd. Consult the 
	xinetd manual for configuration information.
	</para>

	<note><para>Some UNIXes already have entries like netbios_ns 
	(note the underscore) in <filename moreinfo="none">/etc/services</filename>. 
	You must edit <filename moreinfo="none">/etc/services</filename> or
	<filename moreinfo="none">/etc/inetd.conf</filename> to make them consistent.
	</para></note>

	<note><para>
	<indexterm significance="normal"><primary>ifconfig</primary></indexterm>
	On many systems you may need to use the
	<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="INTERFACES" xlink:href="smb.conf.5.html#INTERFACES">interfaces</link> option in <filename moreinfo="none">smb.conf</filename> to specify
	the IP address and netmask of your interfaces. Run 
	<application moreinfo="none">ifconfig</application> as root if you do
	not know what the broadcast is for your net. <application moreinfo="none">nmbd</application> tries
	to determine it at runtime, but fails on some UNIXes. 
	</para></note>

	<warning><para>
	Many UNIXes only accept around five parameters on the command
	line in <filename moreinfo="none">inetd.conf</filename>.  This means you shouldn't
	use spaces between the options and arguments, or you should use
	a script and start the script from <literal>inetd</literal>.
	</para></warning>

	<para>
	Restart <application moreinfo="none">inetd</application>, perhaps just send it a HUP,
	like this:
<indexterm significance="normal"><primary>killall</primary></indexterm>
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt><userinput moreinfo="none">killall -HUP inetd</userinput>
</screen>
	</para>
		
	</sect2>
	
	<sect2>
	<title>Alternative: Starting <application moreinfo="none">smbd</application> as a Daemon</title>
		
	<para>
	<indexterm significance="normal"><primary>daemon</primary></indexterm>
<indexterm significance="normal"><primary>startsmb</primary></indexterm>
	To start the server as a daemon, you should create a script something
	like this one, perhaps calling it <filename moreinfo="none">startsmb</filename>.
	</para>

<para><programlisting format="linespecific">
#!/bin/sh
/usr/local/samba/sbin/smbd -D 
/usr/local/samba/sbin/winbindd -B
/usr/local/samba/sbin/nmbd -D 
</programlisting></para>

	<para>
	Make it executable with <literal>chmod +x startsmb</literal>.
	</para>

	<para>
	You can then run <literal>startsmb</literal> by hand or execute
	it from <filename moreinfo="none">/etc/rc.local</filename>.
	</para>

	<para>
	To kill it, send a kill signal to the processes <application moreinfo="none">nmbd</application> and <application moreinfo="none">smbd</application>.
	</para>

	<note><para>
	If you use the SVR4-style init system, you may like to look at the
	<filename moreinfo="none">examples/svr4-startup</filename> script to make Samba fit
	into that system.
	</para></note>

	<sect3>
	<title>Starting Samba for Red Hat Linux</title>

	<para>
	Red Hat Linux has not always included all Samba components in the standard installation.
	So versions of Red Hat Linux do not install the winbind utility, even though it is present
	on the installation CDROM media. Check to see if the <literal>winbindd</literal> is present
	on the system:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> ls /usr/sbin/winbindd
/usr/sbin/winbindd
</screen>
	This means that the appropriate RPM package was installed. The following response means
	that it is not installed:
<screen format="linespecific">
/bin/ls: /usr/sbin/winbind: No such file or directory
</screen>
	In this case, it should be installed if you intend to use <literal>winbindd</literal>. Search
	the CDROM installation media for the samba-winbind RPM and install it following Red Hat
	guidelines.
	</para>

	<para>
	The process for starting Samba will now be outlined. Be sure to configure Samba's <filename moreinfo="none">smb.conf</filename>
	file before starting Samba. When configured, start Samba by executing:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> service smb start
<prompt moreinfo="none">root# </prompt> service winbind start
</screen>
	These steps will start <application moreinfo="none">nmbd</application>, <application moreinfo="none">smbd</application> and <application moreinfo="none">winbindd</application>.
	</para>

	<para>
	To ensure that these services will be automatically restarted when the system is rebooted
	execute:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> chkconfig smb on
<prompt moreinfo="none">root# </prompt> chkconfig winbind on
</screen>
	Samba will be started automatically at every system reboot.
	</para>

	</sect3>

	<sect3>
	<title>Starting Samba for Novell SUSE Linux</title>

	<para>
	Novell SUSE Linux products automatically install all essential Samba components in a default installation.
	Configure your <filename moreinfo="none">smb.conf</filename> file, then execute the following to start Samba:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> rcnmb start
<prompt moreinfo="none">root# </prompt> rcsmb start
<prompt moreinfo="none">root# </prompt> rcwinbind start
</screen>
	Now execute these commands so that Samba will be started automatically following a system
	reboot:
<screen format="linespecific">
<prompt moreinfo="none">root# </prompt> chkconfig nmb on
<prompt moreinfo="none">root# </prompt> chkconfig smb on
<prompt moreinfo="none">root# </prompt> chkconfig winbind on
</screen>
	The Samba services will now be started automatically following a system reboot.
	</para>

	</sect3>

	</sect2>

</sect1>

</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="Portability">
<chapterinfo>
	<author>
<firstname>Jelmer</firstname><othername>R.</othername><surname>Vernooij</surname><othername>R.</othername>
<affiliation>
	<orgname>The Samba Team</orgname>
	<address format="linespecific"><email>jelmer@samba.org</email></address>
</affiliation></author>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
	<!-- Some other people as well, but there were no author names in the text files this file is based on-->
</chapterinfo>

<title>Portability</title>

<para>
<indexterm significance="normal"><primary>platforms</primary></indexterm>
<indexterm significance="normal"><primary>compatible</primary></indexterm>
Samba works on a wide range of platforms, but the interface all the 
platforms provide is not always compatible. This chapter contains 
platform-specific information about compiling and using Samba.</para>

<sect1>
<title>HPUX</title>

<para>
<indexterm significance="normal"><primary>/etc/logingroup</primary></indexterm>
<indexterm significance="normal"><primary>/etc/group</primary></indexterm>
Hewlett-Packard's implementation of supplementary groups is nonstandard (for
historical reasons). There are two group files, <filename moreinfo="none">/etc/group</filename> and
<filename moreinfo="none">/etc/logingroup</filename>; the system maps UIDs to numbers using the former, but
initgroups() reads the latter. Most system admins who know the ropes
symlink <filename moreinfo="none">/etc/group</filename> to <filename moreinfo="none">/etc/logingroup</filename>
(hard-link does not work for reasons too obtuse to go into here). initgroups() will complain if one of the
groups you're in, in <filename moreinfo="none">/etc/logingroup</filename>, has what it considers to be an invalid
ID, which means outside the range <constant>[0..UID_MAX]</constant>, where <constant>UID_MAX</constant> is
60000 currently on HP-UX. This precludes -2 and 65534, the usual <constant>nobody</constant>
GIDs.
</para>

<para>
If you encounter this problem, make sure the programs that are failing 
to initgroups() are run as users, not in any groups with GIDs outside the 
allowed range.
</para>

<para>
This is documented in the HP manual pages under setgroups(2) and passwd(4).
</para>

<para>
<indexterm significance="normal"><primary>gcc</primary></indexterm>
<indexterm significance="normal"><primary>ANSI compiler</primary></indexterm>
On HP-UX you must use gcc or the HP ANSI compiler. The free compiler
that comes with HP-UX is not ANSI compliant and cannot compile Samba.
</para>

</sect1>

<sect1>
<title>SCO UNIX</title>

<para> 
If you run an old version of SCO UNIX, you may need to get important 
TCP/IP patches for Samba to work correctly. Without the patch, you may 
encounter corrupt data transfers using Samba.
</para>

<para>
The patch you need is UOD385 Connection Drivers SLS. It is available from
SCO <ulink noescape="1" url="ftp://ftp.sco.com/">ftp.sco.com</ulink>, directory SLS,
files uod385a.Z and uod385a.ltr.Z).
</para>

<para>
The information provided here refers to an old version of SCO UNIX. If you require
binaries for more recent SCO UNIX products, please contact SCO to obtain packages that are
ready to install. You should also verify with SCO that your platform is up to date for the
binary packages you will install. This is important if you wish to avoid data corruption
problems with your installation. To build Samba for SCO UNIX products  may
require significant patching of Samba source code. It is much easier to obtain binary
packages directly from SCO.
</para>

</sect1>

<sect1>
<title>DNIX</title>

<para>
DNIX has a problem with seteuid() and setegid(). These routines are
needed for Samba to work correctly, but they were left out of the DNIX
C library for some reason.
</para>

<para>
For this reason Samba by default defines the macro NO_EID in the DNIX
section of includes.h. This works around the problem in a limited way,
but it is far from ideal, and some things still will not work right.
</para>

<para> 
To fix the problem properly, you need to assemble the following two
functions and then either add them to your C library or link them into
Samba. Put the following in the file <filename moreinfo="none">setegid.s</filename>:
</para>

<para><programlisting format="linespecific">
        .globl  _setegid
_setegid:
        moveq   #47,d0
        movl    #100,a0
        moveq   #1,d1
        movl    4(sp),a1
        trap    #9
        bccs    1$
        jmp     cerror
1$:
        clrl    d0
        rts
</programlisting></para>

<para>
Put this in the file <filename moreinfo="none">seteuid.s</filename>:
</para>

<para><programlisting format="linespecific">
        .globl  _seteuid
_seteuid:
        moveq   #47,d0
        movl    #100,a0
        moveq   #0,d1
        movl    4(sp),a1
        trap    #9
        bccs    1$
        jmp     cerror
1$:
        clrl    d0
        rts
</programlisting></para>

<para>
After creating the files, you then assemble them using
</para>

<screen format="linespecific">
<prompt moreinfo="none">$ </prompt><userinput moreinfo="none">as seteuid.s</userinput>
<prompt moreinfo="none">$ </prompt><userinput moreinfo="none">as setegid.s</userinput>
</screen>

<para>
which should produce the files <filename moreinfo="none">seteuid.o</filename> and
<filename moreinfo="none">setegid.o</filename>.
</para>

<para>
Next you need to add these to the LIBSM line in the DNIX section of
the Samba Makefile. Your LIBSM line will look something like this:
</para>

<para><programlisting format="linespecific">
LIBSM = setegid.o seteuid.o -ln
</programlisting></para>

<para>
You should then remove the line:
</para>

<para><programlisting format="linespecific">
#define NO_EID
</programlisting></para>

<para>from the DNIX section of <filename moreinfo="none">includes.h</filename>.</para>

</sect1>

<sect1>
<title>Red Hat Linux</title>

<para>
By default during installation, some versions of Red Hat Linux add an
entry to <filename moreinfo="none">/etc/hosts</filename> as follows:
<programlisting format="linespecific">
127.0.0.1 loopback "hostname"."domainname"
</programlisting>
</para>

<para>
<indexterm significance="normal"><primary>loopback interface</primary></indexterm>
This causes Samba to loop back onto the loopback interface.
The result is that Samba fails to communicate correctly with
the world and therefore may fail to correctly negotiate who
is the master browse list holder and who is the master browser.
</para>

<para>
Corrective action: Delete the entry after the word "loopback"
in the line starting 127.0.0.1.
</para>
</sect1>

<sect1>
<title>AIX: Sequential Read Ahead</title>
<!-- From an email by William Jojo <jojowil@hvcc.edu> -->
<para>
Disabling sequential read ahead can improve Samba performance significantly
when there is a relatively high level of multiprogramming (many smbd processes
or mixed with another workload), not an abundance of physical memory or slower
disk technology. These can cause AIX to have a higher WAIT values. Disabling
sequential read-ahead can also have an adverse affect on other workloads in the
system so you will need to evaluate other applications for impact.
</para>

<para>
It is recommended to use the defaults provided by IBM, but if you experience a
high amount of wait time, try disabling read-ahead with the following commands:
</para>

<para>
For AIX 5.1 and earlier: <userinput moreinfo="none">vmtune -r 0</userinput>
</para>

<para>
For AIX 5.2 and later jfs filesystems: <userinput moreinfo="none">ioo -o minpgahead=0</userinput>
</para>

<para>
For AIX 5.2 and later jfs2 filesystems: <userinput moreinfo="none">ioo -o j2_minPageReadAhead=0</userinput>
</para>

<para>
If you have a mix of jfs and jfs2 filesystems on the same host, simply use both
ioo commands.
</para>
</sect1>

<sect1>
<title>Solaris</title>

<sect2>
<title>Locking Improvements</title>

<para>Some people have been experiencing problems with F_SETLKW64/fcntl 
when running Samba on Solaris. The built-in file-locking mechanism was
not scalable. Performance would degrade to the point where processes would
get into loops of trying to lock a file. It would try a lock, then fail,
then try again. The lock attempt was failing before the grant was
occurring. The visible manifestation of this was a handful of
processes stealing all of the CPU, and when they were trussed, they would
be stuck in F_SETLKW64 loops.
</para>

<para>
Please check with Sun support for current patches needed to fix this bug.
The patch revision for 2.6 is 105181-34, for 8 is 108528-19, and for 9 is 112233-04.
After the installation of these patches, it is recommended to reconfigure
and rebuild Samba.
</para>

<para>Thanks to Joe Meslovich for reporting this.</para>

</sect2>

<sect2 id="winbind-solaris9">
<title>Winbind on Solaris 9</title>
<para>
Nsswitch on Solaris 9 refuses to use the Winbind NSS module. This behavior
is fixed by Sun in patch <ulink url="http://sunsolve.sun.com/search/advsearch.do?collection=PATCH&amp;type=collections&amp;max=50&amp;language=en&amp;queryKey5=112960;rev=14&amp;toDocument=yes">112960-14</ulink>.
</para>
</sect2>
</sect1>

</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="Other-Clients">
<chapterinfo>
	<author>
<firstname>Jelmer</firstname><othername>R.</othername><surname>Vernooij</surname><othername>R.</othername>
<affiliation>
	<orgname>The Samba Team</orgname>
	<address format="linespecific"><email>jelmer@samba.org</email></address>
</affiliation></author>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
	<author>
<firstname>Dan</firstname><surname>Shearer</surname>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>dan@samba.org</email></address>
</affiliation>
</author>
	<author>
	<firstname>Jim</firstname><surname>McDonough</surname>
	<affiliation>
		<orgname>IBM</orgname>
		<address format="linespecific"><email>jmcd@us.ibm.com</email></address>
	</affiliation><contrib>OS/2</contrib></author>
	<pubdate>5 Mar 2001</pubdate>
</chapterinfo>

<title>Samba and Other CIFS Clients</title>

<para>This chapter contains client-specific information.</para>

<sect1>
<title>Macintosh Clients</title>

<para>
<indexterm significance="normal"><primary>DAVE</primary></indexterm>
Yes. <ulink url="http://www.thursby.com/">Thursby</ulink> has a CIFS client/server called <ulink url="http://www.thursby.com/products/dave.html">DAVE</ulink>.  They test it against Windows 95, Windows
NT/200x/XP, and Samba for compatibility issues. At the time of this writing, DAVE was at version 5.1. Please
refer to Thursby's Web site for more information regarding this product.
</para>

<para> 
<indexterm significance="normal"><primary>Netatalk</primary></indexterm>
<indexterm significance="normal"><primary>CAP</primary></indexterm>
Alternatives include two free implementations of AppleTalk for several kinds of UNIX machines and several more
commercial ones.  These products allow you to run file services and print services natively to Macintosh
users, with no additional support required on the Macintosh. The two free implementations are <ulink url="http://www.umich.edu/~rsug/netatalk/">Netatalk</ulink> and <ulink url="http://www.cs.mu.oz.au/appletalk/atalk.html">CAP</ulink>.  What Samba offers MS Windows users, these
packages offer to Macs.  For more info on these packages, Samba, and Linux (and other UNIX-based systems), see
<ulink noescape="1" url="http://www.eats.com/linux_mac_win.html">http://www.eats.com/linux_mac_win.html.</ulink>
</para>

<para>Newer versions of the Macintosh (Mac OS X) include Samba.</para>

</sect1>

<sect1>
<title>OS2 Client</title>

	<sect2>
		<title>Configuring OS/2 Warp Connect or OS/2 Warp 4</title>

		<para>Basically, you need three components:</para>
		
		<itemizedlist>
			<listitem><para>The File and Print Client (IBM peer)</para></listitem>
			<listitem><para>TCP/IP (Internet support) </para></listitem>
			<listitem><para>The <quote>NetBIOS over TCP/IP</quote> driver (TCPBEUI)</para></listitem>
		</itemizedlist>
		
		<para>Installing the first two together with the base operating 
		system on a blank system is explained in the Warp manual. If Warp 
		has already been installed, but you now want to install the 
		networking support, use the <quote>Selective Install for Networking</quote> 
		object in the <quote>System Setup</quote> folder.</para>

		<para>Adding the <quote>NetBIOS over TCP/IP</quote> driver is not described 
		in the manual and just barely in the online documentation. Start 
		<literal>MPTS.EXE</literal>, click on <guiicon moreinfo="none">OK</guiicon>, click on <guimenu moreinfo="none">Configure LAPS</guimenu>, and click 
		on <guimenu moreinfo="none">IBM OS/2 NETBIOS OVER TCP/IP</guimenu> in  <guilabel moreinfo="none">Protocols</guilabel>. This line 
		is then moved to <guilabel moreinfo="none">Current Configuration</guilabel>. Select that line, 
		click on <guimenuitem moreinfo="none">Change number</guimenuitem>, and increase it from 0 to 1. Save this
		configuration.</para>

		<para>If the Samba server is not on your local subnet, you 
		can optionally add IP names and addresses of these servers 
		to the <guimenu moreinfo="none">Names List</guimenu> or specify a  WINS server (NetBIOS 
		Nameserver in IBM and RFC terminology). For Warp Connect, you 
		may need to download an update for <constant>IBM Peer</constant> to bring it on 
		the same level as Warp 4. See the IBM OS/2 Warp Web page</para>
	</sect2>
	
	<sect2>
		<title>Configuring Other Versions of OS/2</title>
	
		<para>This sections deals with configuring OS/2 Warp 3 (not Connect), OS/2 1.2, 1.3 or 2.x.</para>
		
		<para>You can use the free Microsoft LAN Manager 2.2c Client for OS/2 that is
		available from 
		<ulink noescape="1" url="ftp://ftp.microsoft.com/BusSys/Clients/LANMAN.OS2/">
		ftp://ftp.microsoft.com/BusSys/Clients/LANMAN.OS2/</ulink>. In a nutshell, edit
	the file <filename moreinfo="none">\OS2VER</filename> in the root directory of the OS/2 boot partition and add the lines:</para>
		
		<para><programlisting format="linespecific">
		20=setup.exe
		20=netwksta.sys
		20=netvdd.sys
		</programlisting></para>
		
		<para>before you install the client. Also, do not use the included NE2000 driver because it is buggy.
		Try the NE2000 or NS2000 driver from <ulink noescape="1" url="ftp://ftp.cdrom.com/pub/os2/network/ndis/">
 		ftp://ftp.cdrom.com/pub/os2/network/ndis/</ulink> instead.
		</para>
	</sect2>
	
	<sect2>
		<title>Printer Driver Download for OS/2 Clients</title>

		<para>Create a share called <parameter>[PRINTDRV]</parameter> that is 
		world-readable. Copy your OS/2 driver files there. The <filename moreinfo="none">.EA_</filename>
		files must still be separate, so you will need to use the original install files
		and not copy an installed driver from an OS/2 system.</para>
		
		<para>Install the NT driver first for that printer. Then, add to your <filename moreinfo="none">smb.conf</filename> a parameter,
		<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="OS2DRIVERMAP" xlink:href="smb.conf.5.html#OS2DRIVERMAP">os2 driver map</link>. 
		Next, in the file specified by <replaceable>filename</replaceable>, map the 
		name of the NT driver name to the OS/2 driver name as follows:</para>
		
		<para><parameter moreinfo="none"><replaceable>nt driver name</replaceable> = <replaceable>os2 driver name</replaceable>.<replaceable>device name</replaceable></parameter>, e.g.,</para>

		<para><parameter moreinfo="none">
		HP LaserJet 5L = LASERJET.HP LaserJet 5L</parameter></para>

		<para>You can have multiple drivers mapped in this file.</para>
	
		<para>If you only specify the OS/2 driver name, and not the 
		device name, the first attempt to download the driver will 
		actually download the files, but the OS/2 client will tell 
		you the driver is not available. On the second attempt, it 
		will work. This is fixed simply by adding the device name
  		 to the mapping, after which it will work on the first attempt.
		</para>
	</sect2>
</sect1>

<sect1>
<title>Windows for Workgroups</title>

<sect2>
<title>Latest TCP/IP Stack from Microsoft</title>

<para>Use the latest TCP/IP stack from Microsoft if you use Windows
for Workgroups. The early TCP/IP stacks had lots of bugs.</para>

<para> 
Microsoft has released an incremental upgrade to its TCP/IP 32-bit VxD drivers. The latest release can be
found at ftp.microsoft.com, located in <filename moreinfo="none">/Softlib/MSLFILES/TCP32B.EXE</filename>.  There is an
update.txt file there that describes the problems that were fixed. New files include
<filename moreinfo="none">WINSOCK.DLL</filename>, <filename moreinfo="none">TELNET.EXE</filename>, <filename moreinfo="none">WSOCK.386</filename>,
<filename moreinfo="none">VNBT.386</filename>, <filename moreinfo="none">WSTCP.386</filename>, <filename moreinfo="none">TRACERT.EXE</filename>,
<filename moreinfo="none">NETSTAT.EXE</filename>, and <filename moreinfo="none">NBTSTAT.EXE</filename>.
</para>

<para>
More information about this patch is available in <ulink url="http://support.microsoft.com/kb/q99891/">Knowledge Base article 99891</ulink>.
</para>

</sect2>

<sect2>
<title>Delete .pwl Files After Password Change</title>

<para>
Windows for Workgroups does a lousy job with passwords. When you change passwords on either
the UNIX box or the PC, the safest thing to do is delete the .pwl files in the Windows
directory. The PC will complain about not finding the files, but will soon get over it,
allowing you to enter the new password.
</para>

<para> 
If you do not do this, you may find that Windows for Workgroups remembers and uses the old
password, even if you told it a new one.
</para>

<para> 
Often Windows for Workgroups will totally ignore a password you give it in a dialog box.
</para>

</sect2>

<sect2>
<title>Configuring Windows for Workgroups Password Handling</title>

<para>
<indexterm significance="normal"><primary>admincfg.exe</primary></indexterm>
There is a program call <filename moreinfo="none">admincfg.exe</filename> on the last disk (disk 8) of the WFW 3.11 disk set.
To install it, type <userinput moreinfo="none">EXPAND A:\ADMINCFG.EX_ C:\WINDOWS\ADMINCFG.EXE</userinput>.  Then add an icon
for it via the <application moreinfo="none">Program Manager</application> <guimenu moreinfo="none">New</guimenu> menu.  This program allows
you to control how WFW handles passwords, Disable Password Caching and so on, for use with <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SECURITY" xlink:href="smb.conf.5.html#SECURITY">security = user</link>.
</para>

</sect2>

<sect2>
<title>Password Case Sensitivity</title>

<para>Windows for Workgroups uppercases the password before sending it to the server.
UNIX passwords can be case-sensitive though. Check the <filename moreinfo="none">smb.conf</filename> information on
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PASSWORDLEVEL" xlink:href="smb.conf.5.html#PASSWORDLEVEL">password level</link> to specify what characters
Samba should try to uppercase when checking.</para>

</sect2>

<sect2>
<title>Use TCP/IP as Default Protocol</title>

<para>To support print queue reporting, you may find
that you have to use TCP/IP as the default protocol under
Windows for Workgroups. For some reason, if you leave NetBEUI as the default,
it may break the print queue reporting on some systems.
It is presumably a Windows for Workgroups bug.</para>

</sect2>

<sect2 id="speedimpr">
<title>Speed Improvement</title>

<para>
Note that some people have found that setting <parameter moreinfo="none">DefaultRcvWindow</parameter> in
the <parameter>[MSTCP]</parameter> section of the 
<filename moreinfo="none">SYSTEM.INI</filename> file under Windows for Workgroups to 3072 gives a
big improvement.
</para>

<para>
My own experience with DefaultRcvWindow is that I get a much better
performance with a large value (16384 or larger). Other people have
reported that anything over 3072 slows things down enormously. One
person even reported a speed drop of a factor of 30 when he went from
3072 to 8192.
</para>
</sect2>
</sect1>

<sect1>
<title>Windows 95/98</title>

<para>
When using Windows 95 OEM SR2, the following updates are recommended where Samba
is being used. Please note that the changes documented in 
<link linkend="speedimpr">Speed Improvement</link> will affect you once these
updates  have been installed.
</para>

<para> 
There are more updates than the ones mentioned here. Refer to the
Microsoft Web site for all currently available updates to your specific version
of Windows 95.
</para>

<simplelist type="vert">
<member>Kernel Update: KRNLUPD.EXE</member>
<member>Ping Fix: PINGUPD.EXE</member>
<member>RPC Update: RPCRTUPD.EXE</member>
<member>TCP/IP Update: VIPUPD.EXE</member>
<member>Redirector Update: VRDRUPD.EXE</member>
</simplelist>

<para>
Also, if using <application moreinfo="none">MS Outlook,</application> it is desirable to 
install the <literal>OLEUPD.EXE</literal> fix. This
fix may stop your machine from hanging for an extended period when exiting
Outlook, and you may notice a significant speedup when accessing network
neighborhood services.
</para>

<sect2>
<title>Speed Improvement</title>

<para>
Configure the Windows 95 TCP/IP registry settings to give better
performance. I use a program called <literal>MTUSPEED.exe</literal> that I got off the
Internet. There are various other utilities of this type freely available.
</para>

</sect2>

</sect1>

<sect1>
<title>Windows 2000 Service Pack 2</title>

<para> 
There are several annoyances with Windows 2000 SP2, one of which
only appears when using a Samba server to host user profiles
to Windows 2000 SP2 clients in a Windows domain. This assumes
that Samba is a member of the domain, but the problem will
most likely occur if it is not.
</para>

<para> 
In order to serve profiles successfully to Windows 2000 SP2 
clients (when not operating as a PDC), Samba must have 
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="NTACLSUPPORT" xlink:href="smb.conf.5.html#NTACLSUPPORT">nt acl support = no</link>
added to the file share that houses the roaming profiles.
If this is not done, then the Windows 2000 SP2 client will
complain about not being able to access the profile (Access 
Denied) and create multiple copies of it on disk (DOMAIN.user.001,
DOMAIN.user.002, and so on). See the <filename moreinfo="none">smb.conf</filename> man page
for more details on this option. Also note that the 
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="NTACLSUPPORT" xlink:href="smb.conf.5.html#NTACLSUPPORT">nt acl support</link> parameter was formally a global parameter in
releases prior to Samba 2.2.2.
</para>

<para> 
<link linkend="minimalprofile">Following example</link> provides a minimal profile share.
</para>

<example id="minimalprofile">
<title>Minimal Profile Share</title>
<simplelist>
<member> </member><member><parameter>[profile]</parameter></member>
<member><indexterm>path</indexterm><parameter>path = /export/profile</parameter></member>
<member><indexterm>create mask</indexterm><parameter>create mask = 0600</parameter></member>
<member><indexterm>directory mask</indexterm><parameter>directory mask = 0700</parameter></member>
<member><indexterm>nt acl support</indexterm><parameter>nt acl support = no</parameter></member>
<member><indexterm>read only</indexterm><parameter>read only = no</parameter></member>
</simplelist>
</example>

<para>
The reason for this bug is that the Windows 200x SP2 client copies
the security descriptor for the profile that contains
the Samba server's SID, and not the domain SID. The client
compares the SID for SAMBA\user and realizes it is
different from the one assigned to DOMAIN\user; hence,
<errorname>access denied</errorname> message.
</para>

<para>
When the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="NTACLSUPPORT" xlink:href="smb.conf.5.html#NTACLSUPPORT">nt acl support</link> parameter is disabled, Samba will send
the Windows 200x client a response to the QuerySecurityDescriptor trans2 call, which causes the client
to set a default ACL for the profile. This default ACL includes:
</para>

<para><emphasis>DOMAIN\user 	<quote>Full Control</quote></emphasis>&gt;</para>

<note><para>This bug does not occur when using Winbind to
create accounts on the Samba host for Domain users.</para></note>

</sect1>

<sect1>
<title>Windows NT 3.1</title>

<para>If you have problems communicating across routers with Windows 
NT 3.1 workstations, read <ulink url="http://support.microsoft.com/default.aspx?scid=kb;Q103765">this Microsoft Knowledge Base article:</ulink>.

</para>

</sect1>

</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="speed">

<chapterinfo>
	<author>
		<firstname>Paul</firstname><surname>Cochrane</surname>
		<affiliation>
			<orgname>Dundee Limb Fitting Centre</orgname>
			<address format="linespecific"><email>paulc@dth.scot.nhs.uk</email></address>
		</affiliation>
	</author>
	<author>
<firstname>Jelmer</firstname><othername>R.</othername><surname>Vernooij</surname><othername>R.</othername>
<affiliation>
	<orgname>The Samba Team</orgname>
	<address format="linespecific"><email>jelmer@samba.org</email></address>
</affiliation></author>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
</chapterinfo>

<title>Samba Performance Tuning</title>

<sect1>
<title>Comparisons</title>

<para>
The Samba server uses TCP to talk to the client, so if you are
trying to see if it performs well, you should really compare it to
programs that use the same protocol. The most readily available
programs for file transfer that use TCP are ftp or another TCP-based
SMB server.
</para>

<para>
If you want to test against something like an NT or Windows for Workgroups server, then
you will have to disable all but TCP on either the client or
server. Otherwise, you may well be using a totally different protocol
(such as NetBEUI) and comparisons may not be valid.
</para>

<para>
Generally, you should find that Samba performs similarly to ftp at raw
transfer speed. It should perform quite a bit faster than NFS,
although this depends on your system.
</para>

<para>
Several people have done comparisons between Samba and Novell, NFS, or
Windows NT. In some cases Samba performed the best, in others the worst. I
suspect the biggest factor is not Samba versus some other system, but the
hardware and drivers used on the various systems. Given similar
hardware, Samba should certainly be competitive in speed with other
systems.
</para>

</sect1>

<sect1>
<title>Socket Options</title>

<para>
There are a number of socket options that can greatly affect the
performance of a TCP-based server like Samba.
</para>

<para>
The socket options that Samba uses are settable both on the command
line with the <option>-O</option> option and in the <filename moreinfo="none">smb.conf</filename> file.
</para>

<para>
The <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SOCKETOPTIONS" xlink:href="smb.conf.5.html#SOCKETOPTIONS">socket options</link> section of the <filename moreinfo="none">smb.conf</filename> manual page describes how
to set these and gives recommendations.
</para>

<para>
Getting the socket options correct can make a big difference to your
performance, but getting them wrong can degrade it by just as
much. The correct settings are very dependent on your local network.
</para>

<para>
The socket option TCP_NODELAY is the one that seems to make the biggest single difference
for most networks. Many people report that adding
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="SOCKETOPTIONS" xlink:href="smb.conf.5.html#SOCKETOPTIONS">socket options = TCP_NODELAY</link>
doubles the read performance of a Samba drive. The best explanation I have seen for
this is that the Microsoft TCP/IP stack is slow in sending TCP ACKs.
</para>

<para>
There have been reports that setting <parameter moreinfo="none">socket options = SO_RCVBUF=8192</parameter> in smb.conf
can seriously degrade Samba performance on the loopback adaptor (IP Address 127.0.0.1). It is strongly
recommended that before specifying any settings for <parameter moreinfo="none">socket options</parameter>, the effect
first be quantitatively measured on the server being configured.
</para>

</sect1>

<sect1>
<title>Read Size</title>

<para>
The option <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="READSIZE" xlink:href="smb.conf.5.html#READSIZE">read size</link> affects the overlap of disk
reads/writes with network reads/writes. If the amount of data being
transferred in several of the SMB commands (currently SMBwrite, SMBwriteX, and
SMBreadbraw) is larger than this value, then the server begins writing
the data before it has received the whole packet from the network, or
in the case of SMBreadbraw, it begins writing to the network before
all the data has been read from disk.
</para>

<para>
This overlapping works best when the speeds of disk and network access
are similar, having little effect when the speed of one is much
greater than the other.
</para>

<para>
The default value is 16384, but little experimentation has been
done as yet to determine the optimal value, and it is likely that the best
value will vary greatly between systems anyway. A value over 65536 is
pointless and will cause you to allocate memory unnecessarily.
</para>

</sect1>

<sect1>
<title>Max Xmit</title>

<para>
	At startup the client and server negotiate a <parameter moreinfo="none">maximum transmit</parameter> size,
which limits the size of nearly all SMB commands. You can set the
maximum size that Samba will negotiate using the <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="MAXXMIT" xlink:href="smb.conf.5.html#MAXXMIT">max xmit</link> option
in <filename moreinfo="none">smb.conf</filename>. Note that this is the maximum size of SMB requests that 
Samba will accept, but not the maximum size that the client will accept.
The client maximum receive size is sent to Samba by the client, and Samba
honors this limit.
</para>

<para>
It defaults to 65536 bytes (the maximum), but it is possible that some
clients may perform better with a smaller transmit unit. Trying values
of less than 2048 is likely to cause severe problems.
In most cases the default is the best option.
</para>

</sect1>

<sect1>
<title>Log Level</title>

<para>
If you set the log level (also known as <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="DEBUGLEVEL" xlink:href="smb.conf.5.html#DEBUGLEVEL">debug level</link>) higher than 2,
then you may suffer a large drop in performance. This is because the
server flushes the log file after each operation, which can be quite
expensive. 
</para>
</sect1>

<sect1>
<title>Read Raw</title>

<para>
The <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="READRAW" xlink:href="smb.conf.5.html#READRAW">read raw</link> operation is designed to be an optimized, low-latency
file read operation. A server may choose to not support it,
however, and Samba makes support for <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="READRAW" xlink:href="smb.conf.5.html#READRAW">read raw</link> optional, with it
being enabled by default.
</para>

<para>
In some cases clients do not handle <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="READRAW" xlink:href="smb.conf.5.html#READRAW">read raw</link> very well and actually
get lower performance using it than they get using the conventional
read operations, so you might like to try <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="READRAW" xlink:href="smb.conf.5.html#READRAW">read raw = no</link> and see what happens on your
network. It might lower, raise, or not affect your performance. Only
testing can really tell.
</para>

</sect1>

<sect1>
<title>Write Raw</title>

<para>
The <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="WRITERAW" xlink:href="smb.conf.5.html#WRITERAW">write raw</link> operation is designed to be an optimized, low-latency
file write operation. A server may choose to not support it, however, and Samba makes support for
<link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="WRITERAW" xlink:href="smb.conf.5.html#WRITERAW">write raw</link> optional, with it being enabled by default.
</para>

<para>
Some machines may find <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="WRITERAW" xlink:href="smb.conf.5.html#WRITERAW">write raw</link> slower than normal write, in which
case you may wish to change this option.
</para>

</sect1>

<sect1>
<title>Slow Logins</title>

<para>
Slow logins are almost always due to the password checking time. Using
the lowest practical <link xmlns:xlink="http://www.w3.org/1999/xlink" linkend="PASSWORDLEVEL" xlink:href="smb.conf.5.html#PASSWORDLEVEL">password level</link> will improve things. 
</para>

</sect1>

<sect1>
<title>Client Tuning</title>

<para>
Often a speed problem can be traced to the client. The client (for
example Windows for Workgroups) can often be tuned for better TCP
performance. Check the sections on the various clients in 
<link linkend="Other-Clients">Samba and Other CIFS Clients</link>.
</para>

</sect1>

<sect1>
<title>Samba Performance Problem Due to Changing Linux Kernel</title>

<para>
A user wrote the following to the mailing list:
</para>

<blockquote>
<para>
<indexterm significance="normal"><primary>Gentoo</primary></indexterm>
<indexterm significance="normal"><primary>slow network</primary></indexterm>
I am running Gentoo on my server and Samba 2.2.8a. Recently I changed kernel versions from
<filename moreinfo="none">linux-2.4.19-gentoo-r10</filename> to <filename moreinfo="none">linux-2.4.20-wolk4.0s</filename>. Now I have a
performance issue with Samba.  Many of you will probably say,  <quote>Move to vanilla sources!</quote> Well, I
tried that and it didn't work. I have a 100MB LAN and two computers (Linux and Windows 2000). The Linux server
shares directories with DivX files, the client (Windows 2000) plays them via LAN. Before, when I was running
the 2.4.19 kernel, everything was fine, but now movies freeze and stop. I tried moving files between the
server and Windows, and it is terribly slow.
</para>
</blockquote>

<para>
The answer he was given is:
</para>

<blockquote>
<para>
<indexterm significance="normal"><primary>ifconfig</primary></indexterm>
<indexterm significance="normal"><primary>framing error</primary></indexterm>
<indexterm significance="normal"><primary>collisions</primary></indexterm>
Grab the mii-tool and check the duplex settings on the NIC.  My guess is that it is a link layer issue, not an
application layer problem. Also run ifconfig and verify that the framing error, collisions, and so on, look
normal for ethernet.
</para>
</blockquote>

</sect1>

<sect1>
<title>Corrupt tdb Files</title>

<para>
<indexterm significance="normal"><primary>PDC</primary></indexterm>
<indexterm significance="normal"><primary>mbd kept spawning</primary></indexterm>
<indexterm significance="normal"><primary>/var/locks/*.tdb</primary></indexterm>
Our Samba PDC server has been hosting three TB of data to our 500+ users [Windows NT/XP]  for the last three
years using Samba without a problem.  Today all shares went very slow. Also, the main smbd kept spawning new
processes, so we had 1600+ running SMDB's (normally we average 250).  It crashed the SUN E3500 cluster twice.
After a lot of searching, I decided to <literal>rm /var/locks/*.tdb</literal>. Happy again.
</para>

<para>
<emphasis>Question:</emphasis> Is there any method of keeping the *.tdb files in top condition, or
how can I detect early corruption?
</para>

<para>
<indexterm significance="normal"><primary>tdbbackup</primary></indexterm>
<indexterm significance="normal"><primary>nmbd</primary></indexterm>
<emphasis>Answer:</emphasis> Yes, run <literal>tdbbackup</literal> each time after stopping nmbd and before starting nmbd.
</para>

<para>
<emphasis>Question:</emphasis> What I also would like to mention is that the service latency seems
a lot lower than before the locks cleanup. Any ideas on keeping it top notch?
</para>

<para>
<emphasis>Answer:</emphasis> Yes. Same answer as for previous question!
</para>

</sect1>

<sect1>
<title>Samba Performance is Very Slow</title>

<para>
<indexterm significance="normal"><primary>slow performance</primary></indexterm>
A site reported experiencing very baffling symptoms with MYOB Premier opening and
accessing its data files. Some  operations on the file would take between 40 and
45 seconds.
</para>

<para>
<indexterm significance="normal"><primary>printer monitor</primary></indexterm>
<indexterm significance="normal"><primary>pauses</primary></indexterm>
It turned out that the printer monitor program running on the Windows
clients was causing the problems. From the logs, we saw activity coming
through with pauses of about 1 second.
</para>

<para>
<indexterm significance="normal"><primary>networks access</primary></indexterm>
<indexterm significance="normal"><primary>printing now</primary></indexterm>
Stopping the monitor software resulted in the networks access at normal
(quick) speed. Restarting the program caused the speed to slow down
again. The printer was a Canon LBP-810 and the relevant task was
something like CAPON (not sure on spelling). The monitor software
displayed a "printing now" dialog on the client during printing.
</para>

<para>
We discovered this by starting with a clean install of Windows and
trying the application at every step of the installation of other software
process (we had to do this many times).
</para>

<para>
Moral of the story: Check everything (other software included)!
</para>

</sect1>

</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="ch-ldap-tls">
<chapterinfo>
        
<author>
	<firstname>Gavin</firstname><surname>Henry</surname>
	<affiliation>
		<orgname>Suretec Systems Limited, UK</orgname>
		<address format="linespecific"><email>ghenry@suretecsystems.com</email></address>
	</affiliation>
</author>
        <pubdate>July 8, 2005</pubdate>
</chapterinfo>
<title>LDAP and Transport Layer Security</title>

<sect1 id="s1-intro-ldap-tls">
<title>Introduction</title>

	<para>
	<indexterm significance="normal"><primary>Transport Layer Seccurity, TLS</primary><secondary>Introduction</secondary></indexterm>
<indexterm significance="normal"><primary>ACL</primary></indexterm>
	Up until now, we have discussed the straightforward configuration of <trademark class="trade">OpenLDAP</trademark>,
	with some advanced features such as ACLs. This does not however, deal with the fact that the network
	transmissions are still in plain text. This is where <firstterm>Transport Layer Security (TLS)</firstterm>
	comes in.
	</para> 

	<para>
<indexterm significance="normal"><primary>RFC 2830</primary></indexterm>
	<trademark class="trade">OpenLDAP</trademark> clients and servers are capable of using the Transport Layer Security (TLS)
	framework to provide integrity and confidentiality protections in accordance with <ulink url="http://rfc.net/rfc2830.html">RFC 2830</ulink>; <emphasis>Lightweight Directory Access Protocol (v3):
	Extension for Transport Layer Security.</emphasis>
	</para>

	<para>
<indexterm significance="normal"><primary>X.509 certificates</primary></indexterm>
	TLS uses X.509 certificates. All servers are required to have valid certificates, whereas client certificates
	are optional. We will only be discussing server certificates.
	</para>

	<tip><para>
<indexterm significance="normal"><primary>DN</primary></indexterm>
<indexterm significance="normal"><primary>CN</primary></indexterm>
<indexterm significance="normal"><primary>FQDN</primary></indexterm>
	The DN of a server certificate must use the CN attribute to name the server, and the CN must carry the
	server's fully qualified domain name (FQDN). Additional alias names and wildcards may be present in the
	<option>subjectAltName</option> certificate extension. More details on server certificate names are in <ulink url="http://rfc.net/rfc2830.html">RFC2830</ulink>.
	</para></tip>

	<para>
	We will discuss this more in the next sections.
	</para>

	</sect1>

	<sect1 id="s1-config-ldap-tls">
	<title>Configuring</title>

	<para>
	<indexterm significance="normal"><primary>Transport Layer Seccurity, TLS</primary><secondary>Configuring</secondary></indexterm>
	Now on to the good bit.
	</para>

	<sect2 id="s1-config-ldap-tls-certs">
	<title>Generating the Certificate Authority</title>

	<para>
<indexterm significance="normal"><primary>Certificate Authority</primary><see>CA</see></indexterm>
	In order to create the relevant certificates, we need to become our own Certificate Authority (CA).
	<footnote><para>We could however, get our generated server certificate signed by proper CAs, like <ulink url="http://www.thawte.com/">Thawte</ulink> and <ulink url="http://www.verisign.com/">VeriSign</ulink>, which
	you pay for, or the free ones, via <ulink url="http://www.cacert.org/">CAcert</ulink>
	</para></footnote> This is necessary, so we can sign the server certificate.
	</para>

	<para>
<indexterm significance="normal"><primary>OpenSSL</primary></indexterm>
	We will be using the <ulink url="http://www.openssl.org">OpenSSL</ulink> <footnote><para>The downside to
	making our own CA, is that the certificate is not automatically recognized by clients, like the commercial
	ones are.</para></footnote> software for this, which is included with every great <trademark class="registered">Linux</trademark> distribution.
	</para>

	<para>
	TLS is used for many types of servers, but the instructions<footnote><para>For information straight from the
	horse's mouth, please visit <ulink url="http://www.openssl.org/docs/HOWTO/">http://www.openssl.org/docs/HOWTO/</ulink>; the main OpenSSL
	site.</para></footnote> presented here, are tailored for <application moreinfo="none">OpenLDAP</application>.
	</para>

	<note><para>
	The <emphasis>Common Name (CN)</emphasis>, in the following example, <emphasis>MUST</emphasis> be
	the fully qualified domain name (FQDN) of your ldap server.
	</para></note>

	<para>
	First we need to generate the CA:
<screen width="90" format="linespecific">
<computeroutput moreinfo="none">
<prompt moreinfo="none">root# </prompt> mkdir myCA
</computeroutput>
</screen>
	Move into that directory:
<screen width="90" format="linespecific">
<computeroutput moreinfo="none">
<prompt moreinfo="none">root# </prompt> cd myCA
</computeroutput>
</screen>
	Now generate the CA:<footnote><para>Your <filename moreinfo="none">CA.pl</filename> or <filename moreinfo="none">CA.sh</filename> might not be
	in the same location as mine is, you can find it by using the <literal>locate</literal> command, i.e.,
	<literal>locate CA.pl</literal>.  If the command complains about the database being too old, run
	<literal>updatedb</literal> as <emphasis>root</emphasis> to update it.</para></footnote>
<screen width="90" format="linespecific">
<computeroutput moreinfo="none">
<prompt moreinfo="none">root# </prompt> /usr/share/ssl/misc/CA.pl -newca
CA certificate filename (or enter to create)
  
Making CA certificate ...
Generating a 1024 bit RSA private key
.......................++++++
.............................++++++
writing new private key to './demoCA/private/cakey.pem'
Enter PEM pass phrase:
Verifying - Enter PEM pass phrase:
-----
You are about to be asked to enter information that will be incorporated
into your certificate request.
What you are about to enter is what is called a Distinguished Name or a DN.
There are quite a few fields but you can leave some blank
For some fields there will be a default value,
If you enter '.', the field will be left blank.
-----
Country Name (2 letter code) [AU]:AU
State or Province Name (full name) [Some-State]:NSW
Locality Name (eg, city) []:Sydney
Organization Name (eg, company) [Internet Widgits Pty Ltd]:Abmas
Organizational Unit Name (eg, section) []:IT
Common Name (eg, YOUR name) []:ldap.abmas.biz
Email Address []:support@abmas.biz
</computeroutput>
</screen>
	</para>

	<para>
	There are some things to note here. 
	</para>

	<orderedlist inheritnum="ignore" continuation="restarts">
		<listitem>
			<para>
			You <emphasis>MUST</emphasis> remember the password, as we will need
			it to sign the server certificate..
			</para>
		</listitem>

		<listitem>
			<para>
			The <emphasis>Common Name (CN)</emphasis>, <emphasis>MUST</emphasis> be the
			fully qualified domain name (FQDN) of your ldap server.
			</para>
		</listitem>
	</orderedlist>

	</sect2>

	<sect2 id="s1-config-ldap-tls-server">
	<title>Generating the Server Certificate</title>

	<para>
	Now we need to generate the server certificate:
<screen width="90" format="linespecific">
<computeroutput moreinfo="none">
<prompt moreinfo="none">root# </prompt> openssl req -new -nodes -keyout newreq.pem -out newreq.pem
Generating a 1024 bit RSA private key
.............++++++
........................................................++++++
writing new private key to 'newreq.pem'
-----
You are about to be asked to enter information that will be incorporated
into your certificate request.
What you are about to enter is what is called a Distinguished Name or a DN.
There are quite a few fields but you can leave some blank
For some fields there will be a default value,
If you enter '.', the field will be left blank.
-----
Country Name (2 letter code) [AU]:AU
State or Province Name (full name) [Some-State]:NSW
Locality Name (eg, city) []:Sydney
Organization Name (eg, company) [Internet Widgits Pty Ltd]:Abmas
Organizational Unit Name (eg, section) []:IT
Common Name (eg, YOUR name) []:ldap.abmas.biz
Email Address []:support@abmas.biz
  
Please enter the following 'extra' attributes
to be sent with your certificate request
A challenge password []:
An optional company name []:
</computeroutput>
</screen>
	</para>	

	<para>
	Again, there are some things to note here. 
	</para>		

	<orderedlist inheritnum="ignore" continuation="restarts">
		<listitem>
			<para>
			You should <emphasis>NOT</emphasis> enter a password.
			</para>
		</listitem>

		<listitem>
			<para>
			The <emphasis>Common Name (CN)</emphasis>, <emphasis>MUST</emphasis> be
			the fully qualified domain name (FQDN) of your ldap server.
			</para>
		</listitem>
	</orderedlist>

	<para>
	Now we sign the certificate with the new CA:
<screen width="90" format="linespecific">
<computeroutput moreinfo="none">
<prompt moreinfo="none">root# </prompt> /usr/share/ssl/misc/CA.pl -sign
Using configuration from /etc/ssl/openssl.cnf
Enter pass phrase for ./demoCA/private/cakey.pem:
Check that the request matches the signature
Signature ok
Certificate Details:
Serial Number: 1 (0x1)
Validity
	Not Before: Mar  6 18:22:26 2005 EDT
	Not After : Mar  6 18:22:26 2006 EDT
Subject:
	countryName               = AU
	stateOrProvinceName       = NSW
	localityName              = Sydney
	organizationName          = Abmas
	organizationalUnitName    = IT
	commonName                = ldap.abmas.biz
	emailAddress              = support@abmas.biz
X509v3 extensions:
	X509v3 Basic Constraints:
	    CA:FALSE
	Netscape Comment:
	    OpenSSL Generated Certificate
	X509v3 Subject Key Identifier:
	    F7:84:87:25:C4:E8:46:6D:0F:47:27:91:F0:16:E0:86:6A:EE:A3:CE
	X509v3 Authority Key Identifier:
	    keyid:27:44:63:3A:CB:09:DC:B1:FF:32:CC:93:23:A4:F1:B4:D5:F0:7E:CC
	    DirName:/C=AU/ST=NSW/L=Sydney/O=Abmas/OU=IT/
						CN=ldap.abmas.biz/emailAddress=support@abmas.biz
	    serial:00

Certificate is to be certified until Mar  6 18:22:26 2006 EDT (365 days)
Sign the certificate? [y/n]:y


1 out of 1 certificate requests certified, commit? [y/n]y
Write out database with 1 new entries
Data Base Updated
Signed certificate is in newcert.pem
</computeroutput>
</screen>
	</para>

	<para>
	That completes the server certificate generation. 
	</para>

	</sect2>

	<sect2 id="s1-config-ldap-tls-install">
	<title>Installing the Certificates</title>

	<para>
	Now we need to copy the certificates to the right configuration directories,
	rename them at the same time (for convenience), change the ownership and
	finally the permissions:
<screen width="90" format="linespecific">
<computeroutput moreinfo="none">
<prompt moreinfo="none">root# </prompt> cp demoCA/cacert.pem /etc/openldap/
<prompt moreinfo="none">root# </prompt> cp newcert.pem /etc/openldap/servercrt.pem
<prompt moreinfo="none">root# </prompt> cp newreq.pem /etc/openldap/serverkey.pem
<prompt moreinfo="none">root# </prompt> chown ldap.ldap /etc/openldap/*.pem
<prompt moreinfo="none">root# </prompt> chmod 640 /etc/openldap/cacert.pem;
<prompt moreinfo="none">root# </prompt> chmod 600 /etc/openldap/serverkey.pem
</computeroutput>
</screen>
	</para>

	<para>
	Now we just need to add these locations to <filename moreinfo="none">slapd.conf</filename>,
	anywhere before the <option>database</option> declaration as shown here:
<screen width="90" format="linespecific">
<computeroutput moreinfo="none">
TLSCertificateFile /etc/openldap/servercrt.pem
TLSCertificateKeyFile /etc/openldap/serverkey.pem
TLSCACertificateFile /etc/openldap/cacert.pem
</computeroutput>
</screen>
	</para>

	<para>
	Here is the declaration and <filename moreinfo="none">ldap.conf</filename>:
<filename moreinfo="none">ldap.conf</filename>
<screen width="90" format="linespecific">
<computeroutput moreinfo="none">
TLS_CACERT /etc/openldap/cacert.pem
</computeroutput>
</screen>
	</para>

	<para>
	That's all there is to it. Now on to <xref linkend="s1-test-ldap-tls"/>
	</para>

	</sect2>

</sect1>

<sect1 id="s1-test-ldap-tls">
<title>Testing</title>

<para>
<indexterm significance="normal"><primary>Transport Layer Security, TLS</primary><secondary>Testing</secondary></indexterm>
This is the easy part. Restart the server:
<screen width="90" format="linespecific">
<computeroutput moreinfo="none">
<prompt moreinfo="none">root# </prompt> /etc/init.d/ldap restart
Stopping slapd:                                            [  OK  ]
Checking configuration files for slapd: config file testing succeeded
Starting slapd:                                            [  OK  ]
</computeroutput>
</screen>
	Then, using <literal>ldapsearch</literal>, test an anonymous search with the
	<option>-ZZ</option><footnote><para>See <literal>man ldapsearch</literal></para></footnote> option:
<screen width="90" format="linespecific">
<computeroutput moreinfo="none">
<prompt moreinfo="none">root# </prompt> ldapsearch -x -b "dc=ldap,dc=abmas,dc=biz" \
        -H 'ldap://ldap.abmas.biz:389' -ZZ
</computeroutput>
</screen>
	Your results should be the same as before you restarted the server, for example:
<screen width="90" format="linespecific">
<computeroutput moreinfo="none">
<prompt moreinfo="none">root# </prompt> ldapsearch -x -b "dc=ldap,dc=abmas,dc=biz" \
    -H 'ldap://ldap.abmas.biz:389' -ZZ

# extended LDIF
#
# LDAPv3
# base &lt;&gt; with scope sub
# filter: (objectclass=*)
# requesting: ALL
#

# abmas.biz
dn: dc=ldap,dc=abmas,dc=biz
objectClass: dcObject
objectClass: organization
o: Abmas
dc: abmas

# Manager, ldap.abmas.biz
dn: cn=Manager,dc=ldap,dc=abmas,dc=biz
objectClass: organizationalRole
cn: Manager

# ABMAS, abmas.biz
dn: sambaDomainName=ABMAS,dc=ldap,dc=abmas,dc=biz
sambaDomainName: ABMAS
sambaSID: S-1-5-21-238355452-1056757430-1592208922
sambaAlgorithmicRidBase: 1000
objectClass: sambaDomain
sambaNextUserRid: 67109862
sambaNextGroupRid: 67109863
</computeroutput>
</screen>
	If you have any problems, please read <xref linkend="s1-int-ldap-tls"/>
</para>

</sect1>

<sect1 id="s1-int-ldap-tls">
<title>Troubleshooting</title>

<para>
<indexterm significance="normal"><primary>Transport Layer Security, TLS</primary><secondary>Troubleshooting</secondary></indexterm>
The most common error when configuring TLS, as I have already mentioned numerous times, is that the
<emphasis>Common Name (CN)</emphasis> you entered in <xref linkend="s1-config-ldap-tls-server"/> is
<emphasis>NOT</emphasis> the Fully Qualified Domain Name (FQDN) of your ldap server.
</para>

<para>
Other errors could be that you have a typo somewhere in your <literal>ldapsearch</literal> command, or that
your have the wrong permissions on the <filename moreinfo="none">servercrt.pem</filename> and <filename moreinfo="none">cacert.pem</filename>
files. They should be set with <literal>chmod 640</literal>, as per <xref linkend="s1-config-ldap-tls-install"/>.
</para>

<para>
For anything else, it's best to read through your ldap logfile or join the <application moreinfo="none">OpenLDAP</application> mailing list.
</para>

</sect1>

</chapter>
	<?latex \cleardoublepage ?>
	<chapter lang="en-US">
<title>Samba Support</title>

<para>
<indexterm significance="normal"><primary>support</primary></indexterm>
One of the most difficult to answer questions in the information technology industry is, <quote>What is
support?</quote>. That question irritates some folks, as much as common answers may annoy others.
</para>

<para>
<indexterm significance="normal"><primary>customers</primary></indexterm>
The most aggravating situation pertaining to support is typified when, as a Linux user, a call is made to
an Internet service provider who, instead of listening to the problem to find a solution, blandly replies:
<quote>Oh, Linux? We do not support Linux!</quote>. It has happened to me, and similar situations happen
through-out the IT industry. Answers like that are designed to inform us that there are some customers
that a business just does not want to deal with, and well may we feel the anguish of the rejection that 
is dished out.
</para>

<para>
One way to consider support is to view it as consisting of the right answer, in the right place,
at the right time, no matter the situation. Support is all that it takes to take away pain, disruption,
inconvenience, loss of productivity, disorientation, uncertainty, and real or perceived risk.
</para>

<para>
<indexterm significance="normal"><primary>provided services</primary></indexterm>
<indexterm significance="normal"><primary>services provided</primary></indexterm>
<indexterm significance="normal"><primary>customer expected</primary></indexterm>
One of the forces that has become a driving force for the adoption of open source software is the fact that
many IT businesses have provided services that have perhaps failed to deliver what the customer expected, or
that have been found wanting for other reasons.
</para>

<para>
<indexterm significance="normal"><primary>consumer expects</primary></indexterm>
<indexterm significance="normal"><primary>problem resolution</primary></indexterm>
In recognition of the need for needs satisfaction as the primary experience an information technology user or
consumer expects, the information provided in this chapter may help someone to avoid an unpleasant experience
in respect of problem resolution.
</para>

<para>
<indexterm significance="normal"><primary>free support</primary></indexterm>
<indexterm significance="normal"><primary>paid-for support</primary></indexterm>
<indexterm significance="normal"><primary>commercial support</primary></indexterm>
In the open source software arena there are two support options: free support and paid-for (commercial)
support.
</para>

	<sect1>
	<title>Free Support</title>

	<para>
<indexterm significance="normal"><primary>user groups</primary></indexterm>
<indexterm significance="normal"><primary>mailing lists</primary></indexterm>
<indexterm significance="normal"><primary>interactive help</primary></indexterm>
<indexterm significance="normal"><primary>help</primary></indexterm>
<indexterm significance="normal"><primary>mutual assistance</primary></indexterm>
<indexterm significance="normal"><primary>assistance</primary></indexterm>
	Free support may be obtained from friends, colleagues, user groups, mailing lists, and interactive help
	facilities. An example of an interactive dacility is the Internet relay chat (IRC) channels that host user
	supported mutual assistance.
	</para>

	<para>
<indexterm significance="normal"><primary>mailing list</primary></indexterm>
<indexterm significance="normal"><primary>deployment</primary></indexterm>
<indexterm significance="normal"><primary>subscription</primary></indexterm>
<indexterm significance="normal"><primary>IRC</primary></indexterm>
<indexterm significance="normal"><primary>project</primary></indexterm>
	The Samba project maintains a mailing list that is commonly used to discuss solutions to Samba deployments.
	Information regarding subscription to the Samba mailing list can be found on the Samba <ulink url="https://lists.samba.org/mailman/">web</ulink> site. The public mailing list that can be used to obtain
	free, user contributed, support is called the <literal moreinfo="none">samba</literal> list. The email address for this list
	is at <literal moreinfo="none">mail:samba@samba.org</literal>.  Information regarding the Samba IRC channels may be found on
	the Samba <ulink url="http://www.samba.org/samba.irc.html">IRC</ulink> web page.
	</para>

	<para>
<indexterm significance="normal"><primary>free support</primary></indexterm>
<indexterm significance="normal"><primary>qualified problem</primary></indexterm>
<indexterm significance="normal"><primary>requesting payment</primary></indexterm>
<indexterm significance="normal"><primary>professional support</primary></indexterm>
	As a general rule, it is considered poor net behavior to contact a Samba Team member directly
	for free support. Most active members of the Samba Team work exceptionally long hours to assist
	users who have demonstrated a qualified problem. Some team members may respond to direct email
	or telephone contact, with requests for assistance, by requesting payment. A few of the Samba
	Team members actually provide professional paid-for Samba support and it is therefore wise
	to show appropriate discretion and reservation in all direct contact.
	</para>

	<para>
<indexterm significance="normal"><primary>bug report</primary></indexterm>
<indexterm significance="normal"><primary>problem report</primary></indexterm>
<indexterm significance="normal"><primary>code maintainer</primary></indexterm>
	When you stumble across a Samba bug, often the quickest way to get it resolved is by posting
	a bug <ulink url="https://bugzilla.samba.org/">report</ulink>. All such reports are mailed to
	the responsible code maintainer for action. The better the report, and the more serious it is,
	the sooner it will be dealt with. On the other hand, if the responsible person can not duplicate
	the reported bug it is likely to be rejected. It is up to you to provide sufficient information
	that will permit the problem to be reproduced.
	</para>

	<para>
<indexterm significance="normal"><primary>purchase support</primary></indexterm>
	We all recognize that sometimes free support does not provide the answer that is sought within
	the time-frame required. At other times the problem is elusive and you may lack the experience
	necessary to isolate the problem and thus to resolve it. This is a situation where is may be
	prudent to purchase paid-for support.
	</para>

	</sect1>

	<sect1>
	<title>Commercial Support</title>

	<para>
	There are six basic support oriented services that are most commonly sought by Samba sites:
	</para>

	<itemizedlist>
		<listitem><para>Assistance with network design</para></listitem>
		<listitem><para>Staff Training</para></listitem>
		<listitem><para>Assistance with Samba network deployment and installation</para></listitem>
		<listitem><para>Priority telephone or email Samba configuration assistance</para></listitem>
		<listitem><para>Trouble-shooting and diagnostic assistance</para></listitem>
		<listitem><para>Provision of quality assured ready-to-install Samba binary packages</para></listitem>
	</itemizedlist>

	<para>
<indexterm significance="normal"><primary>commercial support</primary></indexterm>
<indexterm significance="normal"><primary>country of origin</primary></indexterm>
	Information regarding companies that provide professional Samba support can be obtained by performing a Google
	search, as well as by reference to the Samba <ulink url="http://www.samba.org/samba/support.html">Support</ulink> web page. Companies who notify the Samba Team
	that they provide commercial support are given a free listing that is sorted by the country of origin.
	Multiple listings are permitted, however no guarantee is offered. It is left to you to qualify a support
	provider and to satisfy yourself that both the company and its staff are able to deliver what is required of
	them.
	</para>

	<para>
<indexterm significance="normal"><primary>commercial support</primary></indexterm>
	The policy within the Samba Team is to treat all commercial support providers equally and to show no
	preference. As a result, Samba Team members who provide commercial support are lumped in with everyone else.
	You are encouraged to obtain the services needed from a company in your local area. The open source movement
	is pro-community; so do what you can to help a local business to prosper.
	</para>

	<para>
<indexterm significance="normal"><primary>unsupported software</primary></indexterm>
	Open source software support can be found in any quality, at any price and in any place you can
	to obtain it. Over 180 companies around the world provide Samba support, there is no excuse for
	suffering in the mistaken belief that Samba is unsupported software <?latex --- ?> it is supported.
	</para>

	</sect1>

</chapter>
	<?latex \cleardoublepage ?>
	<chapter id="DNSDHCP">
<chapterinfo>
	<author>
<firstname>John</firstname><surname>Terpstra</surname><othername>H.</othername>
<affiliation>
	<orgname>Samba Team</orgname>
	<address format="linespecific"><email>jht@samba.org</email></address>
</affiliation></author>
</chapterinfo>

<title>DNS and DHCP Configuration Guide</title>

<sect1>
<title>Features and Benefits</title>

<para>
<indexterm significance="normal"><primary>Dynamic Host Configuration Protocol</primary><see>DHCP</see></indexterm>
<indexterm significance="normal"><primary>Domain Name System</primary><see>DNS</see></indexterm>
There are few subjects in the UNIX world that might raise as much contention as
Domain Name System (DNS) and Dynamic Host Configuration Protocol (DHCP).
Not all opinions held for or against particular implementations of DNS and DHCP
are valid.
</para>

<para>
We live in a modern age where many information technology users demand mobility
and freedom. Microsoft Windows users in particular expect to be able to plug their
notebook computer into a network port and have things <quote>just work.</quote>
</para>

<para>
<indexterm significance="normal"><primary>ADS</primary></indexterm>
UNIX administrators have a point. Many of the normative practices in the Microsoft
Windows world at best border on bad practice from a security perspective.
Microsoft Windows networking protocols allow workstations to arbitrarily register
themselves on a network. Windows 2000 Active Directory registers entries in the DNS namespace
that are equally perplexing to UNIX administrators. Welcome to the new world!
</para>


<para>
<indexterm significance="normal"><primary>ISC</primary><secondary>DNS</secondary></indexterm>
<indexterm significance="normal"><primary>ISC</primary><secondary>DHCP</secondary></indexterm>
<indexterm significance="normal"><primary>Dynamic DNS</primary><see>DDNS</see></indexterm>
The purpose of this chapter is to demonstrate the configuration of the Internet
Software Consortium (ISC) DNS and DHCP servers to provide dynamic services that are
compatible with their equivalents in the Microsoft Windows 2000 Server products.
</para>

<para>
This chapter provides no more than a working example of configuration files for both DNS and DHCP servers. The
examples used match configuration examples used elsewhere in this document.
</para>

<para>
<indexterm significance="normal"><primary>DNS</primary></indexterm>
<indexterm significance="normal"><primary>DHCP</primary></indexterm>
<indexterm significance="normal"><primary>BIND9.NET</primary></indexterm>
This chapter explicitly does not provide a tutorial, nor does it pretend to be a reference guide on DNS and
DHCP, as this is well beyond the scope and intent of this document as a whole. Anyone who wants more detailed
reference materials on DNS or DHCP should visit the ISC Web site at <ulink noescape="1" url="http://www.isc.org"> http://www.isc.org</ulink>.  Those wanting a written text might also be interested
in the O'Reilly publications on DNS, see the <ulink url="http://www.oreilly.com/catalog/dns/index.htm">O'Reilly</ulink> web site, and the <ulink url="http://www.bind9.net/books-dhcp">BIND9.NET</ulink> web site for details.
The books are:
</para>

<orderedlist inheritnum="ignore" continuation="restarts">
	<listitem><para>DNS and BIND, By Cricket Liu, Paul Albitz, ISBN: 1-56592-010-4</para></listitem>
	<listitem><para>DNS &amp; Bind Cookbook, By Cricket Liu, ISBN: 0-596-00410-9</para></listitem>
	<listitem><para>The DHCP Handbook (2nd Edition), By: Ralph Droms, Ted Lemon, ISBN 0-672-32327-3</para></listitem>
</orderedlist>

</sect1>

<sect1>
<title>Example Configuration</title>

<para>
<indexterm significance="normal"><primary>WINS</primary></indexterm>
<indexterm significance="normal"><primary>DNS</primary></indexterm>
The DNS is to the Internet what water is to life. Nearly all information resources (host names) are resolved
to their Internet protocol (IP) addresses through DNS.  Windows networking tried hard to avoid the
complexities of DNS, but alas, DNS won.  <indexterm significance="normal"><primary>WINS</primary></indexterm> The alternative to
DNS, the Windows Internet Name Service (WINS) <?latex --- ?> an artifact of NetBIOS networking over the TCP/IP
protocols <?latex --- ?> has demonstrated scalability problems as well as a flat, nonhierarchical namespace that
became unmanageable as the size and complexity of information technology networks grew.
</para>

<para>
<indexterm significance="normal"><primary>RFC 1001</primary></indexterm>
<indexterm significance="normal"><primary>RFC 1002</primary></indexterm>
WINS is a Microsoft implementation of the RFC1001/1002 NetBIOS Name Service (NBNS).
It allows NetBIOS clients (like Microsoft Windows machines) to register an arbitrary
machine name that the administrator or user has chosen together with the IP
address that the machine has been given. Through the use of WINS, network client machines
could resolve machine names to their IP address.
</para>

<para>
The demand for an alternative to the limitations of NetBIOS networking finally drove
Microsoft to use DNS and Active Directory. Microsoft's new implementation attempts
to use DNS in a manner similar to the way that WINS is used for NetBIOS networking.
Both WINS and Microsoft DNS rely on dynamic name registration.
</para> 

<para>
Microsoft Windows clients can perform dynamic name registration to the DNS server
on startup. Alternatively, where DHCP is used to assign workstation IP addresses,
it is possible to register hostnames and their IP address by the DHCP server as
soon as a client acknowledges an IP address lease. Finally, Microsoft DNS can resolve
hostnames via Microsoft WINS.
</para>

<para>
The following configurations demonstrate a simple, insecure dynamic DNS server and
a simple DHCP server that matches the DNS configuration.
</para>

	<sect2>
	<title>Dynamic DNS</title>

	<para>
	<indexterm significance="normal"><primary>DNS</primary><secondary>Dynamic</secondary></indexterm>
	The example DNS configuration is for a private network in the IP address
	space for network 192.168.1.0/24. The private class network address space
	is set forth in RFC1918.
	</para>


	<para>
	<indexterm significance="normal"><primary>BIND</primary></indexterm>
	It is assumed that this network will be situated behind a secure firewall.
	The files that follow work with ISC BIND version 9. BIND is the Berkeley
	Internet Name Daemon.
	</para>

	<para>
	The master configuration file <filename moreinfo="none">/etc/named.conf</filename>
	determines the location of all further configuration files used.
	The location and name of this file is specified in the startup script
	that is part of the operating system.
<programlisting format="linespecific">
# Quenya.Org configuration file

acl mynet {
	192.168.1.0/24;
	127.0.0.1;
};

options {

	directory "/var/named";
	listen-on-v6 { any; };
	notify no;
	forward first;
	forwarders {
		192.168.1.1;
		};
	auth-nxdomain yes;
	multiple-cnames yes;
	listen-on {
		mynet;
		};
};

# The following three zone definitions do not need any modification.
# The first one defines localhost while the second defines the
# reverse lookup for localhost. The last zone "." is the
# definition of the root name servers.

zone "localhost" in {
	type master;
	file "localhost.zone";
};

zone "0.0.127.in-addr.arpa" in {
	type master;
	file "127.0.0.zone";
};

zone "." in {
	type hint;
	file "root.hint";
};

# You can insert further zone records for your own domains below.

zone "quenya.org" {
	type master;
	file "/var/named/quenya.org.hosts";
	allow-query {
		mynet;
		};
	allow-transfer {
		mynet;
		};
	allow-update {
		mynet;
		};
	};

zone "1.168.192.in-addr.arpa" {
	type master;
	file "/var/named/192.168.1.0.rev";
	allow-query {
		mynet;
	};
	allow-transfer {
		mynet;
	};
	allow-update {
		mynet;
	};
};
</programlisting>
	</para>

	<para>
	The following files are all located in the directory <filename moreinfo="none">/var/named</filename>.
	This is the <filename moreinfo="none">/var/named/localhost.zone</filename> file:
<programlisting format="linespecific">
$TTL 1W
@               IN SOA  @   root (
				42              ; serial (d. adams)
				2D              ; refresh
				4H              ; retry
				6W              ; expiry
				1W )            ; minimum

		IN NS           @
		IN A            127.0.0.1
	</programlisting>
	</para>

	<para>
	The <filename moreinfo="none">/var/named/127.0.0.zone</filename> file:
<programlisting format="linespecific">
$TTL 1W
@               IN SOA          localhost.  root.localhost. (
				42              ; serial (d. adams)
				2D              ; refresh
				4H              ; retry
				6W              ; expiry
				1W )            ; minimum

				IN NS           localhost.
1               IN PTR          localhost.
</programlisting>
	</para>

	<para>
		The <filename moreinfo="none">/var/named/quenya.org.host</filename> file:
<programlisting format="linespecific">
$ORIGIN .
$TTL 38400      ; 10 hours 40 minutes
quenya.org      IN SOA  marvel.quenya.org. root.quenya.org. (
				2003021832 ; serial
				10800      ; refresh (3 hours)
				3600       ; retry (1 hour)
				604800     ; expire (1 week)
				38400      ; minimum (10 hours 40 minutes)
				)
			NS      marvel.quenya.org.
			MX      10 mail.quenya.org.
$ORIGIN quenya.org.
frodo                   A       192.168.1.1
marvel                  A       192.168.1.2
;
mail                    CNAME   marvel
www                     CNAME   marvel
</programlisting>
</para>

<para>
	The <filename moreinfo="none">/var/named/192.168.1.0.rev</filename> file:
<programlisting format="linespecific">
$ORIGIN .
$TTL 38400      ; 10 hours 40 minutes
1.168.192.in-addr.arpa  IN SOA  marvel.quenya.org. root.quenya.org. (
				2003021824 ; serial
				10800      ; refresh (3 hours)
				3600       ; retry (1 hour)
				604800     ; expire (1 week)
				38400      ; minimum (10 hours 40 minutes)
				)
			NS      marvel.quenya.org.
$ORIGIN 1.168.192.in-addr.arpa.
1                       PTR     frodo.quenya.org.
2                       PTR     marvel.quenya.org.
</programlisting>
	</para>

	<para>
<indexterm significance="normal"><primary>BIND</primary></indexterm>
<indexterm significance="normal"><primary>dynamic registration files</primary></indexterm>
	The configuration files shown here were copied from a fully working system. All dynamically registered
	entries have been removed. In addition to these files, BIND version 9 will
	create for each of the dynamic registration files a file that has a 
	<filename moreinfo="none">.jnl</filename> extension. Do not edit or tamper with the configuration
	files or with the <filename moreinfo="none">.jnl</filename> files that are created.
	</para>

	</sect2>

	<sect2 id="DHCP">
	<title>DHCP Server</title>

	<para>
	The following file is used with the ISC DHCP Server version 3.
	The file is located in <filename moreinfo="none">/etc/dhcpd.conf</filename>:
	</para>

	<para>
	<programlisting format="linespecific">
ddns-updates on;
ddns-domainname "quenya.org";
option ntp-servers 192.168.1.2;
ddns-update-style ad-hoc;
allow unknown-clients;
default-lease-time 86400;
max-lease-time 172800;

option domain-name "quenya.org";
option domain-name-servers 192.168.1.2;
option netbios-name-servers 192.168.1.2;
option netbios-dd-server 192.168.1.2;
option netbios-node-type 8;

subnet 192.168.1.0 netmask 255.255.255.0 {
	range dynamic-bootp 192.168.1.60 192.168.1.254;
	option subnet-mask 255.255.255.0;
	option routers 192.168.1.2;
	allow unknown-clients;
}
</programlisting>
	</para>

	<para>
	In this example, IP addresses between 192.168.1.1 and 192.168.1.59 are
	reserved for fixed-address (commonly called <constant>hard-wired</constant>) IP addresses. The
	addresses between 192.168.1.60 and 192.168.1.254 are allocated for dynamic use.
	</para>

	</sect2>

</sect1>
</chapter>
	<?latex \cleardoublepage ?>
</part>

<!-- Comment out the following line to include the manpages.
	 *Please* do not commit with the two lines below enabled! -->
<!--
	<?latex \cleardoublepage ?>
	<xi:include href="manpages.xml"/>
-->
	<?latex \cleardoublepage ?>

<appendix>
  <title>
    <acronym>GNU</acronym> General Public License version 3
  </title>
  <para>
    Version 3, 29 June 2007
  </para>
  <para>
    Copyright © 2007 Free Software Foundation, Inc.
    <ulink url="http://fsf.org/">http://fsf.org/</ulink>
  </para>
  <para>
    Everyone is permitted to copy and distribute verbatim copies of this license
    document, but changing it is not allowed.
  </para>
  <bridgehead renderas="sect1">
    Preamble
  </bridgehead>
  <para>
    The <acronym>GNU</acronym> General Public License is a free, copyleft
    license for software and other kinds of works.
  </para>
  <para>
    The licenses for most software and other practical works are designed to
    take away your freedom to share and change the works.  By contrast, the
    <acronym>GNU</acronym> General Public License is intended to guarantee your
    freedom to share and change all versions of a program—to make sure it
    remains free software for all its users.  We, the Free Software Foundation,
    use the <acronym>GNU</acronym> General Public License for most of our
    software; it applies also to any other work released this way by its
    authors.  You can apply it to your programs, too.
  </para>
  <para>
    When we speak of free software, we are referring to freedom, not price.  Our
    General Public Licenses are designed to make sure that you have the freedom
    to distribute copies of free software (and charge for them if you wish),
    that you receive source code or can get it if you want it, that you can
    change the software or use pieces of it in new free programs, and that you
    know you can do these things.
  </para>
  <para>
    To protect your rights, we need to prevent others from denying you these
    rights or asking you to surrender the rights.  Therefore, you have certain
    responsibilities if you distribute copies of the software, or if you modify
    it: responsibilities to respect the freedom of others.
  </para>
  <para>
    For example, if you distribute copies of such a program, whether gratis or
    for a fee, you must pass on to the recipients the same freedoms that you
    received.  You must make sure that they, too, receive or can get the source
    code.  And you must show them these terms so they know their rights.
  </para>
  <para>
    Developers that use the <acronym>GNU</acronym> <acronym>GPL</acronym>
    protect your rights with two steps: (1) assert copyright on the software,
    and (2) offer you this License giving you legal permission to copy,
    distribute and/or modify it.
  </para>
  <para>
    For the developers’ and authors’ protection, the
    <acronym>GPL</acronym> clearly explains that there is no warranty for this
    free software.  For both users’ and authors’ sake, the
    <acronym>GPL</acronym> requires that modified versions be marked as changed,
    so that their problems will not be attributed erroneously to authors of
    previous versions.
  </para>
  <para>
    Some devices are designed to deny users access to install or run modified
    versions of the software inside them, although the manufacturer can do so.
    This is fundamentally incompatible with the aim of protecting users’
    freedom to change the software.  The systematic pattern of such abuse occurs
    in the area of products for individuals to use, which is precisely where it
    is most unacceptable.  Therefore, we have designed this version of the
    <acronym>GPL</acronym> to prohibit the practice for those products.  If such
    problems arise substantially in other domains, we stand ready to extend this
    provision to those domains in future versions of the <acronym>GPL</acronym>,
    as needed to protect the freedom of users.
  </para>
  <para>
    Finally, every program is threatened constantly by software patents.  States
    should not allow patents to restrict development and use of software on
    general-purpose computers, but in those that do, we wish to avoid the
    special danger that patents applied to a free program could make it
    effectively proprietary.  To prevent this, the <acronym>GPL</acronym>
    assures that patents cannot be used to render the program non-free.
  </para>
  <para>
    The precise terms and conditions for copying, distribution and modification
    follow.
  </para>
  <bridgehead>
    TERMS AND CONDITIONS
  </bridgehead>
  <bridgehead renderas="sect1">
    0. Definitions.
  </bridgehead>
  <para>
    “This License” refers to version 3 of the <acronym>GNU</acronym>
    General Public License.
  </para>
  <para>
    “Copyright” also means copyright-like laws that apply to other
    kinds of works, such as semiconductor masks.
  </para>
  <para>
    “The Program” refers to any copyrightable work licensed under
    this License.  Each licensee is addressed as “you”.
    “Licensees” and “recipients” may be individuals or
    organizations.
  </para>
  <para>
    To “modify” a work means to copy from or adapt all or part of
    the work in a fashion requiring copyright permission, other than the making
    of an exact copy.  The resulting work is called a “modified
    version” of the earlier work or a work “based on” the
    earlier work.
  </para>
  <para>
    A “covered work” means either the unmodified Program or a work
    based on the Program.
  </para>
  <para>
    To “propagate” a work means to do anything with it that, without
    permission, would make you directly or secondarily liable for infringement
    under applicable copyright law, except executing it on a computer or
    modifying a private copy.  Propagation includes copying, distribution (with
    or without modification), making available to the public, and in some
    countries other activities as well.
  </para>
  <para>
    To “convey” a work means any kind of propagation that enables
    other parties to make or receive copies.  Mere interaction with a user
    through a computer network, with no transfer of a copy, is not conveying.
  </para>
  <para>
    An interactive user interface displays “Appropriate Legal
    Notices” to the extent that it includes a convenient and prominently
    visible feature that (1) displays an appropriate copyright notice, and (2)
    tells the user that there is no warranty for the work (except to the extent
    that warranties are provided), that licensees may convey the work under this
    License, and how to view a copy of this License.  If the interface presents
    a list of user commands or options, such as a menu, a prominent item in the
    list meets this criterion.
  </para>
  <bridgehead renderas="sect1">
    1. Source Code.
  </bridgehead>
  <para>
    The “source code” for a work means the preferred form of the
    work for making modifications to it.  “Object code” means any
    non-source form of a work.
  </para>
  <para>
    A “Standard Interface” means an interface that either is an
    official standard defined by a recognized standards body, or, in the case of
    interfaces specified for a particular programming language, one that is
    widely used among developers working in that language.
  </para>
  <para>
    The “System Libraries” of an executable work include anything,
    other than the work as a whole, that (a) is included in the normal form of
    packaging a Major Component, but which is not part of that Major Component,
    and (b) serves only to enable use of the work with that Major Component, or
    to implement a Standard Interface for which an implementation is available
    to the public in source code form.  A “Major Component”, in this
    context, means a major essential component (kernel, window system, and so
    on) of the specific operating system (if any) on which the executable work
    runs, or a compiler used to produce the work, or an object code interpreter
    used to run it.
  </para>
  <para>
    The “Corresponding Source” for a work in object code form means
    all the source code needed to generate, install, and (for an executable
    work) run the object code and to modify the work, including scripts to
    control those activities.  However, it does not include the work’s
    System Libraries, or general-purpose tools or generally available free
    programs which are used unmodified in performing those activities but which
    are not part of the work.  For example, Corresponding Source includes
    interface definition files associated with source files for the work, and
    the source code for shared libraries and dynamically linked subprograms that
    the work is specifically designed to require, such as by intimate data
    communication or control flow between those subprograms and other parts of
    the work.
  </para>
  <para>
    The Corresponding Source need not include anything that users can regenerate
    automatically from other parts of the Corresponding Source.
  </para>
  <para>
    The Corresponding Source for a work in source code form is that same work.
  </para>
  <bridgehead renderas="sect1">
    2. Basic Permissions.
  </bridgehead>
  <para>
    All rights granted under this License are granted for the term of copyright
    on the Program, and are irrevocable provided the stated conditions are met.
    This License explicitly affirms your unlimited permission to run the
    unmodified Program.  The output from running a covered work is covered by
    this License only if the output, given its content, constitutes a covered
    work.  This License acknowledges your rights of fair use or other
    equivalent, as provided by copyright law.
  </para>
  <para>
    You may make, run and propagate covered works that you do not convey,
    without conditions so long as your license otherwise remains in force.  You
    may convey covered works to others for the sole purpose of having them make
    modifications exclusively for you, or provide you with facilities for
    running those works, provided that you comply with the terms of this License
    in conveying all material for which you do not control copyright.  Those
    thus making or running the covered works for you must do so exclusively on
    your behalf, under your direction and control, on terms that prohibit them
    from making any copies of your copyrighted material outside their
    relationship with you.
  </para>
  <para>
    Conveying under any other circumstances is permitted solely under the
    conditions stated below.  Sublicensing is not allowed; section 10 makes it
    unnecessary.
  </para>
  <bridgehead renderas="sect1">
    3. Protecting Users’ Legal Rights From Anti-Circumvention Law.
  </bridgehead>
  <para>
    No covered work shall be deemed part of an effective technological measure
    under any applicable law fulfilling obligations under article 11 of the WIPO
    copyright treaty adopted on 20 December 1996, or similar laws prohibiting or
    restricting circumvention of such measures.
  </para>
  <para>
    When you convey a covered work, you waive any legal power to forbid
    circumvention of technological measures to the extent such circumvention is
    effected by exercising rights under this License with respect to the covered
    work, and you disclaim any intention to limit operation or modification of
    the work as a means of enforcing, against the work’s users, your or
    third parties’ legal rights to forbid circumvention of technological
    measures.
  </para>
  <bridgehead renderas="sect1">
    4. Conveying Verbatim Copies.
  </bridgehead>
  <para>
    You may convey verbatim copies of the Program’s source code as you
    receive it, in any medium, provided that you conspicuously and appropriately
    publish on each copy an appropriate copyright notice; keep intact all
    notices stating that this License and any non-permissive terms added in
    accord with section 7 apply to the code; keep intact all notices of the
    absence of any warranty; and give all recipients a copy of this License
    along with the Program.
  </para>
  <para>
    You may charge any price or no price for each copy that you convey, and you
    may offer support or warranty protection for a fee.
  </para>
  <bridgehead renderas="sect1">
    5. Conveying Modified Source Versions.
  </bridgehead>
  <para>
    You may convey a work based on the Program, or the modifications to produce
    it from the Program, in the form of source code under the terms of section
    4, provided that you also meet all of these conditions:
  </para>
  <orderedlist numeration="loweralpha" inheritnum="ignore" continuation="restarts">
    <listitem>
      <para>
        The work must carry prominent notices stating that you modified it, and
        giving a relevant date.
      </para>
    </listitem>
    <listitem>
      <para>
        The work must carry prominent notices stating that it is released under
        this License and any conditions added under section 7.  This requirement
        modifies the requirement in section 4 to “keep intact all
        notices”.
      </para>
    </listitem>
    <listitem>
      <para>
        You must license the entire work, as a whole, under this License to
        anyone who comes into possession of a copy.  This License will therefore
        apply, along with any applicable section 7 additional terms, to the
        whole of the work, and all its parts, regardless of how they are
        packaged.  This License gives no permission to license the work in any
        other way, but it does not invalidate such permission if you have
        separately received it.
      </para>
    </listitem>
    <listitem>
      <para>
        If the work has interactive user interfaces, each must display
        Appropriate Legal Notices; however, if the Program has interactive
        interfaces that do not display Appropriate Legal Notices, your work need
        not make them do so.
      </para>
    </listitem>
  </orderedlist>
  <para>
    A compilation of a covered work with other separate and independent works,
    which are not by their nature extensions of the covered work, and which are
    not combined with it such as to form a larger program, in or on a volume of
    a storage or distribution medium, is called an “aggregate” if
    the compilation and its resulting copyright are not used to limit the access
    or legal rights of the compilation’s users beyond what the individual works
    permit.  Inclusion of a covered work in an aggregate does not cause
    this License to apply to the other parts of the aggregate.
  </para>
  <bridgehead renderas="sect1">
    6. Conveying Non-Source Forms.
  </bridgehead>
  <para>
    You may convey a covered work in object code form under the terms of
    sections 4 and 5, provided that you also convey the machine-readable
    Corresponding Source under the terms of this License, in one of these ways:
  </para>
  <orderedlist numeration="loweralpha" inheritnum="ignore" continuation="restarts">
    <listitem>
      <para>
        Convey the object code in, or embodied in, a physical product (including
        a physical distribution medium), accompanied by the Corresponding Source
        fixed on a durable physical medium customarily used for software
        interchange.
      </para>
    </listitem>
    <listitem>
      <para>
        Convey the object code in, or embodied in, a physical product (including
        a physical distribution medium), accompanied by a written offer, valid
        for at least three years and valid for as long as you offer spare parts
        or customer support for that product model, to give anyone who possesses
        the object code either (1) a copy of the Corresponding Source for all
        the software in the product that is covered by this License, on a
        durable physical medium customarily used for software interchange, for a
        price no more than your reasonable cost of physically performing this
        conveying of source, or (2) access to copy the Corresponding Source from
        a network server at no charge.
      </para>
    </listitem>
    <listitem>
      <para>
        Convey individual copies of the object code with a copy of the written
        offer to provide the Corresponding Source.  This alternative is allowed
        only occasionally and noncommercially, and only if you received the
        object code with such an offer, in accord with subsection 6b.
      </para>
    </listitem>
    <listitem>
      <para>
        Convey the object code by offering access from a designated place
        (gratis or for a charge), and offer equivalent access to the
        Corresponding Source in the same way through the same place at no
        further charge.  You need not require recipients to copy the
        Corresponding Source along with the object code.  If the place to copy
        the object code is a network server, the Corresponding Source may be on
        a different server (operated by you or a third party) that supports
        equivalent copying facilities, provided you maintain clear directions
        next to the object code saying where to find the Corresponding Source.
        Regardless of what server hosts the Corresponding Source, you remain
        obligated to ensure that it is available for as long as needed to
        satisfy these requirements.
      </para>
    </listitem>
    <listitem>
      <para>
        Convey the object code using peer-to-peer transmission, provided you
        inform other peers where the object code and Corresponding Source of the
        work are being offered to the general public at no charge under
        subsection 6d.
      </para>
    </listitem>
  </orderedlist>
  <para>
    A separable portion of the object code, whose source code is excluded from
    the Corresponding Source as a System Library, need not be included in
    conveying the object code work.
  </para>
  <para>
    A “User Product” is either (1) a “consumer product”,
    which means any tangible personal property which is normally used for
    personal, family, or household purposes, or (2) anything designed or sold
    for incorporation into a dwelling.  In determining whether a product is a
    consumer product, doubtful cases shall be resolved in favor of coverage.
    For a particular product received by a particular user, “normally
    used” refers to a typical or common use of that class of product,
    regardless of the status of the particular user or of the way in which the
    particular user actually uses, or expects or is expected to use, the
    product.  A product is a consumer product regardless of whether the product
    has substantial commercial, industrial or non-consumer uses, unless such
    uses represent the only significant mode of use of the product.
  </para>
  <para>
    “Installation Information” for a User Product means any methods,
    procedures, authorization keys, or other information required to install and
    execute modified versions of a covered work in that User Product from a
    modified version of its Corresponding Source.  The information must suffice
    to ensure that the continued functioning of the modified object code is in
    no case prevented or interfered with solely because modification has been
    made.
  </para>
  <para>
    If you convey an object code work under this section in, or with, or
    specifically for use in, a User Product, and the conveying occurs as part of
    a transaction in which the right of possession and use of the User Product
    is transferred to the recipient in perpetuity or for a fixed term
    (regardless of how the transaction is characterized), the Corresponding
    Source conveyed under this section must be accompanied by the Installation
    Information.  But this requirement does not apply if neither you nor any
    third party retains the ability to install modified object code on the User
    Product (for example, the work has been installed in
    <acronym>ROM</acronym>).
  </para>
  <para>
    The requirement to provide Installation Information does not include a
    requirement to continue to provide support service, warranty, or updates for
    a work that has been modified or installed by the recipient, or for the User
    Product in which it has been modified or installed.  Access to a network may
    be denied when the modification itself materially and adversely affects the
    operation of the network or violates the rules and protocols for
    communication across the network.
  </para>
  <para>
    Corresponding Source conveyed, and Installation Information provided, in
    accord with this section must be in a format that is publicly documented
    (and with an implementation available to the public in source code form),
    and must require no special password or key for unpacking, reading or
    copying.
  </para>
   <bridgehead renderas="sect1">
     7. Additional Terms.
   </bridgehead>
   <para>
     “Additional permissions” are terms that supplement the terms of
     this License by making exceptions from one or more of its conditions.
     Additional permissions that are applicable to the entire Program shall be
     treated as though they were included in this License, to the extent that
     they are valid under applicable law.  If additional permissions apply only
     to part of the Program, that part may be used separately under those
     permissions, but the entire Program remains governed by this License
     without regard to the additional permissions.
   </para>
   <para>
     When you convey a copy of a covered work, you may at your option remove any
     additional permissions from that copy, or from any part of it.  (Additional
     permissions may be written to require their own removal in certain cases
     when you modify the work.)  You may place additional permissions on
     material, added by you to a covered work, for which you have or can give
     appropriate copyright permission.
   </para>
   <para>
     Notwithstanding any other provision of this License, for material you add
     to a covered work, you may (if authorized by the copyright holders of that
     material) supplement the terms of this License with terms:
   </para>
   <orderedlist numeration="loweralpha" inheritnum="ignore" continuation="restarts">
     <listitem>
       <para>
         Disclaiming warranty or limiting liability differently from the terms
         of sections 15 and 16 of this License; or
       </para>
     </listitem>
     <listitem>
       <para>
         Requiring preservation of specified reasonable legal notices or author
         attributions in that material or in the Appropriate Legal Notices
         displayed by works containing it; or
       </para>
     </listitem>
     <listitem>
       <para>
         Prohibiting misrepresentation of the origin of that material, or
         requiring that modified versions of such material be marked in
         reasonable ways as different from the original version; or
       </para>
     </listitem>
     <listitem>
       <para>
         Limiting the use for publicity purposes of names of licensors or
         authors of the material; or
       </para>
     </listitem>
     <listitem>
       <para>
         Declining to grant rights under trademark law for use of some trade
         names, trademarks, or service marks; or
       </para>
     </listitem>
     <listitem>
       <para>
         Requiring indemnification of licensors and authors of that material by
         anyone who conveys the material (or modified versions of it) with
         contractual assumptions of liability to the recipient, for any
         liability that these contractual assumptions directly impose on those
         licensors and authors.
       </para>
     </listitem>
   </orderedlist>
   <para>
     All other non-permissive additional terms are considered “further
     restrictions” within the meaning of section 10.  If the Program as
     you received it, or any part of it, contains a notice stating that it is
     governed by this License along with a term that is a further restriction,
     you may remove that term.  If a license document contains a further
     restriction but permits relicensing or conveying under this License, you
     may add to a covered work material governed by the terms of that license
     document, provided that the further restriction does not survive such
     relicensing or conveying.
   </para>
   <para>
     If you add terms to a covered work in accord with this section, you must
     place, in the relevant source files, a statement of the additional terms
     that apply to those files, or a notice indicating where to find the
     applicable terms.
   </para>
   <para>
     Additional terms, permissive or non-permissive, may be stated in the form
     of a separately written license, or stated as exceptions; the above
     requirements apply either way.
   </para>
   <bridgehead renderas="sect1">
     8. Termination.
   </bridgehead>
   <para>
     You may not propagate or modify a covered work except as expressly provided
     under this License.  Any attempt otherwise to propagate or modify it is
     void, and will automatically terminate your rights under this License
     (including any patent licenses granted under the third paragraph of section
     11).
   </para>
   <para>
     However, if you cease all violation of this License, then your license from
     a particular copyright holder is reinstated (a) provisionally, unless and
     until the copyright holder explicitly and finally terminates your license,
     and (b) permanently, if the copyright holder fails to notify you of the
     violation by some reasonable means prior to 60 days after the cessation.
   </para>
   <para>
     Moreover, your license from a particular copyright holder is reinstated
     permanently if the copyright holder notifies you of the violation by some
     reasonable means, this is the first time you have received notice of
     violation of this License (for any work) from that copyright holder, and
     you cure the violation prior to 30 days after your receipt of the notice.
   </para>
   <para>
     Termination of your rights under this section does not terminate the
     licenses of parties who have received copies or rights from you under this
     License.  If your rights have been terminated and not permanently
     reinstated, you do not qualify to receive new licenses for the same
     material under section 10.
   </para>
   <bridgehead renderas="sect1">
     9. Acceptance Not Required for Having Copies.
   </bridgehead>
   <para>
     You are not required to accept this License in order to receive or run a
     copy of the Program.  Ancillary propagation of a covered work occurring
     solely as a consequence of using peer-to-peer transmission to receive a
     copy likewise does not require acceptance.  However, nothing other than
     this License grants you permission to propagate or modify any covered work.
     These actions infringe copyright if you do not accept this License.
     Therefore, by modifying or propagating a covered work, you indicate your
     acceptance of this License to do so.
   </para>
   <bridgehead renderas="sect1">
     10. Automatic Licensing of Downstream Recipients.
   </bridgehead>
   <para>
     Each time you convey a covered work, the recipient automatically receives a
     license from the original licensors, to run, modify and propagate that
     work, subject to this License.  You are not responsible for enforcing
     compliance by third parties with this License.
   </para>
   <para>
     An “entity transaction” is a transaction transferring control
     of an organization, or substantially all assets of one, or subdividing an
     organization, or merging organizations.  If propagation of a covered work
     results from an entity transaction, each party to that transaction who
     receives a copy of the work also receives whatever licenses to the work the
     party’s predecessor in interest had or could give under the previous
     paragraph, plus a right to possession of the Corresponding Source of the
     work from the predecessor in interest, if the predecessor has it or can get
     it with reasonable efforts.
   </para>
   <para>
     You may not impose any further restrictions on the exercise of the rights
     granted or affirmed under this License.  For example, you may not impose a
     license fee, royalty, or other charge for exercise of rights granted under
     this License, and you may not initiate litigation (including a cross-claim
     or counterclaim in a lawsuit) alleging that any patent claim is infringed
     by making, using, selling, offering for sale, or importing the Program or
     any portion of it.
   </para>
  <bridgehead renderas="sect1">
    11. Patents.
  </bridgehead>
  <para>
    A “contributor” is a copyright holder who authorizes use under
    this License of the Program or a work on which the Program is based.  The
    work thus licensed is called the contributor’s “contributor
    version”.
  </para>
  <para>
    A contributor’s “essential patent claims” are all patent
    claims owned or controlled by the contributor, whether already acquired or
    hereafter acquired, that would be infringed by some manner, permitted by
    this License, of making, using, or selling its contributor version, but do
    not include claims that would be infringed only as a consequence of further
    modification of the contributor version.  For purposes of this definition,
    “control” includes the right to grant patent sublicenses in a
    manner consistent with the requirements of this License.
  </para>
  <para>
    Each contributor grants you a non-exclusive, worldwide, royalty-free patent
    license under the contributor’s essential patent claims, to make, use,
    sell, offer for sale, import and otherwise run, modify and propagate the
    contents of its contributor version.
  </para>
  <para>
    In the following three paragraphs, a “patent license” is any
    express agreement or commitment, however denominated, not to enforce a
    patent (such as an express permission to practice a patent or covenant not
    to sue for patent infringement).  To “grant” such a patent
    license to a party means to make such an agreement or commitment not to
    enforce a patent against the party.
  </para>
  <para>
    If you convey a covered work, knowingly relying on a patent license, and the
    Corresponding Source of the work is not available for anyone to copy, free
    of charge and under the terms of this License, through a publicly available
    network server or other readily accessible means, then you must either (1)
    cause the Corresponding Source to be so available, or (2) arrange to deprive
    yourself of the benefit of the patent license for this particular work, or
    (3) arrange, in a manner consistent with the requirements of this License,
    to extend the patent license to downstream recipients.  “Knowingly
    relying” means you have actual knowledge that, but for the patent
    license, your conveying the covered work in a country, or your
    recipient’s use of the covered work in a country, would infringe one
    or more identifiable patents in that country that you have reason to believe
    are valid.
  </para>
  <para>
    If, pursuant to or in connection with a single transaction or arrangement,
    you convey, or propagate by procuring conveyance of, a covered work, and
    grant a patent license to some of the parties receiving the covered work
    authorizing them to use, propagate, modify or convey a specific copy of the
    covered work, then the patent license you grant is automatically extended to
    all recipients of the covered work and works based on it.
  </para>
  <para>
    A patent license is “discriminatory” if it does not include
    within the scope of its coverage, prohibits the exercise of, or is
    conditioned on the non-exercise of one or more of the rights that are
    specifically granted under this License.  You may not convey a covered work
    if you are a party to an arrangement with a third party that is in the
    business of distributing software, under which you make payment to the third
    party based on the extent of your activity of conveying the work, and under
    which the third party grants, to any of the parties who would receive the
    covered work from you, a discriminatory patent license (a) in connection
    with copies of the covered work conveyed by you (or copies made from those
    copies), or (b) primarily for and in connection with specific products or
    compilations that contain the covered work, unless you entered into that
    arrangement, or that patent license was granted, prior to 28 March 2007.
  </para>
  <para>
    Nothing in this License shall be construed as excluding or limiting any
    implied license or other defenses to infringement that may otherwise be
    available to you under applicable patent law.
  </para>
  <bridgehead renderas="sect1">
    12. No Surrender of Others’ Freedom.
  </bridgehead>
  <para>
    If conditions are imposed on you (whether by court order, agreement or
    otherwise) that contradict the conditions of this License, they do not
    excuse you from the conditions of this License.  If you cannot convey a
    covered work so as to satisfy simultaneously your obligations under this
    License and any other pertinent obligations, then as a consequence you may
    not convey it at all.  For example, if you agree to terms that obligate you
    to collect a royalty for further conveying from those to whom you convey the
    Program, the only way you could satisfy both those terms and this License
    would be to refrain entirely from conveying the Program.
  </para>
  <bridgehead renderas="sect1">
    13. Use with the <acronym>GNU</acronym> Affero General Public License.
  </bridgehead>
  <para>
    Notwithstanding any other provision of this License, you have permission to
    link or combine any covered work with a work licensed under version 3 of the
    <acronym>GNU</acronym> Affero General Public License into a single combined
    work, and to convey the resulting work.  The terms of this License will
    continue to apply to the part which is the covered work, but the special
    requirements of the <acronym>GNU</acronym> Affero General Public License,
    section 13, concerning interaction through a network will apply to the
    combination as such.
  </para>
  <bridgehead renderas="sect1">
    14. Revised Versions of this License.
  </bridgehead>
  <para>
    The Free Software Foundation may publish revised and/or new versions of the
    <acronym>GNU</acronym> General Public License from time to time.  Such new
    versions will be similar in spirit to the present version, but may differ in
    detail to address new problems or concerns.
  </para>
  <para>
    Each version is given a distinguishing version number.  If the Program
    specifies that a certain numbered version of the <acronym>GNU</acronym>
    General Public License “or any later version” applies to it, you
    have the option of following the terms and conditions either of that
    numbered version or of any later version published by the Free Software
    Foundation.  If the Program does not specify a version number of the
    <acronym>GNU</acronym> General Public License, you may choose any version
    ever published by the Free Software Foundation.
  </para>
  <para>
    If the Program specifies that a proxy can decide which future versions of
    the <acronym>GNU</acronym> General Public License can be used, that
    proxy’s public statement of acceptance of a version permanently
    authorizes you to choose that version for the Program.
  </para>
  <para>
    Later license versions may give you additional or different permissions.
    However, no additional obligations are imposed on any author or copyright
    holder as a result of your choosing to follow a later version.
  </para>
  <bridgehead renderas="sect1">
    15. Disclaimer of Warranty.
  </bridgehead>
  <para>
    THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE
    LAW.  EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR
    OTHER PARTIES PROVIDE THE PROGRAM “AS IS” WITHOUT WARRANTY OF
    ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE
    IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.
    THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH
    YOU.  SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL
    NECESSARY SERVICING, REPAIR OR CORRECTION.
  </para>
  <bridgehead renderas="sect1">
    16. Limitation of Liability.
  </bridgehead>
  <para>
    IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING WILL
    ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MODIFIES AND/OR CONVEYS THE
    PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY
    GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE
    OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED TO LOSS OF DATA
    OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD
    PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS),
    EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF
    SUCH DAMAGES.
  </para>
  <bridgehead renderas="sect1">
    17. Interpretation of Sections 15 and 16.
  </bridgehead>
  <para>
    If the disclaimer of warranty and limitation of liability provided above
    cannot be given local legal effect according to their terms, reviewing
    courts shall apply local law that most closely approximates an absolute
    waiver of all civil liability in connection with the Program, unless a
    warranty or assumption of liability accompanies a copy of the Program in
    return for a fee.
  </para>
  <bridgehead>
    END OF TERMS AND CONDITIONS
  </bridgehead>
  <bridgehead renderas="sect1">
    How to Apply These Terms to Your New Programs
  </bridgehead>
  <para>
    If you develop a new program, and you want it to be of the greatest possible
    use to the public, the best way to achieve this is to make it free software
    which everyone can redistribute and change under these terms.
  </para>
  <para>
    To do so, attach the following notices to the program.  It is safest to
    attach them to the start of each source file to most effectively state the
    exclusion of warranty; and each file should have at least the
    “copyright” line and a pointer to where the full notice is
    found.
  </para>
  <screen format="linespecific">
<replaceable>one line to give the program’s name and a brief idea of what it does.</replaceable>
Copyright (C) <replaceable>year</replaceable> <replaceable>name of author</replaceable>

This program is free software: you can redistribute it and/or modify
it under the terms of the <acronym>GNU</acronym> General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.

This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
<acronym>GNU</acronym> General Public License for more details.

You should have received a copy of the <acronym>GNU</acronym> General Public License
along with this program.  If not, see <ulink url="http://www.gnu.org/licenses/">http://www.gnu.org/licenses/</ulink>.
  </screen>
  <para>
    Also add information on how to contact you by electronic and paper mail.
  </para>
  <para>
    If the program does terminal interaction, make it output a short notice like
    this when it starts in an interactive mode:
  </para>
  <screen format="linespecific">
<replaceable>program</replaceable> Copyright (C) <replaceable>year</replaceable> <replaceable>name of author</replaceable>
This program comes with ABSOLUTELY NO WARRANTY; for details type ‘<literal moreinfo="none">show w</literal>’.
This is free software, and you are welcome to redistribute it
under certain conditions; type ‘<literal moreinfo="none">show c</literal>’ for details.
  </screen>
  <para>
    The hypothetical commands ‘<literal moreinfo="none">show w</literal>’ and
    ‘<literal moreinfo="none">show c</literal>’ should show the appropriate parts of
    the General Public License.  Of course, your program’s commands might be
    different; for a GUI interface, you would use an “about box”.
  </para>
  <para>
    You should also get your employer (if you work as a programmer) or school,
    if any, to sign a “copyright disclaimer” for the program, if
    necessary.  For more information on this, and how to apply and follow the
    <acronym>GNU</acronym> <acronym>GPL</acronym>, see <ulink url="http://www.gnu.org/licenses/">http://www.gnu.org/licenses/</ulink>.
  </para>
  <para>
    The <acronym>GNU</acronym> General Public License does not permit
    incorporating your program into proprietary programs.  If your program is a
    subroutine library, you may consider it more useful to permit linking
    proprietary applications with the library.  If this is what you want to do,
    use the <acronym>GNU</acronym> Lesser General Public License instead of this
    License.  But first, please read <ulink url="http://www.gnu.org/philosophy/why-not-lgpl.html">http://www.gnu.org/philosophy/why-not-lgpl.html</ulink>.
  </para>
</appendix>

	<?latex \cleardoublepage ?>
<glossary>
	<title>Glossary</title>

	<glossentry>
		<glossterm>Access Control List</glossterm>
		<acronym>ACL</acronym>
		<glossdef><para>
		A detailed list of permissions granted to users or groups with respect to file and network resource access.
		 See <link linkend="AccessControls"/>, 
		for details.</para></glossdef>
	</glossentry>

	<glossentry>
		<glossterm>Active Directory Service</glossterm>
		<acronym>ADS</acronym>
		<glossdef><para>
		A service unique to Microsoft Windows 200x servers that provides a centrally managed
		directory for management of user identities and computer objects, as well as the permissions
		each user or computer may be granted to access
		distributed network resources. ADS uses Kerberos-based
		authentication and LDAP over Kerberos for directory access.
		</para></glossdef>
        </glossentry>

	<glossentry>
		<glossterm>Common Internet File System</glossterm>
		<acronym>CIFS</acronym>
		<glossdef><para>The new name for SMB. Microsoft renamed the 
		SMB protocol to CIFS during the Internet hype in the nineties.
		At about the time that the SMB protocol was renamed to CIFS, an
		additional dialect of the SMB protocol was in development.
		The need for the deployment of the NetBIOS layer was also
		removed, thus paving the way for use of the SMB protocol natively
		over TCP/IP (known as NetBIOS-less SMB or <quote>naked</quote>  TCP transport).
		</para></glossdef>
	</glossentry>

	<glossentry>
		<glossterm>Common UNIX Printing System</glossterm>
		<acronym>CUPS</acronym>
		<glossdef><para>
		A recent implementation of a high capability printing system for UNIX developed by
		<ulink url="http://www.easysw.com/"/>. The design objective of CUPS was to provide
		a rich print processing system that has built-in intelligence capable of correctly rendering (processing)
		a file that is submitted for printing even if it was formatted for an entirely different printer.
		</para>
		</glossdef>
	</glossentry>

	<glossentry>
		<glossterm>Domain Master Browser</glossterm>
		<acronym>DMB</acronym>
		<glossdef><para>The domain master browser maintains a list of all the servers that 
		have announced their services within a given workgroup or NT domain. See <link linkend="DMB"/> for details.
		</para></glossdef>
	</glossentry>

	<glossentry>
                <glossterm>Domain Name Service</glossterm>
		<acronym>DNS</acronym>
		<glossdef><para>
		A protocol by which computer hostnames may be resolved to the matching IP address/es. DNS is implemented
		by the Berkeley Internet Name Daemon. There exists a recent version of DNS that allows dynamic name registration
		by network clients or by a DHCP server. This recent protocol is known as dynamic DNS (DDNS).
		</para></glossdef>
        </glossentry>

	<glossentry>
                <glossterm>Dynamic Host Configuration Protocol</glossterm>
		<acronym>DHCP</acronym>
		<glossdef><para>
		A protocol that was based on the BOOTP protocol that may be used to dynamically assign an IP address,
		from a reserved pool of addresses, to a network client or device. Additionally, DHCP may assign all
		network configuration settings and may be used to register a computer name and its address with a
		dynamic DNS server.
		</para></glossdef>
        </glossentry>
	<glossentry>
		<glossterm>Extended Meta-file Format</glossterm>
		<acronym>EMF</acronym>
		<glossdef>
			<para>
			An intermediate file format used by Microsoft Windows-based servers and clients. EMF files may be
			rendered into a page description language by a print processor.
		</para>
		</glossdef>
	</glossentry>

	<glossentry>
		<glossterm>Graphical Device Interface</glossterm>
		<acronym>GDI</acronym>
		<glossdef><para>
		Device-independent format for printing used by Microsoft Windows. 
		It is quite similar to what PostScript is for UNIX. Printing jobs are first generated in GDI and
		then converted to a device-specific format. See <link linkend="gdipost"/> for details.
		</para></glossdef>
	</glossentry>

	<glossentry>
		<glossterm>Group IDentifier</glossterm>
		<acronym>GID</acronym>
		<glossdef><para>
		The UNIX system group identifier; on older systems, a 32-bit unsigned integer, and on newer systems
		an unsigned 64-bit integer. The GID is used in UNIX-like operating systems for all group-level access
		control.
		</para></glossdef>
	</glossentry>

	<glossentry>
		<glossterm>Internet Print Protocol</glossterm>
		<acronym>IPP</acronym>
		<glossdef><para>An IETF standard for network printing. CUPS
		implements IPP.</para></glossdef>
	</glossentry>

	<glossentry>
		<glossterm>Key Distribution Center</glossterm>
		<acronym>KDC</acronym>
		<glossdef><para>The Kerberos authentication protocol makes use of security keys (also called a ticket)
		by which access to network resources is controlled. The issuing of Kerberos tickets is effected by
		a KDC.</para></glossdef>
	</glossentry>

	<glossentry>
		<glossterm>NetBIOS Extended User Interface</glossterm>
		<acronym>NetBEUI</acronym>
		<glossdef><para>
		Very simple network protocol invented by IBM and Microsoft. It is used 
		to do NetBIOS over Ethernet with low overhead. NetBEUI is a nonroutable
		protocol.
		</para></glossdef>
	</glossentry>

	<glossentry>
		<glossterm>Network Basic Input/Output System</glossterm>
		<acronym>NetBIOS</acronym>
		<glossdef><para>
		NetBIOS is a simple application programming interface (API) invented in the 1980s 
		that allows programs to send data to certain network names. 
		NetBIOS is always run over another network protocol such 
		as IPX/SPX, TCP/IP, or Logical Link Control (LLC). NetBIOS run over LLC
		is best known as NetBEUI (NetBIOS Extended User Interface <?latex --- ?> a complete misnomer!).
		</para></glossdef>
	</glossentry>


	<glossentry>
		<glossterm>NetBT</glossterm>
		<acronym>NBT</acronym>
		<glossdef><para>Protocol for transporting NetBIOS frames over TCP/IP. Uses ports 137, 138, and 139. 
		NetBT is a fully routable protocol.
		</para></glossdef>
	</glossentry>

	<glossentry>
		<glossterm>Local Master Browser</glossterm>
		<acronym>LMB</acronym>
		<glossdef><para>The local master browser maintains a list 
		of all servers that have announced themselves within a given workgroup or NT domain on a particular
		broadcast-isolated subnet. See <link linkend="DMB"/> for details.
		</para></glossdef>
	</glossentry>

	<glossentry>
		<glossterm>Printer Command Language</glossterm>
		<acronym>PCL</acronym>
		<glossdef><para>
		A printer page description language that was developed by Hewlett-Packard
		and is in common use today.
		</para></glossdef>
	</glossentry>

	<glossentry>
		<glossterm>Portable Document Format</glossterm>
		<acronym>PDF</acronym>
		<glossdef>
		<para>
		A highly compressed document format, based on PostScript, used as a document distribution format
		that is supported by Web browsers as well as many applications. Adobe also distributes an application
		called <quote>Acrobat,</quote> which is a PDF reader.
		</para>
		</glossdef>
	</glossentry>

	<glossentry>
		<glossterm>Page Description Language</glossterm>
		<acronym>PDL</acronym>
		<glossdef><para>A language for describing the layout and contents of a printed page.
		The best-known PDLs are Adobe PostScript and Hewlett-Packard PCL (Printer Control Language),
		both of which are used to control laser printers.</para></glossdef>
	</glossentry>

	<glossentry>
		<glossterm>PostScript Printer Description</glossterm>
		<acronym>PPD</acronym>
		<glossdef><para>
		PPDs specify and control options supported by PostScript printers, such as duplexing, stapling, 
		and DPI. See also <link linkend="post-and-ghost"/>. PPD files can be read by printing applications
		to enable correct PostScript page layout for a particular PostScript printer.
		</para></glossdef>
	</glossentry>

	<glossentry>
        <glossterm>Remote Procedure Call</glossterm>
        <acronym>RPC</acronym>
        <glossdef><para>
		RPCs are a means for executing network operations. The RPC protocol is independent of transport protocols. RPC
		does not try to implement any kind of reliability and the application that uses RPCs must be aware of the type
		of transport protocol underneath RPC. An RPC is like a programmatic jump subroutine over a network. RPCs used
		in the UNIX environment are specified in RFC 1050. RPC is a powerful technique for constructing distributed,
		client-server based applications. It is based on extending the notion of conventional, or local procedure
		calling, so that the called procedure need not exist in the same address space as the calling procedure. The
		two processes may be on the same system, or they may be on different systems with a network connecting them.
		By using RPC, programmers of distributed applications avoid the details of the interface with the network. The
		transport independence of RPC isolates the application from the physical and logical elements of the data
		communications mechanism and allows the application to use a variety of transports.
        </para></glossdef>
    </glossentry>

	<glossentry>
		<glossterm>Server Message Block</glossterm>
		<acronym>SMB</acronym>
		<glossdef><para>
		SMB was the original name of the protocol `spoken' by 
		Samba. It was invented in the 1980s by IBM and adopted 
		and extended further by Microsoft. Microsoft 
		renamed the protocol to CIFS during the Internet hype in the 
		1990s.
		</para></glossdef>
	</glossentry>

	<glossentry>
		<glossterm>User IDentifier</glossterm>
		<acronym>UID</acronym>
		<glossdef><para>
		The UNIX system user identifier; on older systems a 32-bit unsigned integer, and on newer systems,
		an unsigned 64-bit integer. The UID is used in UNIX-like operating systems for all user-level access
		control.
		</para></glossdef>
	</glossentry>

	<glossentry>
		<glossterm>Universal Naming Convention</glossterm>
		<acronym>UNC</acronym>
		<glossdef><para>A syntax for specifying the location of network resources (such as file shares).
		The UNC syntax was developed in the early days of MS DOS 3.x and is used internally by the SMB protocol.
		</para></glossdef>
	</glossentry>

	

</glossary>

	<?latex \cleardoublepage ?>
<?latex \chaptermark{Subject index} ?>

<index/>

</book>