<h1 id="overview">Overview</h1>
<p><code>l2tpns</code> is half of a complete L2TP implementation. It
supports only the LNS side of the connection.</p>
<p>L2TP (Layer 2 Tunneling Protocol) is designed to allow any layer 2
protocol (e.g. Ethernet, PPP) to be tunneled over an IP connection.
<code>l2tpns</code> implements PPP over L2TP only.</p>
<p>There are a couple of other L2TP implementations, of which <a
href="http://sourceforge.net/projects/l2tpd">l2tpd</a> is probably the
most popular. l2tpd also will handle being either end of a tunnel, and
is a lot more configurable than <code>l2tpns</code>. However, due to the
way it works, it is nowhere near as scalable.</p>
<p><code>l2tpns</code> uses the TUN/TAP interface provided by the Linux
kernel to receive and send packets. Using some packet manipulation it
doesn't require a single interface per connection, as l2tpd does.</p>
<p>This allows it to scale extremely well to very high loads and very
high numbers of connections.</p>
<p>It also has a plugin architecture which allows custom code to be run
during processing. An example of this is in the walled garden module
included.</p>
<h1 id="installation">Installation</h1>
<h2 id="installation-requirements">Requirements</h2>
<ul>
<li><p>Linux kernel version 2.4 or above, with the Tun/Tap interface
either compiled in, or as a module.</p></li>
<li><p>libcli 1.8.5 or greater. You can get this from <a
href="http://sourceforge.net/projects/libcli">SourceForge</a></p></li>
</ul>
<h2 id="installation-compile">Compiling</h2>
<p>You can generally get away with just running <code>make</code> from
the source directory. This will compile the daemon, associated tools and
any modules shipped with the distribution.</p>
<h2 id="installation-install">Installing</h2>
<p>After you have successfully compiled everything, run
<code>make install</code> to install it. By default, the binaries are
installed into <code>/usr/sbin</code>, the configuration into
<code>/etc/l2tpns</code>, and the modules into
<code>/usr/lib/l2tpns</code>.</p>
<p>You will definately need to edit the configuration files before you
start. See <a href="#configuration">Configuration</a> for more
information.</p>
<h2 id="installation-run">Running</h2>
<p>You only need to run <code>/usr/sbin/l2tpns</code> as root to start
it. It does not normally detach to a daemon process (see the
<code>-d</code> option), so you should perhaps run it from
<code>init</code>.</p>
<p>By default there is no log destination set, so all log messages will
go to stdout.</p>
<h1 id="configuration">Configuration</h1>
<p>All configuration of the software is done from the files installed
into <code>/etc/l2tpns</code>.</p>
<h2 id="config-startup"><code>startup-config</code></h2>
<p>This is the main configuration file for <code>l2tpns</code>. The
format of the file is a list of commands that can be run through the
command-line interface. This file can also be written directly by the
<code>l2tpns</code> process if a user runs the <code>write memory</code>
command, so any comments will be lost. However if your policy is not to
write the config by the program, then feel free to comment the file with
a <code>#</code> or <code>!</code> at the beginning of the line.</p>
<p>A list of the possible configuration directives follows. Each of
these should be set by a line like: set configstring "value" set
ipaddress 192.168.1.1 set boolean true</p>
<dl>
<dt><code>debug</code> (int)</dt>
<dd>
<p>Sets the level of messages that will be written to the log file. The
value should be between 0 and 5, with 0 being no debugging, and 5 being
the highest. A rough description of the levels is:</p>
<dl>
<dt><code>0</code>: Critical Errors</dt>
<dd>
<p>Things are probably broken</p>
</dd>
<dt><code>1</code>: Errors</dt>
<dd>
<p>Things might have gone wrong, but probably will recover</p>
</dd>
<dt><code>2</code>: Warnings</dt>
<dd>
<p>Just in case you care what is not quite perfect</p>
</dd>
<dt><code>3</code>: Information</dt>
<dd>
<p>Parameters of control packets</p>
</dd>
<dt><code>4</code>: Calls</dt>
<dd>
<p>For tracing the execution of the code</p>
</dd>
<dt><code>5</code>: Packets</dt>
<dd>
<p>Everything, including a hex dump of all packets processed... probably
twice</p>
</dd>
</dl>
<p>Note that the higher you set the debugging level, the slower the
program will run. Also, at level 5 a <em>lot</em> of information will be
logged. This should only ever be used for working out why it doesn't
work at all.</p>
</dd>
<dt><code>log_file</code> (string)</dt>
<dd>
<p>This will be where all logging and debugging information is written
to. This may be either a filename, such as <code>/var/log/l2tpns</code>,
or the special magic string <code>syslog:</code>facility, where facility
is any one of the syslog logging facilities, such as
<code>local5</code>.</p>
</dd>
<dt><code>pid_file</code> (string)</dt>
<dd>
<p>If set, the process id will be written to the specified file. The
value must be an absolute path.</p>
</dd>
<dt><code>random_device</code> (string)</dt>
<dd>
<p>Path to random data source (default <code>/dev/urandom</code>). Use
"" to use the rand() library function.</p>
</dd>
<dt><code>l2tp_secret</code> (string)</dt>
<dd>
<p>The secret used by <code>l2tpns</code> for authenticating tunnel
request. Must be the same as the LAC, or authentication will fail. Only
actually be used if the LAC requests authentication.</p>
</dd>
<dt><code>l2tp_mtu</code> (int)</dt>
<dd>
<p>MTU of interface for L2TP traffic (default: <code>1500</code>). Used
to set link MRU and adjust TCP MSS.</p>
</dd>
<dt><code>mp_mrru</code> (int)</dt>
<dd>
<p>MRRU for MP traffic (default: <code>1614</code>). Can be set to 0 to
disable MP negociation.</p>
</dd>
<dt><code>ppp_restart_time</code> (int); <code>ppp_max_configure</code>
(int); <code>ppp_max_failure</code> (int)</dt>
<dd>
<p>PPP counter and timer values, as described in §4.1 of <a
href="ftp://ftp.rfc-editor.org/in-notes/rfc1661.txt">RFC1661</a>.</p>
</dd>
<dt><code>lcp_renegotiation</code> (string)</dt>
<dd>
<p>By default (<code>always</code>), we renegotiate LCP even if the LAC
already did with the client.</p>
<p>We can avoid the LCP renegotiation (proxy LCP negotiation), unless
what was already negotiated is not fine for us
(<code>on-mismatch</code>). To be noted: we will accept not using our
preferred RADIUS authentication method. This is notably useful when the
LAC cannot pass LCP configuration through, and thus we have to try to
accept what was already negotiated by the LAC with the client.</p>
</dd>
<dt><code>primary_dns</code> (ip address); <code>econdary_dns</code> (ip
address)</dt>
<dd>
<p>Whenever a PPP connection is established, DNS servers will be sent to
the user, both a primary and a secondary. If either is set to 0.0.0.0,
then that one will not be sent.</p>
</dd>
<dt><code>primary_radius</code> (ip address);
<code>secondary_radius</code> (ip address)</dt>
<dd>
<p>Sets the RADIUS servers used for both authentication and accounting.
If the primary server does not respond, then the secondary RADIUS server
will be tried.</p>
<div class="note">
<p>In addition to the source IP address and identifier, the RADIUS
server <em>must</em> include the source port when detecting duplicates
to suppress (in order to cope with a large number of sessions coming
on-line simultaneously <code>l2tpns</code> uses a set of udp sockets,
each with a separate identifier).</p>
</div>
</dd>
<dt><code>primary_radius_port</code> (short);
<code>secondary_radius_port</code> (short)</dt>
<dd>
<p>Sets the authentication ports for the primary and secondary RADIUS
servers. The accounting port is one more than the authentication port.
If no RADIUS ports are given, the authentication port defaults to 1812,
and the accounting port to 1813.</p>
</dd>
<dt><code>radius_accounting</code> (boolean)</dt>
<dd>
<p>If set to true, then RADIUS accounting packets will be sent. This
means that a Start record will be sent when the session is successfully
authenticated, and a Stop record will be sent when the session is
closed.</p>
</dd>
<dt><code>radius_interim</code> (int)</dt>
<dd>
<p>If <code>radius_accounting</code> is on, defines the interval between
sending of RADIUS interim accounting records (in seconds).</p>
</dd>
<dt><code>radius_secret</code> (string)</dt>
<dd>
<p>This secret will be used in all RADIUS queries. If this is not set
then RADIUS queries will fail.</p>
</dd>
<dt><code>radius_require_message_authenticator</code> (string)</dt>
<dd>
<p>If set to true, RADIUS answers to AccessRequests will have to contain
a valid MessageAuthenticator. If set to auto (default), if the first
RADIUS answer to AccessRequests contains a valid MessageAuthenticator,
subsequent answers will have to contain one. If set to no (not
recommended), RADIUS answers to AccessRequests do not have to contain a
valid MessageAuthenticator. It is advised to set this to true after
checking that your RADIUS server does send MessageAuthenticator.</p>
</dd>
<dt><code>radius_authtypes</code> (string)</dt>
<dd>
<p>A comma separated list of supported RADIUS authentication methods
(<code>pap</code> or <code>chap</code>), in order of preference (default
<code>pap</code>).</p>
</dd>
<dt><code>radius_bind_min</code> (short); <code>radius_bind_max</code>
(short)</dt>
<dd>
<p>Define a port range in which to bind sockets used to send and receive
RADIUS packets. Must be at least RADIUS_FDS (64) wide. Simplifies
firewalling of RADIUS ports (default: dynamically assigned).</p>
</dd>
<dt><code>radius_dae_port</code> (short)</dt>
<dd>
<p>Port for DAE RADIUS (Packet of Death/Disconnect, Change of
Authorization) requests (default: <code>3799</code>).</p>
</dd>
<dt><code>allow_duplicate_users</code> (boolean)</dt>
<dd>
<p>Allow multiple logins with the same username. If false (the default),
any prior session with the same username will be dropped when a new
session is established.</p>
</dd>
<dt><code>guest_account</code> (string)</dt>
<dd>
<p>Allow multiple logins matching this specific username.</p>
</dd>
<dt><code>bind_address</code> (ip address)</dt>
<dd>
<p>When the tun interface is created, it is assigned the address
specified here. If no address is given, 1.1.1.1 is used. Packets
containing user traffic should be routed via this address if given,
otherwise the primary address of the machine.</p>
</dd>
<dt><code>peer_address</code> (ip address)</dt>
<dd>
<p>Address to send to clients as the default gateway.</p>
</dd>
<dt><code>route_protocol</code> (short)</dt>
<dd>
<p>Protocol number to record when adding a route (see
<code>/usr/share/iproute2/rt_protos</code>). This allows to separate
routes added by several l2tpns instances. The default is 42.</p>
</dd>
<dt><code>route_metric</code> (int)</dt>
<dd>
<p>Metric of the route. Routes with lower values are preferred. This
should not be changed during execution, since l2tpns would then not
recognize the routes it added itself. The default is 1.</p>
</dd>
<dt><code>ipv6_prefix</code> (ipv6 address)</dt>
<dd>
<p>Enable negotiation of IPv6. This forms the the first 64 bits of the
client allocated address. The remaining 64 come from the allocated IPv4
address and 4 bytes of 0s.</p>
</dd>
<dt><code>send_garp</code> (boolean)</dt>
<dd>
<p>Determines whether or not to send a gratuitous ARP for the
bind_address when the server is ready to handle traffic (default:
<code>true</code>). This value is ignored if BGP is configured.</p>
</dd>
<dt><code>throttle_speed</code> (int)</dt>
<dd>
<p>Sets the default speed (in kbits/s) which sessions will be limited
to. If this is set to 0, then throttling will not be used at all. Note:
You can set this by the CLI, but changes will not affect currently
connected users.</p>
</dd>
<dt><code>throttle_buckets</code> (int)</dt>
<dd>
<p>Number of token buckets to allocate for throttling. Each throttled
session requires two buckets (in and out).</p>
</dd>
<dt><code>accounting_dir</code> (string)</dt>
<dd>
<p>If set to a directory, then every 5 minutes the current usage for
every connected use will be dumped to a file in this directory. Each
file dumped begins with a header, where each line is prefixed by
<code>#</code>. Following the header is a single line for every
connected user, fields separated by a space.</p>
<p>The fields are username, ip, qos, uptxoctets, downrxoctets. The qos
field is 1 if a standard user, and 2 if the user is throttled.</p>
</dd>
<dt><code>dump_speed</code> (boolean)</dt>
<dd>
<p>If set to true, then the current bandwidth utilization will be logged
every second. Even if this is disabled, you can see this information by
running the <code>uptime</code> command on the CLI.</p>
</dd>
<dt><code>multi_read_count</code> (int)</dt>
<dd>
<p>Number of packets to read off each of the UDP and TUN fds when
returned as readable by select (default: 10). Avoids incurring the
unnecessary system call overhead of select on busy servers.</p>
</dd>
<dt><code>scheduler_fifo</code> (boolean)</dt>
<dd>
<p>Sets the scheduling policy for the <code>l2tpns</code> process to
<code>SCHED_FIFO</code>. This causes the kernel to immediately preempt
any currently running <code>SCHED_OTHER</code> (normal) process in
favour of <code>l2tpns</code> when it becomes runnable. Ignored on
uniprocessor systems.</p>
</dd>
<dt><code>lock_pages</code> (boolean)</dt>
<dd>
<p>Keep all pages mapped by the <code>l2tpns</code> process in
memory.</p>
</dd>
<dt><code>icmp_rate</code> (int)</dt>
<dd>
<p>Maximum number of host unreachable ICMP packets to send per
second.</p>
</dd>
<dt><code>packet_limit</code> (int)</dt>
<dd>
<p>Maximum number of packets of downstream traffic to be handled each
tenth of a second per session. If zero, no limit is applied (default:
0). Intended as a DoS prevention mechanism and not a general throttling
control (packets are dropped, not queued).</p>
</dd>
<dt><code>cluster_address</code> (ip address)</dt>
<dd>
<p>Multicast cluster address (default: 239.192.13.13). See <a
href="#clustering">Clustering</a> for more information.</p>
</dd>
<dt><code>cluster_port</code> (udp port)</dt>
<dd>
<p>UDP cluster port (default: 32792). See <a
href="#clustering">Clustering</a> for more information.</p>
</dd>
<dt><code>cluster_interface</code> (string)</dt>
<dd>
<p>Interface for cluster packets (default: none). An IPv4 address can
also be specified, e.g. to run several l2tpns on the same host (but
beware of enabling kernel acceleration only on one instance, to set a
different <code>bind_portremotelns</code>, and to set a different
<code>route_metric</code>. It is also useful to set a different
<code>route_protocol</code> number)</p>
</dd>
<dt><code>cluster_mcast_ttl</code> (int)</dt>
<dd>
<p>TTL for multicast packets (default: 1).</p>
</dd>
<dt><code>cluster_hb_interval</code> (int)</dt>
<dd>
<p>Interval in tenths of a second between cluster heartbeat/pings.</p>
</dd>
<dt><code>cluster_hb_timeout</code> (int)</dt>
<dd>
<p>Cluster heartbeat timeout in tenths of a second. A new master will be
elected when this interval has been passed without seeing a heartbeat
from the master.</p>
</dd>
<dt><code>cluster_master_min_adv</code> (int)</dt>
<dd>
<p>Determines the minimum number of up to date slaves required before
the master will drop routes (default: 1).</p>
</dd>
</dl>
<h3 id="config-startup-bgp">BGP</h3>
<p>BGP routing configuration is entered by the command: router bgp as
where as specifies the local AS number.</p>
<p>Subsequent lines prefixed with neighbour peer define the attributes
of BGP neighhbours. Valid commands are: neighbour peer remote-as as
neighbour peer timers keepalive hold</p>
<p>Where peer specifies the BGP neighbour as either a hostname or IP
address, as is the remote AS number and keepalive, hold are the timer
values in seconds.</p>
<h3 id="config-startup-acl">Access Lists</h3>
<p>Named access-lists are configured using one of the commands: ip
access-list standard name ip access-list extended name</p>
<p>Subsequent lines prefixed with <code>permit</code> or
<code>deny</code> define the body of the access-list. Standard
access-list syntax:</p>
<p>{<code>permit</code>|<code>deny</code>} {host|source
source-wildcard|<code>any</code>} [{host|destination
destination-wildcard|<code>any</code>}]</p>
<p>Extended access-lists:</p>
<p>{<code>permit</code>|<code>deny</code>} <code>ip</code> {host|source
source-wildcard|<code>any</code>} {host|destination
destination-wildcard|<code>any</code>} [<code>fragments</code>]</p>
<p>{<code>permit</code>|<code>deny</code>} <code>udp</code> {host|source
source-wildcard|<code>any</code>}
[{<code>eq</code>|<code>neq</code>|<code>gt</code>|<code>lt</code>}
port|<code>range</code> from to] {host|destination
destination-wildcard|<code>any</code>}
[{<code>eq</code>|<code>neq</code>|<code>gt</code>|<code>lt</code>}
port|<code>range</code> from to] [<code>fragments</code>]</p>
<p>{<code>permit</code>|<code>deny</code>} <code>tcp</code> {host|source
source-wildcard|<code>any</code>}
[{<code>eq</code>|<code>neq</code>|<code>gt</code>|<code>lt</code>}
port|<code>range</code> from to] {host|destination
destination-wildcard|<code>any</code>}
[{<code>eq</code>|<code>neq</code>|<code>gt</code>|<code>lt</code>}
port|<code>range</code> from to]
[{<code>established</code>|{<code>match-any</code>|<code>match-all</code>}
{<code>+</code>|<code>-</code>}{<code>fin</code>|<code>syn</code>|<code>rst</code>|<code>psh</code>|<code>ack</code>|<code>urg</code>}
...|<code>fragments</code>]</p>
<h2 id="config-users"><code>users</code></h2>
<p>Usernames and passwords for the command-line interface are stored in
this file. The format is username:password where password may either by
plain text, an MD5 digest (prefixed by
<code>$1</code>salt<code>$</code>) or a DES password, distinguished from
plain text by the prefix <code>{crypt}</code>.</p>
<p>The username <code>enable</code> has a special meaning and is used to
set the enable password.</p>
<div class="important">
<p>If this file doesn't exist, then anyone who can get to port 23 will
be allowed access without a username or password.</p>
</div>
<h2 id="config-ip-pool"><code>ip_pool</code></h2>
<p>This file is used to configure the IP address pool which user
addresses are assigned from. This file should contain either an IP
address or a CIDR network per line. e.g.:</p>
<pre><code>192.168.1.1
192.168.1.2
192.168.1.3
192.168.4.0/24
172.16.0.0/16
10.0.0.0/8</code></pre>
<p>Keep in mind that <code>l2tpns</code> can only handle 65535
connections per process, so don't put more than 65535 IP addresses in
the configuration file. They will be wasted.</p>
<h2 id="config-build-garden"><code>build-garden</code></h2>
<p>The garden plugin on startup creates a NAT table called "garden" then
sources the <code>build-garden</code> script to populate that table. All
packets from gardened users will be sent through this table.
Example:</p>
<pre><code>iptables -t nat -A garden -p tcp -m tcp --dport 25 -j DNAT --to 192.168.1.1
iptables -t nat -A garden -p udp -m udp --dport 53 -j DNAT --to 192.168.1.1
iptables -t nat -A garden -p tcp -m tcp --dport 53 -j DNAT --to 192.168.1.1
iptables -t nat -A garden -p tcp -m tcp --dport 80 -j DNAT --to 192.168.1.1
iptables -t nat -A garden -p tcp -m tcp --dport 110 -j DNAT --to 192.168.1.1
iptables -t nat -A garden -p tcp -m tcp --dport 443 -j DNAT --to 192.168.1.1
iptables -t nat -A garden -p icmp -m icmp --icmp-type echo-request -j DNAT --to 192.168.1.1
iptables -t nat -A garden -p icmp -j ACCEPT
iptables -t nat -A garden -j DROP</code></pre>
<h1 id="operation">Operation</h1>
<p>A running l2tpns process can be controlled in a number of ways. The
primary method of control is by the Command-Line Interface (CLI).</p>
<p>You can also remotely send commands to modules via the
<code>nsctl</code> client provided.</p>
<p>There are also a number of signals that l2tpns understands and takes
action when it receives them.</p>
<h2 id="operation-cli">Command-Line Interface</h2>
<p>You can access the command line interface by telneting to port 23.
There is no IP address restriction, so it's a good idea to firewall this
port off from anyone who doesn't need access to it. See <a
href="#config-users"></a> for information on restricting access based on
a username and password.</p>
<p>The CLI gives you real-time control over almost everything in the
process. The interface is designed to look like a Cisco device, and
supports things like command history, line editing and context sensitive
help. This is provided by linking with the <a
href="http://sourceforge.net/projects/libcli">libcli</a> library. Some
general documentation of the interface is <a
href="http://sourceforge.net/docman/display_doc.php?docid=20501&amp;group_id=79019">here</a>.</p>
<p>After you have connected to the telnet port (and perhaps logged in),
you will be presented with a <code>hostname&gt;</code> prompt.</p>
<p>Enter <code>help</code> to get a list of possible commands, or press
<code>?</code> for context-specific help.</p>
<p>A brief overview of the more important commands follows:</p>
<p><code>show session [ID]</code></p>
<p>: Detailed information for a specific session is presented if you
specify a session ID argument.</p>
<pre><code>If no ID is given, a summary of all connected sessions is produced.
Note that this summary list can be around 185 columns wide, so you
should probably use a wide terminal.

The columns listed in the summary are:

  -------------- -------------------------------------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------------------------------------
  `SID`          Session ID                                                                                                     
  `TID`          Tunnel ID                                                                                                      See also the [show tunnel](#operation-cli-show-tunnel) CLI command.
  `Username`     The username given in the PPP authentication.                                                                  If this is \*, then LCP authentication has not completed.
  `IP`           The IP address given to the session.                                                                           If this is 0.0.0.0, IPCP negotiation has not completed
  `I`            Intercept                                                                                                      Y or N: indicates whether the session is being snooped. See also the [snoop](#operation-cli-snoop) CLI command.
  `T`            Throttled                                                                                                      Y or N: indicates whether the session is currently throttled. See also the [throttle](#operation-cli-throttle) CLI command.
  `G`            Walled Garden                                                                                                  Y or N: indicates whether the user is trapped in the walled garden. This field is present even if the garden module is not loaded.
  `6`            IPv6                                                                                                           Y or N: indicates whether the session has IPv6 active (IPV6CP open)
  `opened`       The number of seconds since the session started                                                                
  `downloaded`   Number of bytes downloaded by the user                                                                         
  `uploaded`     Number of bytes uploaded by the user                                                                           
  `idle`         The number of seconds since traffic was detected on the session                                                
  `LAC`          The IP address of the LAC the session is connected to.                                                         
  `CLI`          The Calling-Line-Identification field provided during the session setup. This field is generated by the LAC.   
  -------------- -------------------------------------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------------------------------------</code></pre>
<p><code>show users</code>; <code>show user username</code></p>
<p>: With no arguments, display a list of currently connected users. If
an argument is given, the session details for the given username are
displayed.</p>
<dl>
<dt><code>show tunnel [ID]</code></dt>
<dd>
<p>Produce a summary list of all open tunnels, or detail on a specific
tunnel ID.</p>
<p>The columns listed in the summary are:</p>
<table>
<tbody>
<tr class="odd">
<td style="text-align: left;">TID</td>
<td style="text-align: left;">Tunnel ID</td>
</tr>
<tr class="even">
<td style="text-align: left;">Hostname</td>
<td style="text-align: left;">The hostname for the tunnel as provided by
the LAC. This has no relation to DNS, it is just a text field.</td>
</tr>
<tr class="odd">
<td style="text-align: left;">IP</td>
<td style="text-align: left;">The IP address of the LAC</td>
</tr>
<tr class="even">
<td style="text-align: left;">State</td>
<td style="text-align: left;">Tunnel state: Free, Open, Dieing,
Opening</td>
</tr>
<tr class="odd">
<td style="text-align: left;">Sessions</td>
<td style="text-align: left;">The number of open sessions on the
tunnel</td>
</tr>
</tbody>
</table>
</dd>
<dt><code>show pool</code></dt>
<dd>
<p>Displays the current IP address pool allocation. This will only
display addresses that are in use, or are reserved for re-allocation to
a disconnected user.</p>
<p>If an address is not currently in use, but has been used, then in the
User column the username will be shown in square brackets, followed by
the time since the address was used:</p>
<pre><code>IP Address      Used  Session User
192.168.100.6     N           [joe.user] 1548s</code></pre>
</dd>
<dt><code>show radius</code></dt>
<dd>
<p>Show a summary of the in-use RADIUS sessions. This list should not be
very long, as RADIUS sessions should be cleaned up as soon as they are
used. The columns listed are:</p>
<table>
<tbody>
<tr class="odd">
<td style="text-align: left;">Radius</td>
<td style="text-align: left;">The ID of the RADIUS request. This is sent
in the packet to the RADIUS server for identification</td>
</tr>
<tr class="even">
<td style="text-align: left;">State</td>
<td style="text-align: left;">The state of the request: WAIT, CHAP,
AUTH, IPCP, START, STOP or NULL</td>
</tr>
<tr class="odd">
<td style="text-align: left;">Session</td>
<td style="text-align: left;">The session ID that this RADIUS request is
associated with</td>
</tr>
<tr class="even">
<td style="text-align: left;">Retry</td>
<td style="text-align: left;">If a response does not appear to the
request, it will retry at this time. This is a Unix timestamp</td>
</tr>
<tr class="odd">
<td style="text-align: left;">Try</td>
<td style="text-align: left;">Retry count. The RADIUS request is
discarded after 3 retries</td>
</tr>
</tbody>
</table>
</dd>
<dt><code>show running-config</code></dt>
<dd>
<p>This will list the current running configuration. This is in a format
that can either be pasted into the configuration file, or run directly
at the command line.</p>
</dd>
<dt><code>show counters</code></dt>
<dd>
<p>Internally, counters are kept of key values, such as bytes and
packets transferred, as well as function call counters. This function
displays all these counters, and is probably only useful for
debugging.</p>
<p>You can reset these counters by running
<code>clear counters</code>.</p>
</dd>
<dt><code>show cluster</code></dt>
<dd>
<p>Show cluster status. Shows the cluster state for this server
(Master/Slave), information about known peers and (for slaves) the
master IP address, last packet seen and up-to-date status. See <a
href="#clustering">Clustering</a> for more information.</p>
</dd>
<dt><code>write memory</code></dt>
<dd>
<p>This will write the current running configuration to the config file
<code>startup-config</code>, which will be run on a restart.</p>
</dd>
</dl>
<p><code>snoop user         IP         port</code></p>
<p>: You must specify a username, IP address and port. All packets for
the current session for that username will be forwarded to the given
host/port. Specify <code>no snoop             username</code> to disable
interception for the session.</p>
<pre><code>If you want interception to be permanent, you will have to modify
the RADIUS response for the user. See [Interception](#interception).</code></pre>
<p><code>throttle user         [in|out] rate</code></p>
<p>: You must specify a username, which will be throttled for the
current session to rate Kbps. Prefix rate with <code>in</code> or
<code>out</code> to set different upstream and downstream rates.</p>
<pre><code>Specify `no throttle
        username` to disable throttling for the current session.

If you want throttling to be permanent, you will have to modify the
RADIUS response for the user. See [Throttling](#throttling).</code></pre>
<dl>
<dt><code>drop session</code></dt>
<dd>
<p>This will cleanly disconnect the session specified by session ID.</p>
</dd>
<dt><code>drop tunnel</code></dt>
<dd>
<p>This will cleanly disconnect the tunnel specified by tunnel ID, as
well as all sessions on that tunnel.</p>
</dd>
<dt><code>uptime</code></dt>
<dd>
<p>This will show how long the <code>l2tpns</code> process has been
running, and the current bandwidth utilization:</p>
<pre><code>17:10:35 up 8 days, 2212 users, load average: 0.21, 0.17, 0.16
Bandwidth: UDP-ETH:6/6  ETH-UDP:13/13  TOTAL:37.6   IN:3033 OUT:2569</code></pre>
<p>The bandwidth line contains 4 sets of values:</p>
<table>
<tbody>
<tr class="odd">
<td style="text-align: left;">UDP-ETH</td>
<td style="text-align: left;">The current bandwidth going from the LAC
to the ethernet (user uploads), in mbits/sec.</td>
</tr>
<tr class="even">
<td style="text-align: left;">ETH-UDP</td>
<td style="text-align: left;">The current bandwidth going from ethernet
to the LAC (user downloads).</td>
</tr>
<tr class="odd">
<td style="text-align: left;">TOTAL</td>
<td style="text-align: left;">The total aggregate bandwidth in
mbits/s.</td>
</tr>
<tr class="even">
<td style="text-align: left;">IN and OUT</td>
<td style="text-align: left;">Packets/per-second going between UDP-ETH
and ETH-UDP.</td>
</tr>
</tbody>
</table>
<p>These counters are updated every second.</p>
</dd>
<dt><code>configure terminal</code></dt>
<dd>
<p>Enter configuration mode. Use <code>exit</code> or <code>^Z</code> to
exit this mode.</p>
<p>The following commands are valid in this mode:</p>
<p><code>load plugin             name</code></p>
<p>: Load a plugin. You must specify the plugin name, and it will search
in <code>/usr/lib/l2tpns</code> for <code>name.so</code>. You can unload
a loaded plugin with
<code>remove plugin                 name</code>.</p>
<dl>
<dt><code>set</code> ...</dt>
<dd>
<p>Set a configuration variable. You must specify the variable name, and
the value. If the value contains any spaces, you should quote the value
with double (") or single (') quotes.</p>
<p>You can set any configuration value in this way, although some may
require a restart to take effect. See <a
href="#config-startup"></a>.</p>
</dd>
<dt><code>router bgp</code> ...</dt>
<dd>
<p>Configure BGP. See <a href="#config-startup-bgp">BGP</a>.</p>
</dd>
<dt><code>ip access-list</code> ...</dt>
<dd>
<p>Configure a named access list. See <a
href="#config-startup-acl">Access Lists</a>.</p>
</dd>
</dl>
</dd>
</dl>
<h2 id="operation-nsctl">nsctl</h2>
<p><code>nsctl</code> sends messages to a running <code>l2tpns</code>
instance to be control plugins.</p>
<p>Arguments are <code>command</code> and optional args. See
<code>nsctl(8)</code>.</p>
<p>Built-in command are <code>load_plugin</code>,
<code>unload_plugin</code> and <code>help</code>. Any other commands are
passed to plugins for processing by the <code>plugin_control</code>
function.</p>
<h2 id="operation-signals">Signals</h2>
<p>While the process is running, you can send it a few different
signals, using the <code>kill</code> command.</p>
<pre><code>killall -HUP l2tpns</code></pre>
<p>The signals understood are:</p>
<dl>
<dt>SIGHUP</dt>
<dd>
<p>Reload the config from disk and re-open log file.</p>
</dd>
<dt>SIGTERM; SIGINT</dt>
<dd>
<p>Stop process. Tunnels and sessions are not terminated. This signal
should be used to stop <code>l2tpns</code> on a cluster node where there
are other machines to continue handling traffic. See <a
href="#clustering">Clustering</a></p>
</dd>
<dt>SIGQUIT</dt>
<dd>
<p>Shut down tunnels and sessions, exit process when complete.</p>
</dd>
</dl>
<h1 id="throttling">Throttling</h1>
<p><code>l2tpns</code> contains support for slowing down user sessions
to whatever speed you desire. The global setting
<code>throttle_speed</code> defines the default throttle rate.</p>
<p>To throttle a sesion permanently, add a <code>Cisco-AVPair</code>
RADIUS attribute. The <code>autothrotle</code> module interprets the
following attributes:</p>
<table>
<tbody>
<tr class="odd">
<td style="text-align: left;"><code>throttle=yes</code></td>
<td style="text-align: left;">Throttle upstream/downstream traffic to
the configured <code>throttle_speed</code>.</td>
</tr>
<tr class="even">
<td style="text-align: left;"><code>throttle=rate</code></td>
<td style="text-align: left;">Throttle upstream/downstream traffic to
the specified rate Kbps.</td>
</tr>
<tr class="odd">
<td style="text-align: left;">`lcp:interface-config#1=service-policy
input</td>
<td style="text-align: left;">Alternate (Cisco) format: throttle
upstream/downstream to specified rate Kbps.</td>
</tr>
<tr class="even">
<td style="text-align: left;">rate`</td>
<td style="text-align: left;"></td>
</tr>
<tr class="odd">
<td style="text-align: left;">`lcp:interface-config#2=service-policy
output</td>
<td style="text-align: left;"></td>
</tr>
<tr class="even">
<td style="text-align: left;">rate`</td>
<td style="text-align: left;"></td>
</tr>
</tbody>
</table>
<p>You can also enable and disable throttling an active session using
the <a href="#operation-cli-throttle">throttle</a> CLI command.</p>
<h1 id="interception">Interception</h1>
<p>You may have to deal with legal requirements to be able to intercept
a user's traffic at any time. <code>l2tpns</code> allows you to begin
and end interception on the fly, as well as at authentication time.</p>
<p>When a user is being intercepted, a copy of every packet they send
and receive will be sent wrapped in a UDP packet to a specified
host.</p>
<p>The UDP packet contains just the raw IP frame, with no extra headers.
The script <code>scripts/l2tpns-capture</code> may be used as the
end-point for such intercepts, writing the data in PCAP format (suitable
for inspection with <code>tcpdump</code>).</p>
<p>To enable or disable interception of a connected user, use the <a
href="#operation-cli-snoop">snoop</a> and <code>no       snoop</code>
CLI commands. These will enable interception immediately.</p>
<p>If you wish the user to be intercepted whenever they reconnect, you
will need to modify the RADIUS response to include the Vendor-Specific
value <code>Cisco-AVPair="intercept=ip:port"</code>. For this feature to
be enabled, you need to have the <code>autosnoop</code> module
loaded.</p>
<h1 id="plugins">Plugins</h1>
<p>So as to make <code>l2tpns</code> as flexible as possible, a plugin
API is include which you can use to hook into certain events.</p>
<p>There are a some standard modules included which may be used as
examples: <code>autosnoop</code>, <code>autothrottle</code>,
<code>garden</code>, <code>sessionctl</code>, <code>setrxspeed</code>,
<code>snoopctl</code>, <code>stripdomain</code> and
<code>throttlectl</code>.</p>
<p>When an event occurs that has a hook, <code>l2tpns</code> looks for a
predefined function name in every loaded module, and runs them in the
order the modules were loaded.</p>
<p>The function should return <code>PLUGIN_RET_OK</code> if it is all
OK. If it returns <code>PLUGIN_RET_STOP</code>, then it is assumed to
have worked, but that no further modules should be run for this
event.</p>
<p>A return of <code>PLUGIN_RET_ERROR</code> means that this module
failed, and no further processing should be done for this event.</p>
<div class="note">
<p>Use this with care.</p>
</div>
<p>Most event functions take a specific structure named
<code>param_event</code>, which varies in content with each event. The
function name for each event will be <code>plugin_event</code>, so for
the event timer, the function declaration should look like:</p>
<pre><code>int plugin_timer(struct param_timer *data);</code></pre>
<p>A list of the available events follows, with a list of all the fields
in the supplied structure:</p>
<table style="width:97%;">
<colgroup>
<col style="width: 29%" />
<col style="width: 29%" />
<col style="width: 29%" />
<col style="width: 8%" />
</colgroup>
<tbody>
<tr class="odd">
<td rowspan="2"><h1 id="event">Event</h1>
<p><code>plugin_init</code></p></td>
<td rowspan="2"><h1 id="description">Description</h1>
<p>Called when the plugin is loaded. A pointer to a struct containing
function pointers is passed as the only argument, allowing the plugin to
call back into the main code.</p>
<p>Prior to loading the plugin, <code>l2tpns</code> checks the API
version the plugin was compiled against. All plugins should contain:</p>
<pre><code> int</code></pre>
<p>plugin_api_version = PLUGIN_API_VERSION;</p></td>
<td rowspan="2"><h1 id="arguments">Arguments</h1>
<p><code>s  truct pluginfuncs *</code></p></td>
<td></td>
</tr>
<tr class="even">
<td></td>
</tr>
<tr class="odd">
<td>See <code>pluginfuncs</code> structure in <code>plugin.h</code> for
available functions.</td>
<td></td>
<td></td>
<td></td>
</tr>
<tr class="even">
<td><code>plugin_done</code></td>
<td>Called when the plugin is unloaded or <code>l2tpns</code> is
shutdown.</td>
<td><code>void</code></td>
<td></td>
</tr>
<tr class="odd">
<td>No arguments.</td>
<td></td>
<td></td>
<td></td>
</tr>
<tr class="even">
<td><code>plugin_pre_auth</code></td>
<td>Called after a RADIUS response has been received, but before it has
been processed by the code. This will allow you to modify the response
in some way.</td>
<td><code>struct plug in param_pre_auth *</code></td>
<td></td>
</tr>
<tr class="odd">
<td><code>tunnelt *t</code></td>
<td>Tunnel.</td>
<td></td>
<td></td>
</tr>
<tr class="even">
<td><code>sessiont *s</code></td>
<td>Session.</td>
<td></td>
<td></td>
</tr>
<tr class="odd">
<td><code>char *username</code></td>
<td>User name.</td>
<td></td>
<td></td>
</tr>
<tr class="even">
<td><code>char *password</code></td>
<td>Password.</td>
<td></td>
<td></td>
</tr>
<tr class="odd">
<td><code>int protocol</code></td>
<td>Authentication protocol: <code>0xC023</code> for PAP,
<code>0xC223</code> for CHAP.</td>
<td></td>
<td></td>
</tr>
<tr class="even">
<td><code>int continue_auth</code></td>
<td>Set to 0 to stop processing authentication modules.</td>
<td></td>
<td></td>
</tr>
<tr class="odd">
<td><code>plugin_post_auth</code></td>
<td>Called after a RADIUS response has been received, and the basic
checks have been performed. This is what the <code>garden</code> module
uses to force authentication to be accepted.</td>
<td><code>struct plugi n param_post_auth *</code></td>
<td></td>
</tr>
<tr class="even">
<td><code>tunnelt *t</code></td>
<td>Tunnel.</td>
<td></td>
<td></td>
</tr>
<tr class="odd">
<td><code>sessiont *s</code></td>
<td>Session.</td>
<td></td>
<td></td>
</tr>
<tr class="even">
<td><code>char *username</code></td>
<td>User name.</td>
<td></td>
<td></td>
</tr>
<tr class="odd">
<td><code>short auth_allowed</code></td>
<td>Initially true or false depending on whether authentication has been
allowed so far. You can set this to 1 or 0 to force authentication to be
accepted or rejected.</td>
<td></td>
<td></td>
</tr>
<tr class="even">
<td><code>int protocol</code></td>
<td>Authentication protocol: <code>0xC023</code> for PAP,
<code>0xC223</code> for CHAP.</td>
<td></td>
<td></td>
</tr>
<tr class="odd">
<td><code>plugin_timer</code></td>
<td>Run once per second.</td>
<td><code>struct p lugin param_timer *</code></td>
<td></td>
</tr>
<tr class="even">
<td><code>time_t time_now</code></td>
<td>The current unix timestamp.</td>
<td></td>
<td></td>
</tr>
<tr class="odd">
<td><code>plugin_new_session</code></td>
<td>Called after a session is fully set up. The session is now ready to
handle traffic.</td>
<td><code>struct plugin param_new_session *</code></td>
<td></td>
</tr>
<tr class="even">
<td><code>tunnelt *t</code></td>
<td>Tunnel.</td>
<td></td>
<td></td>
</tr>
<tr class="odd">
<td><code>sessiont *s</code></td>
<td>Session.</td>
<td></td>
<td></td>
</tr>
<tr class="even">
<td><code>plugin_kill_session</code></td>
<td>Called when a session is about to be shut down. This may be called
multiple times for the same session.</td>
<td><code>struct plugin p aram_kill_session *</code></td>
<td></td>
</tr>
<tr class="odd">
<td><code>tunnelt *t</code></td>
<td>Tunnel.</td>
<td></td>
<td></td>
</tr>
<tr class="even">
<td><code>sessiont *s</code></td>
<td>Session.</td>
<td></td>
<td></td>
</tr>
<tr class="odd">
<td><code>plugin_control</code></td>
<td><p>Called in whenever a <code>nsctl</code> packet is received. This
should handle the packet and form a response if required.</p>
<p>Plugin-specific help strings may be included in the output of
<code>nsctl help</code> by defining a <code>NULL</code> terminated list
of strings as follows:</p>
<pre><code>char</code></pre>
<p>*plugin_control_hel p[] = { …, NULL };</p></td>
<td><code>struct plu gin param_control *</code></td>
<td></td>
</tr>
<tr class="even">
<td><code>int iam_master</code></td>
<td>If true, this node is the cluster master.</td>
<td></td>
<td></td>
</tr>
<tr class="odd">
<td><code>int argc</code></td>
<td><code>nsctl</code> arguments.</td>
<td></td>
<td></td>
</tr>
<tr class="even">
<td><code>char **argc</code></td>
<td></td>
<td></td>
<td></td>
</tr>
<tr class="odd">
<td><code>int response</code></td>
<td>Response from control message (if handled): should be either
<code>NSCTL_RES_OK</code> or <code>NSCTL_RES_ERR</code>.</td>
<td></td>
<td></td>
</tr>
<tr class="even">
<td><code>char *additional</code></td>
<td>Additional information, output by <code>nsctl</code> on receiving
the response.</td>
<td></td>
<td></td>
</tr>
<tr class="odd">
<td colspan="4"><code>plugin_radius_response</code> | Called whenever a
|
<code>struct plugin para  |                      | RADIUS response      | m_radius_response *</code>
| | includes a | | | <code>Cisco-AVPair</code> | | | value. The value is
| | | split into | | | key<code>=</code>value pairs. | | | Will be
called once | | | for each pair in the | | | response. | |</td>
</tr>
<tr class="even">
<td><code>tunnelt *t</code></td>
<td>Tunnel.</td>
<td></td>
<td></td>
</tr>
<tr class="odd">
<td><code>sessiont *s</code></td>
<td>Session.</td>
<td></td>
<td></td>
</tr>
<tr class="even">
<td><code>char *key</code></td>
<td>Key and value.</td>
<td></td>
<td></td>
</tr>
<tr class="odd">
<td><code>char *value</code></td>
<td></td>
<td></td>
<td></td>
</tr>
<tr class="even">
<td colspan="4"><code>plugin_radius_reset</code> | Called whenever a |
<code>struct p            |                      | RADIUS CoA request   | aram_radius_reset *</code>
| | is received to reset | | | any options to | | | default values | | |
before the new | | | values are applied. | |</td>
</tr>
<tr class="odd">
<td><code>tunnelt *t</code></td>
<td>Tunnel.</td>
<td></td>
<td></td>
</tr>
<tr class="even">
<td><code>sessiont *s</code></td>
<td>Session.</td>
<td></td>
<td></td>
</tr>
<tr class="odd">
<td colspan="4"><code>plugin_radius_account</code> | Called when |
<code>struct par          |                      | preparing a RADIUS   | am_radius_account *</code>
| | accounting record to | | | allow additional | | | data to be added
to | | | the packet. | |</td>
</tr>
<tr class="even">
<td><code>tunnelt *t</code></td>
<td>Tunnel.</td>
<td></td>
<td></td>
</tr>
<tr class="odd">
<td><code>sessiont *s</code></td>
<td>Session.</td>
<td></td>
<td></td>
</tr>
<tr class="even">
<td><code>uint8_t **packet</code></td>
<td>Pointer to the end of the currently assembled packet buffer. The
value should be incremented by the length of any data added.</td>
<td></td>
<td></td>
</tr>
<tr class="odd">
<td colspan="4"><code>plugin_become_master</code> | Called when a node |
<code>void</code> | | elects itself | | | cluster master. | |</td>
</tr>
<tr class="even">
<td>No arguments.</td>
<td></td>
<td></td>
<td></td>
</tr>
<tr class="odd">
<td colspan="4"><code>plugin_new_session_master</code> | Called once for
each | <code>sessiont *</code> | open session on | | | becoming cluster
| | | master. | |</td>
</tr>
<tr class="even">
<td>Session.</td>
<td></td>
<td></td>
<td></td>
</tr>
</tbody>
</table>
<h1 id="walled-garden">Walled Garden</h1>
<p>A "Walled Garden" is implemented so that you can provide perhaps
limited service to sessions that incorrectly authenticate.</p>
<p>Whenever a session provides incorrect authentication, and the RADIUS
server responds with Auth-Reject, the walled garden module (if loaded)
will force authentication to succeed, but set the
<code>walled_garden</code> flag in the session structure, and adds an
<code>iptables</code> rule to the <code>garden_users</code> chain to
cause all packets for the session to traverse the <code>garden</code>
chain.</p>
<p>This doesn't <em>just work</em>. To set this all up, you will to
setup the <code>garden</code> nat table with the <a
href="#config-build-garden">build-garden</a> script with rules to limit
user's traffic.</p>
<p>For example, to force all traffic except DNS to be forwarded to
192.168.1.1, add these entries to your <code>build-garden</code>
script:</p>
<pre><code>iptables -t nat -A garden -p tcp --dport ! 53 -j DNAT --to 192.168.1.1
iptables -t nat -A garden -p udp --dport ! 53 -j DNAT --to 192.168.1.1</code></pre>
<p><code>l2tpns</code> will add entries to the <code>garden_users</code>
chain as appropriate.</p>
<p>You can check the amount of traffic being captured using the
following command:</p>
<pre><code>iptables -t nat -L garden -nvx</code></pre>
<h1 id="filtering">Filtering</h1>
<p>Sessions may be filtered by specifying <code>Filter-Id</code>
attributes in the RADIUS reply. filter.<code>in</code> specifies that
the named access-list filter should be applied to traffic from the
customer, filter.<code>out</code> specifies a list for traffic to the
customer.</p>
<h1 id="clustering">Clustering</h1>
<p>An <code>l2tpns</code> cluster consists of one* or more servers
configured with the same configuration, notably the multicast
<code>cluster_address</code> and the <code>cluster_port</code></p>
<p>*A stand-alone server is simply a degraded cluster.</p>
<p>Initially servers come up as cluster slaves, and periodically (every
<code>cluster_hb_interval</code>/10 seconds) send out ping packets
containing the start time of the process to the multicast
<code>cluster_address</code> on <code>cluster_port</code>.</p>
<p>A cluster master sends heartbeat rather than ping packets, which
contain those session and tunnel changes since the last heartbeat.</p>
<p>When a slave has not seen a heartbeat within
<code>cluster_hb_timeout</code>/10 seconds it "elects" a new master by
examining the list of peers it has seen pings from and determines which
of these and itself is the "best" candidate to be master. "Best" in this
context means the server with the highest uptime (the highest IP address
is used as a tie-breaker in the case of equal uptimes).</p>
<p>After discovering a master, and determining that it is up-to-date
(has seen an update for all in-use sessions and tunnels from heartbeat
packets) will raise a route (see <a href="#routing">Routing</a>) for the
<code>bind_address</code> and for all addresses/networks in
<code>ip_pool</code>.</p>
<p>Any packets received by the slave which would alter the session
state, as well as packets for throttled or gardened sessions are
forwarded to the master for handling. In addition, byte counters for
session traffic are periodically forwarded.</p>
<p>The heartbeat exchange use timing, so the master and slaves need to
have their time synchronized, e.g. with ntp.</p>
<p>Since all the masters and slaves should be able to receive all
packets, <code>bind_address</code> needs to be used to make all of them
agree on the IP to be used. If <code>l2tpns</code> is to behave as LAC,
<code>bind_address_remotelns</code> also needs to be set.</p>
<p>To balance the load, <code>bind_multi_address</code> can be used to
make them support several IPs, and one can then prioritize one or the
other on each server, and make e.g. radius pick up one of them randomly
for <code>Tunnel-Server-Endpoint</code>. There is currently no support
for such load balancing for LAC support.</p>
<p>*Configurable with <code>cluster_master_min_adv</code></p>
<p>Multiple clusters can be run on the same network by just using
different multicast <code>cluster_address</code>. However, for a given
host to be part of multiple clusters without mixing the clusters,
<code>cluster_port</code> must be different for each cluster.</p>
<h1 id="routing">Routing</h1>
<p>If you are running a single instance, you may simply statically route
the IP pools to the <code>bind_address</code> (<code>l2tpns</code> will
send a gratuitous arp).</p>
<p>For a cluster, configure the members as BGP neighbours on your router
and configure multi-path load-balancing. Cisco uses
<code>maximum-paths ibgp</code> for IBGP. If this is not supported by
your IOS revision, you can use <code>maximum-paths</code> (which works
for EBGP) and set <code>as_number</code> to a private value such as
64512.</p>