=head1 NAME

nbdkit-sh-plugin - nbdkit shell, script or executable plugin

=head1 SYNOPSIS

 nbdkit sh /path/to/script [arguments...]

=for paragraph

 nbdkit sh - <<'EOF'
 ... shell script ...
 EOF

=head1 DESCRIPTION

C<nbdkit-sh-plugin> allows you to write plugins for L<nbdkit(1)> using
arbitrary scripting languages, including shells like L<bash(1)>,
L<dash(1)>, L<csh(1)>, L<zsh(1)> etc., other scripting environments,
or any executable.

Note if you want to use an established scripting language like Perl or
Python, then nbdkit has specific plugins to handle those languages and
those will be more efficient (see L<nbdkit(1)> for a complete list).

To use shell script fragments from the nbdkit command line (rather
than a separate script) see L<nbdkit-eval-plugin(1)>.

=head2 If you have been given an nbdkit sh plugin

Assuming you have a shell script which is an nbdkit plugin, you run it
like this:

 nbdkit sh /path/to/script

You may have to add further C<key=value> arguments to the command
line.  The script must be executable (C<chmod +x>).

=head2 Inline shell scripts

It is also possible to write a shell script plugin "inline" using C<->
as the name of the script, like this:

 nbdkit sh - <<'EOF'
   case "$1" in
     get_size) echo 1M ;;
     pread) dd if=/dev/zero count=$3 iflag=count_bytes ;;
     *) exit 2 ;;
   esac
 EOF

By default the inline script runs under F</bin/sh>.  You can add a
shebang (C<#!>) to use other scripting languages.  Of course, reading
an inline script from stdin is incompatible with the I<-s>
(I<--single>) mode of nbdkit that connects a client on stdin.

=head1 WRITING AN NBDKIT SH PLUGIN

For example plugins written in Bash, see:
L<https://gitlab.com/nbdkit/nbdkit/blob/master/plugins/sh/examples/>

Broadly speaking, nbdkit shell plugins work like C ones, so you should
read L<nbdkit-plugin(3)> first.

=head2 Programming model

This plugin has a simple programming model: For every plugin method
that needs to be called, the external script is invoked with
parameters describing the method and its arguments.  The first
parameter is always the method name.  For example:

 /path/to/script config file disk.img
                   │      │   │
                   │      │   └─ value ($3)
                   │      └── key ($2)
               method ($1)

=for paragraph

 /path/to/script pread <handle> <count> <offset>
                   │       │       │       │
                   │       │       │       └─ offset in bytes ($4)
                   │       │       └── request size in bytes ($3)
               method ($1) └── handle ($2) ─ see "Handles" below

Scripts should ignore extra parameters that they don't understand
since we may add new parameters in future.

=head2 Exit codes

The script should exit with specific exit codes:

=over 4

=item S<0>

The method was executed successfully.

=item 1 and 16-255

There was an error.  The script may print on stderr an errno name,
optionally followed by whitespace and a message, for example:

 echo 'ENOSPC Out of space' >&2
 exit 1

or if you don't need the log message:

 echo ENOSPC >&2
 exit 1

If the script doesn't print anything or the output cannot be parsed
then nbdkit assumes error C<EIO>.  Note that output to stderr is
ignored if the command succeeds, so it is acceptable to output a
potential error message prefix prior to attempting a command which
will add further details if a failure occurs.

=item S<2>

The requested method is not supported by the script.

=item S<3>

For methods which return booleans, this code indicates false.

=item 4 and 5

Triggers a call to the C function L<nbdkit_shutdown(3)>, which requests
an asynchronous exit of the nbdkit server (disconnecting all clients).
The client will usually get a response before shutdown is complete
(although this is racy); so once the shutdown is requested, code 4
then behaves like code 0 (stderr is ignored, and the server tries to
return success), and code 5 behaves like code 1 (the server tries to
return an error to the client parsed from stderr, although a missing
error defaults to C<ESHUTDOWN> instead of C<EIO>).

=item S<6>

Triggers a call to the C function L<nbdkit_disconnect(3)> with C<force>
set to true, which requests an abrupt disconnect of the current
client.  The contents of stderr are irrelevant with this status, since
the client will not get a response.

=item 7 and 8

Triggers a call to the C function L<nbdkit_disconnect(3)> with C<force>
set to false, which requests a soft disconnect of the current client
(future client requests are rejected with C<ESHUTDOWN> without calling
into the plugin, but current requests may complete).  Since the client
will likely get the response to this command, code 7 then behaves like
code 0 (stderr is ignored, and the server tries to return success),
and code 8 behaves like code 1 (the server tries to return an error to
the client parsed from stderr, although a missing error defaults to
C<ESHUTDOWN> instead of C<EIO>).

=item 9-15

These exit codes are reserved for future use.  Note that versions of
nbdkit E<lt> 1.34 documented that codes 8 through 15 behaved like code
1; although it is unlikely that many scripts relied on this similarity
in practice.

=back

In nbdkit E<gt> 1.34, it is possible to probe whether additional exit
codes have been assigned meaning, by looking for the line
B<max_known_status=> in the output of B<nbdkit --dump-plugin sh>.  If
this line is not present, exit codes 4 and above behave like status 1.

=head2 Temporary directory

A fresh script is invoked for each method call (ie. scripts are
stateless), so if the script needs to store state it has to store it
somewhere in the filesystem in a format and location which is left up
to the author of the script.

However nbdkit helps by creating a randomly named, empty directory for
the script.  This directory persists for the lifetime of nbdkit and is
deleted when nbdkit exits.  The name of the directory is passed to
each script invocation in the C<$tmpdir> environment variable.

=head2 Handles

Handles are arbitrary strings, but it is best to limit them to short
alphanumeric strings.

=head3 Per-connection state

The temporary directory described above can be used for state for the
lifetime of the nbdkit instance (across multiple connections).  If you
want to store state per connection then one way to do it is to create
a randomly named subdirectory under the temporary directory:

 case "$1" in
   ...
   open)
     mktemp -d $tmpdir/handle-XXXXXX ;;

The handle will be the subdirectory name, returned to the script as
C<$2> in all connected calls (eg. C<pread>, C<get_size>).  You can
delete the subdirectory explicitly in C<close>:

 case "$1" in
   ...
   close)
     rm -rf "$2" ;;

or rely on nbdkit deleting the whole temporary directory including all
per-handle subdirectories when it exits.

=head2 Performance

This plugin has to fork on every request, so performance will never be
great.  For best performance, consider using the L<nbdkit-plugin(3)>
API directly.  Having said that, if you have a sh plugin and want to
improve performance then the following tips may help:

=over 4

=item Relax the thread model.

The default C<thread_model> is C<serialize_all_requests> meaning that
two instances of the script can never be running at the same time.
This is safe but slow.  If your script is safe to be called in
parallel, set this to C<parallel>.

=item Implement the C<zero> method.

If the C<zero> method is not implemented then nbdkit will fall back to
using C<pwrite> which is considerably slower because nbdkit has to
send blocks of zeroes to the script.

=item You don't have to write shell scripts.

This plugin can run any external binary, not only shell scripts.  You
should get more performance by rewriting the shell script as a program
in a compiled language.

=back

=head2 Methods

This just documents the arguments to the script corresponding to each
plugin method, and any way that they differ from the C callbacks.  In
all other respects they work the same way as the C callbacks, so you
should go and read L<nbdkit-plugin(3)>.

=over 4

=item C<load>

 /path/to/script load

=item C<unload>

 /path/to/script unload

This is called just before nbdkit exits.  Errors from this method are
ignored.

=item C<dump_plugin>

 /path/to/script dump_plugin

=item C<config>

 /path/to/script config <key> <value>

=item C<config_complete>

 /path/to/script config_complete

=item C<magic_config_key>

 /path/to/script magic_config_key

If a magic config key is needed, this should echo it to stdout.
See L<nbdkit(1)/Magic parameters>.

=item C<thread_model>

 /path/to/script thread_model

On success this should print the desired thread model of the script,
one of C<"serialize_connections">, C<"serialize_all_requests">,
C<"serialize_requests">, or C<"parallel">.

This method is I<not> required; if omitted, then the plugin will be
executed under the safe C<"serialize_all_requests"> model.  However,
this means that this method B<must> be provided if you want to use the
C<"parallel"> or C<"serialize_requests"> model.  Even then your
request may be restricted for other reasons; look for C<thread_model>
in the output of C<nbdkit --dump-plugin sh script> to see what
actually gets selected.

If an error occurs, the script should output an error message and exit
with status C<1>; unrecognized output is ignored.

=item C<get_ready>

 /path/to/script get_ready

=item C<after_fork>

 /path/to/script after_fork

=item C<preconnect>

 /path/to/script preconnect <readonly>

=item C<list_exports>

 /path/to/script list_exports <readonly> <tls>

The C<readonly> and C<tls> parameters will be C<true> or C<false>.

The first line of output informs nbdkit how to parse the rest of the
output, the remaining lines then supply the inputs of the C
C<nbdkit_add_export> function (see L<nbdkit-plugin(3)>), as follows:

=over 4

=item NAMES

The remaining output provides one export name per line, and no export
will be given a description.  For convenience, this form is also
assumed if the first output line does not match one of the recognized
parse modes.

=item INTERLEAVED

The remaining output provides pairs of lines, the first line being an
export name, and the second the corresponding description.

=item NAMES+DESCRIPTIONS

The number of remaining lines is counted, with the first half being
used as export names, and the second half providing descriptions to
pair with names from the first half.

An example of using this form to list files in the current directory,
followed by their L<ls(1)> long description, would be:

 echo NAMES+DESCRIPTIONS
 ls
 ls -l

=back

Note that other output modes might be introduced in the future; in
particular, none of the existing modes allow a literal newline in an
export name or description, although this could be possible under a
new mode supporting escape sequences.

This method is I<not> required; if it is absent, the list of exports
advertised by nbdkit will be the single name result of C<default_export>
and no description.

=item C<default_export>

 /path/to/script default_export <readonly> <tls>

The C<readonly> and C<tls> parameters will be C<true> or C<false>.

On success this should print a name on stdout to use in place of the
default export C<"">, then exit with code C<0>.  For convenience, the
output can be any of the list forms recognized by C<list_exports>, in
which case the first listed export name is used, and where an empty
list uses C<"">.  Given the current set of recognized export lists, it
is not possible for the resulting name to include a newline.

This method is I<not> required; if it is absent, the default export
name will be the empty string, C<"">.

=item C<open>

 /path/to/script open <readonly> <exportname> <tls>

The C<readonly> parameter will be C<true> or C<false>.  The
C<exportname> parameter, if present, is the export name passed to the
server from the client.  The C<tls> parameter, if present, will be
C<true> or C<false> depending on whether the client is using TLS.

On success this should print the handle (any string) on stdout and
exit with code C<0>.  If the handle ends with a newline character then
the newline is removed.

Unlike C plugins, this method is I<not> required.  If omitted then the
handle will be C<""> (empty string).

=item C<close>

 /path/to/script close <handle>

=item C<export_description>

 /path/to/script export_description <handle>

The script should print a human-readable description of the disk image
on stdout.  If the description ends with a newline character then the
newline is removed.

This method is I<not> required; if it is absent, no export description
will be provided to the client.

=item C<get_size>

 /path/to/script get_size <handle>

The script should print the size of the disk image on stdout.  You can
print the size in bytes, or use any format understood by
L<nbdkit_parse_size(3)> such as C<1M>.

This method is required.

=item C<block_size>

 /path/to/script block_size <handle>

This script should print three numbers on stdout, separated by
whitespace.  These are (in order) the minimum block size, the
preferred block size, and the maximum block size.  You can print the
sizes in bytes or use any format understood by L<nbdkit_parse_size(3)>
such as C<1M>.

=item C<can_write>

=item C<can_flush>

=item C<can_trim>

=item C<can_zero>

=item C<can_extents>

Unlike in other languages, you B<must> provide the C<can_*> methods
otherwise they are assumed to all return false and your C<pwrite>,
C<flush>, C<trim>, C<zero> and C<extents> methods will never be
called.  The reason for this is obscure: In other languages we can
detect if (eg) a C<pwrite> method is defined and synthesize an
appropriate response if no actual C<can_write> method is defined.
However detecting if methods are present without running them is not
possible with this plugin.

 /path/to/script can_write <handle>
 /path/to/script can_flush <handle>
 /path/to/script can_trim <handle>
 /path/to/script can_zero <handle>
 /path/to/script can_extents <handle>

The script should exit with code C<0> for true or code C<3> for false.

=item C<is_rotational>

=item C<can_fast_zero>

 /path/to/script is_rotational <handle>
 /path/to/script can_fast_zero <handle>

The script should exit with code C<0> for true or code C<3> for false.

=item C<can_fua>

=item C<can_cache>

 /path/to/script can_fua <handle>
 /path/to/script can_cache <handle>

These control Forced Unit Access (FUA) and caching behaviour of the
core server.

Unlike the other C<can_*> callbacks, these two are I<not> a boolean.
They must print either "none", "emulate" or "native" to stdout.  The
meaning of these is described in L<nbdkit-plugin(3)>.  Furthermore,
you B<must> provide a C<can_cache> method if you desire the C<cache>
callback to be utilized, similar to the reasoning behind requiring
C<can_write> to utilize C<pwrite>.

=item C<can_multi_conn>

 /path/to/script can_multi_conn <handle>

The script should exit with code C<0> for true or code C<3> for false.

=item C<pread>

 /path/to/script pread <handle> <count> <offset>

The script should print the requested binary data on stdout.  Exactly
C<count> bytes must be printed.

This method is required.

=item C<pwrite>

 /path/to/script pwrite <handle> <count> <offset> <flags>

The script should read the binary data to be written from stdin.

The C<flags> parameter can be an empty string or C<"fua">.  In the
future, a comma-separated list of flags may be present.

Unlike in other languages, if you provide a C<pwrite> method you
B<must> also provide a C<can_write> method which exits with code C<0>
(true).

=item C<flush>

 /path/to/script flush <handle>

Unlike in other languages, if you provide a C<flush> method you
B<must> also provide a C<can_flush> method which exits with code C<0>
(true).

=item C<trim>

 /path/to/script trim <handle> <count> <offset> <flags>

The C<flags> parameter can be an empty string or C<"fua">.  In the
future, a comma-separated list of flags may be present.

Unlike in other languages, if you provide a C<trim> method you B<must>
also provide a C<can_trim> method which exits with code C<0> (true).

=item C<zero>

 /path/to/script zero <handle> <count> <offset> <flags>

The C<flags> parameter can be an empty string or a comma-separated
list of the flags: C<"fua">, C<"may_trim">, and C<"fast"> (eg. C<"">,
C<"fua">, C<"fua,may_trim,fast"> are some of the 8 possible values).

Unlike in other languages, if you provide a C<zero> method you B<must>
also provide a C<can_zero> method which exits with code C<0> (true).

To trigger a fallback to C<pwrite> on a normal zero request, or to
respond quickly to the C<"fast"> flag that a specific zero request is
no faster than a corresponding write, the script must output
C<ENOTSUP> or C<EOPNOTSUPP> to stderr (possibly followed by a
description of the problem) before exiting with code C<1> (failure).

=item C<extents>

 /path/to/script extents <handle> <count> <offset> <flags>

The C<flags> parameter can be an empty string or C<"req_one">.

This must print, one per line on stdout, a list of one or more extents
in the format:

 offset length type

which correspond to the inputs of the C C<nbdkit_add_extent> function
(see L<nbdkit-plugin(3)>).  The C<offset> and C<length> fields may use
any format understood by L<nbdkit_parse_size(3)>.  The optional
C<type> field may be an integer, missing (same as 0), or a
comma-separated list of the words C<hole> and C<zero>.  Blank lines
and lines beginning with C<#> are comments.  An example of a valid set
of extents covering a C<10M> disk where the first megabyte only is
allocated data:

 0  1M
 1M 9M  hole,zero

Unlike in other languages, if you provide an C<extents> method you
B<must> also provide a C<can_extents> method which exits with code
C<0> (true).

=item C<cache>

 /path/to/script cache <handle> <count> <offset>

Unlike in other languages, if you provide a C<cache> method you
B<must> also provide a C<can_cache> method which prints "native" and
exits with code C<0> (true).

=back

=head2 Missing callbacks

=over 4

=item Missing: C<name>, C<version>, C<longname>,
C<description>, C<config_help>

These are not yet supported.

=back

=head1 FILES

=over 4

=item F<$plugindir/nbdkit-sh-plugin.so>

The plugin.

Use C<nbdkit --dump-config> to find the location of C<$plugindir>.

=back

=head1 VERSION

C<nbdkit-sh-plugin> first appeared in nbdkit 1.8.

=head1 SEE ALSO

L<nbdkit(1)>,
L<nbdkit-plugin(3)>,
L<nbdkit-eval-plugin(1)>,
L<nbdkit-cc-plugin(1)>.

=head1 AUTHORS

Richard W.M. Jones

=head1 COPYRIGHT

Copyright Red Hat