3 collectd-perl - Documentation of collectd's C<perl plugin>
12 IncludeDir "/path/to/perl/plugins"
13 BaseName "Collectd::Plugins"
24 The C<perl plugin> embeds a Perl-interpreter into collectd and provides an
25 interface to collectd's plugin system. This makes it possible to write plugins
26 for collectd in Perl. This is a lot more efficient than executing a
27 Perl-script every time you want to read a value with the C<exec plugin> (see
28 L<collectd-exec(5)>) and provides a lot more functionality, too.
30 When loading the C<perl plugin>, the B<Globals> option should be enabled.
31 Else, the perl plugin will fail to load any Perl modules implemented in C,
32 which includes, amongst many others, the B<threads> module used by the plugin
33 itself. See the documentation of the B<Globals> option in L<collectd.conf(5)>
40 =item B<LoadPlugin> I<Plugin>
42 Loads the Perl plugin I<Plugin>. This does basically the same as B<use> would
43 do in a Perl program. As a side effect, the first occurrence of this option
44 causes the Perl-interpreter to be initialized.
46 =item B<BaseName> I<Name>
48 Prepends I<Name>B<::> to all plugin names loaded after this option. This is
49 provided for convenience to keep plugin names short. All Perl-based plugins
50 provided with the I<collectd> distributions reside in the C<Collectd::Plugins>
53 =item E<lt>B<Plugin> I<Name>E<gt> block
55 This block may be used to pass on configuration settings to a Perl plugin. The
56 configuration is converted into a config-item data type which is passed to the
57 registered configuration callback. See below for details about the config-item
58 data type and how to register callbacks.
60 The I<name> identifies the callback. It is used literally and independent of
61 the B<BaseName> setting.
63 =item B<EnableDebugger> I<Package>[=I<option>,...]
65 Run collectd under the control of the Perl source debugger. If I<Package> is
66 not the empty string, control is passed to the debugging, profiling, or
67 tracing module installed as Devel::I<Package>. A comma-separated list of
68 options may be specified after the "=" character. Please note that you may not
69 leave out the I<Package> option even if you specify B<"">. This is the same as
70 using the B<-d:Package> command line option.
72 See L<perldebug> for detailed documentation about debugging Perl.
74 This option does not prevent collectd from daemonizing, so you should start
75 collectd with the B<-f> command line option. Else you will not be able to use
76 the command line driven interface of the debugger.
78 =item B<IncludeDir> I<Dir>
80 Adds I<Dir> to the B<@INC> array. This is the same as using the B<-IDir>
81 command line option or B<use lib Dir> in the source code. Please note that it
82 only has effect on plugins loaded after this option.
86 =head1 WRITING YOUR OWN PLUGINS
88 Writing your own plugins is quite simple. collectd manages plugins by means of
89 B<dispatch functions> which call the appropriate B<callback functions>
90 registered by the plugins. Any plugin basically consists of the implementation
91 of these callback functions and initializing code which registers the
92 functions with collectd. See the section "EXAMPLES" below for a really basic
93 example. The following types of B<callback functions> are known to collectd
94 (all of them are optional):
98 =item configuration functions
100 This type of functions is called during configuration if an appropriate
101 B<Plugin> block has been encountered. It is called once for each B<Plugin>
102 block which matches the name of the callback as provided with the
103 B<plugin_register> method - see below.
107 This type of functions is called once after loading the module and before any
108 calls to the read and write functions. It should be used to initialize the
109 internal state of the plugin (e.E<nbsp>g. open sockets, ...). If the return
110 value evaluates to B<false>, the plugin will be disabled.
114 This type of function is used to collect the actual data. It is called once
115 per interval (see the B<Interval> configuration option of collectd). Usually
116 it will call B<plugin_dispatch_values> to dispatch the values to collectd
117 which will pass them on to all registered B<write functions>. If the return
118 value evaluates to B<false> the plugin will be skipped for an increasing
119 amount of time until it returns B<true> again.
121 =item write functions
123 This type of function is used to write the dispatched values. It is called
124 once for each call to B<plugin_dispatch_values>.
126 =item flush functions
128 This type of function is used to flush internal caches of plugins. It is
129 usually triggered by the user only. Any plugin which caches data before
130 writing it to disk should provide this kind of callback function.
134 This type of function is used to pass messages of plugins or the daemon itself
137 =item notification function
139 This type of function is used to act upon notifications. In general, a
140 notification is a status message that may be associated with a data instance.
141 Usually, a notification is generated by the daemon if a configured threshold
142 has been exceeded (see the section "THRESHOLD CONFIGURATION" in
143 L<collectd.conf(5)> for more details), but any plugin may dispatch
144 notifications as well.
146 =item shutdown functions
148 This type of function is called once before the daemon shuts down. It should
149 be used to clean up the plugin (e.g. close sockets, ...).
153 Any function (except log functions) may set the B<$@> variable to describe
154 errors in more detail. The message will be passed on to the user using
155 collectd's logging mechanism.
157 See the documentation of the B<plugin_register> method in the section
158 "METHODS" below for the number and types of arguments passed to each
159 B<callback function>. This section also explains how to register B<callback
160 functions> with collectd.
162 To enable a plugin, copy it to a place where Perl can find it (i.E<nbsp>e. a
163 directory listed in the B<@INC> array) just as any other Perl plugin and add
164 an appropriate B<LoadPlugin> option to the configuration file. After
165 restarting collectd you're done.
169 The following complex types are used to pass values between the Perl plugin
176 A config-item is one structure which keeps the information provided in the
177 configuration file. The array of children keeps one entry for each
178 configuration option. Each such entry is another config-item structure, which
179 may nest further if nested blocks are used.
183 values => [ val1, val2, ... ],
184 children => [ { ... }, { ... }, ... ]
189 A data-set is a list of one or more data-sources. Each data-source defines a
190 name, type, min- and max-value and the data-set wraps them up into one
191 structure. The general layout looks like this:
194 name => 'data_source_name',
195 type => DS_TYPE_COUNTER || DS_TYPE_GAUGE || DS_TYPE_DERIVE || DS_TYPE_ABSOLUTE,
196 min => value || undef,
197 max => value || undef
202 A value-list is one structure which features an array of values and fields to
203 identify the values, i.E<nbsp>e. time and host, plugin name and
204 plugin-instance as well as a type and type-instance. Since the "type" is not
205 included in the value-list but is passed as an extra argument, the general
206 layout looks like this:
209 values => [123, 0.5],
211 interval => $interval_g,
213 plugin => 'myplugin',
215 plugin_instance => '',
221 A notification is one structure defining the severity, time and message of the
222 status message as well as an identification of a data instance. Also, it
223 includes an optional list of user-defined meta information represented as
227 severity => NOTIF_FAILURE || NOTIF_WARNING || NOTIF_OKAY,
229 message => 'status message',
231 plugin => 'myplugin',
233 plugin_instance => '',
235 meta => [ { name => <name>, value => <value> }, ... ]
240 A match-proc is one structure storing the callbacks of a "match" of the filter
241 chain infrastructure. The general layout looks like this:
244 create => 'my_create',
245 destroy => 'my_destroy',
251 A target-proc is one structure storing the callbacks of a "target" of the
252 filter chain infrastructure. The general layout looks like this:
255 create => 'my_create',
256 destroy => 'my_destroy',
257 invoke => 'my_invoke'
264 The following functions provide the C-interface to Perl-modules. They are
265 exported by the ":plugin" export tag (see the section "EXPORTS" below).
269 =item B<plugin_register> (I<type>, I<name>, I<data>)
271 Registers a callback-function or data-set.
273 I<type> can be one of:
297 I<name> is the name of the callback-function or the type of the data-set,
298 depending on the value of I<type>. (Please note that the type of the data-set
299 is the value passed as I<name> here and has nothing to do with the I<type>
300 argument which simply tells B<plugin_register> what is being registered.)
302 The last argument, I<data>, is either a function name or an array-reference.
303 If I<type> is B<TYPE_DATASET>, then the I<data> argument must be an
304 array-reference which points to an array of hashes. Each hash describes one
305 data-set. For the exact layout see B<Data-Set> above. Please note that
306 there is a large number of predefined data-sets available in the B<types.db>
307 file which are automatically registered with collectd - see L<types.db(5)> for
308 a description of the format of this file.
310 B<Note>: Using B<plugin_register> to register a data-set is deprecated. Add
311 the new type to a custom L<types.db(5)> file instead. This functionality might
312 be removed in a future version of collectd.
314 If the I<type> argument is any of the other types (B<TYPE_INIT>, B<TYPE_READ>,
315 ...) then I<data> is expected to be a function name. If the name is not
316 prefixed with the plugin's package name collectd will add it automatically.
317 The interface slightly differs from the C interface (which expects a function
318 pointer instead) because Perl does not support to share references to
319 subroutines between threads.
321 These functions are called in the various stages of the daemon (see the
322 section "WRITING YOUR OWN PLUGINS" above) and are passed the following
329 The only argument passed is I<config-item>. See above for the layout of this
338 No arguments are passed.
342 The arguments passed are I<type>, I<data-set>, and I<value-list>. I<type> is a
343 string. For the layout of I<data-set> and I<value-list> see above.
347 The arguments passed are I<timeout> and I<identifier>. I<timeout> indicates
348 that only data older than I<timeout> seconds is to be flushed. I<identifier>
349 specifies which values are to be flushed.
353 The arguments are I<log-level> and I<message>. The log level is small for
354 important messages and high for less important messages. The least important
355 level is B<LOG_DEBUG>, the most important level is B<LOG_ERR>. In between there
356 are (from least to most important): B<LOG_INFO>, B<LOG_NOTICE>, and
357 B<LOG_WARNING>. I<message> is simply a string B<without> a newline at the end.
361 The only argument passed is I<notification>. See above for the layout of this
366 =item B<plugin_unregister> (I<type>, I<plugin>)
368 Removes a callback or data-set from collectd's internal list of
369 functionsE<nbsp>/ datasets.
371 =item B<plugin_dispatch_values> (I<value-list>)
373 Submits a I<value-list> to the daemon. If the data-set identified by
374 I<value-list>->{I<type>}
375 is found (and the number of values matches the number of data-sources) then the
376 type, data-set and value-list is passed to all write-callbacks that are
377 registered with the daemon.
379 B<Note>: Prior to version 4.4 of collectd, the data-set type used to be passed
380 as the first argument to B<plugin_register>. This syntax is still supported
381 for backwards compatibility but has been deprecated and will be removed in
382 some future version of collectd.
384 =item B<plugin_write> ([B<plugins> => I<...>][, B<datasets> => I<...>],
385 B<valuelists> => I<...>)
387 Calls the write function of the given I<plugins> with the provided I<data
388 sets> and I<value lists>. In contrast to B<plugin_dispatch_values>, it does
389 not update collectd's internal cache and bypasses the filter mechanism (see
390 L<collectd.conf(5)> for details). If the B<plugins> argument has been omitted,
391 the values will be dispatched to all registered write plugins. If the
392 B<datasets> argument has been omitted, the required data sets are looked up
393 according to the C<type> member in the appropriate value list. The value of
394 all three arguments may either be a single scalar or a reference to an array.
395 If the B<datasets> argument has been specified, the number of data sets has to
396 equal the number of specified value lists.
398 =item B<plugin_flush> ([B<timeout> => I<timeout>][, B<plugins> => I<...>][,
399 B<identifiers> => I<...>])
401 Flush one or more plugins. I<timeout> and the specified I<identifiers> are
402 passed on to the registered flush-callbacks. If omitted, the timeout defaults
403 to C<-1>. The identifier defaults to the undefined value. If the B<plugins>
404 argument has been specified, only named plugins will be flushed. The value of
405 the B<plugins> and B<identifiers> arguments may either be a string or a
406 reference to an array of strings.
408 =item B<plugin_flush_one> (I<timeout>, I<plugin>)
410 This is identical to using "plugin_flush (timeout =E<gt> I<timeout>, plugins
413 B<Note>: Starting with version 4.5 of collectd, B<plugin_flush_one> has been
414 deprecated and will be removed in some future version of collectd. Use
415 B<plugin_flush> instead.
417 =item B<plugin_flush_all> (I<timeout>)
419 This is identical to using "plugin_flush (timeout =E<gt> I<timeout>)".
421 B<Note>: Starting with version 4.5 of collectd, B<plugin_flush_all> has been
422 deprecated and will be removed in some future version of collectd. Use
423 B<plugin_flush> instead.
425 =item B<plugin_dispatch_notification> (I<notification>)
427 Submits a I<notification> to the daemon which will then pass it to all
428 notification-callbacks that are registered.
430 =item B<plugin_log> (I<log-level>, I<message>)
432 Submits a I<message> of level I<log-level> to collectd's logging mechanism.
433 The message is passed to all log-callbacks that are registered with collectd.
435 =item B<ERROR>, B<WARNING>, B<NOTICE>, B<INFO>, B<DEBUG> (I<message>)
437 Wrappers around B<plugin_log>, using B<LOG_ERR>, B<LOG_WARNING>,
438 B<LOG_NOTICE>, B<LOG_INFO> and B<LOG_DEBUG> respectively as I<log-level>.
442 The following function provides the filter chain C-interface to Perl-modules.
443 It is exported by the ":filter_chain" export tag (see the section "EXPORTS"
448 =item B<fc_register> (I<type>, I<name>, I<proc>)
450 Registers filter chain callbacks with collectd.
452 I<type> may be any of:
462 I<name> is the name of the match or target. By this name, the callbacks are
463 identified in the configuration file when specifying a B<Match> or B<Target>
464 block (see L<collectd.conf(5)> for details).
466 I<proc> is a hash reference. The hash includes up to three callbacks: an
467 optional constructor (B<create>) and destructor (B<destroy>) and a mandatory
468 B<match> or B<invoke> callback. B<match> is called whenever processing an
469 appropriate match, while B<invoke> is called whenever processing an
470 appropriate target (see the section "FILTER CONFIGURATION" in
471 L<collectd.conf(5)> for details). Just like any other callbacks, filter chain
472 callbacks are identified by the function name rather than a function pointer
473 because Perl does not support to share references to subroutines between
474 threads. The following arguments are passed to the callbacks:
480 The arguments passed are I<config-item> and I<user-data>. See above for the
481 layout of the config-item data-type. I<user-data> is a reference to a scalar
482 value that may be used to store any information specific to this particular
483 instance. The daemon does not care about this information at all. It's for the
488 The only argument passed is I<user-data> which is a reference to the user data
489 initialized in the B<create> callback. This callback may be used to cleanup
490 instance-specific information and settings.
494 The arguments passed are I<data-set>, I<value-list>, I<meta> and I<user-data>.
495 See above for the layout of the data-set and value-list data-types. I<meta> is
496 a pointer to an array of meta information, just like the B<meta> member of the
497 notification data-type (see above). I<user-data> is a reference to the user
498 data initialized in the B<create> callback.
504 =head1 GLOBAL VARIABLES
510 As the name suggests this variable keeps the hostname of the system collectd
511 is running on. The value might be influenced by the B<Hostname> or
512 B<FQDNLookup> configuration options (see L<collectd.conf(5)> for details).
516 This variable keeps the interval in seconds in which the read functions are
517 queried (see the B<Interval> configuration option).
521 Any changes to these variables will be globally visible in collectd.
525 By default no symbols are exported. However, the following export tags are
526 available (B<:all> will export all of them):
534 =item B<plugin_register> ()
536 =item B<plugin_unregister> ()
538 =item B<plugin_dispatch_values> ()
540 =item B<plugin_flush> ()
542 =item B<plugin_flush_one> ()
544 =item B<plugin_flush_all> ()
546 =item B<plugin_dispatch_notification> ()
548 =item B<plugin_log> ()
566 =item B<TYPE_SHUTDOWN>
570 =item B<TYPE_DATASET>
578 =item B<DS_TYPE_COUNTER>
580 =item B<DS_TYPE_GAUGE>
582 =item B<DS_TYPE_DERIVE>
584 =item B<DS_TYPE_ABSOLUTE>
614 =item B<:filter_chain>
620 =item B<FC_MATCH_NO_MATCH>
622 =item B<FC_MATCH_MATCHES>
624 =item B<FC_TARGET_CONTINUE>
626 =item B<FC_TARGET_STOP>
628 =item B<FC_TARGET_RETURN>
646 =item B<NOTIF_FAILURE>
648 =item B<NOTIF_WARNING>
668 Any Perl plugin will start similar to:
670 package Collectd::Plugins::FooBar;
675 use Collectd qw( :all );
677 A very simple read function might look like:
681 my $vl = { plugin => 'foobar' };
682 $vl->{'values'} = [ rand(42) ];
683 plugin_dispatch_values ('gauge', $vl);
687 A very simple write function might look like:
691 my ($type, $ds, $vl) = @_;
692 for (my $i = 0; $i < scalar (@$ds); ++$i) {
693 print "$vl->{'plugin'} ($vl->{'type'}): $vl->{'values'}->[$i]\n";
698 A very simple match callback might look like:
702 my ($ds, $vl, $meta, $user_data) = @_;
703 if (matches($ds, $vl)) {
704 return FC_MATCH_MATCHES;
706 return FC_MATCH_NO_MATCH;
710 To register those functions with collectd:
712 plugin_register (TYPE_READ, "foobar", "foobar_read");
713 plugin_register (TYPE_WRITE, "foobar", "foobar_write");
715 fc_register (FC_MATCH, "foobar", "foobar_match");
717 See the section "DATA TYPES" above for a complete documentation of the data
718 types used by the read, write and match functions.
726 Please feel free to send in new plugins to collectd's mailing list at
727 E<lt>collectdE<nbsp>atE<nbsp>verplant.orgE<gt> for review and, possibly,
728 inclusion in the main distribution. In the latter case, we will take care of
729 keeping the plugin up to date and adapting it to new versions of collectd.
731 Before submitting your plugin, please take a look at
732 L<http://collectd.org/dev-info.shtml>.
742 collectd is heavily multi-threaded. Each collectd thread accessing the perl
743 plugin will be mapped to a Perl interpreter thread (see L<threads(3perl)>).
744 Any such thread will be created and destroyed transparently and on-the-fly.
746 Hence, any plugin has to be thread-safe if it provides several entry points
747 from collectd (i.E<nbsp>e. if it registers more than one callback or if a
748 registered callback may be called more than once in parallel). Please note
749 that no data is shared between threads by default. You have to use the
750 B<threads::shared> module to do so.
754 Each function name registered with collectd has to be available before the
755 first thread has been created (i.E<nbsp>e. basically at compile time). This
756 basically means that hacks (yes, I really consider this to be a hack) like
757 C<*foo = \&bar; plugin_register (TYPE_READ, "plugin", "foo");> most likely
758 will not work. This is due to the fact that the symbol table is not shared
759 across different threads.
763 Each plugin is usually only loaded once and kept in memory for performance
764 reasons. Therefore, END blocks are only executed once when collectd shuts
765 down. You should not rely on END blocks anyway - use B<shutdown functions>
770 The perl plugin exports the internal API of collectd which is considered
771 unstable and subject to change at any time. We try hard to not break backwards
772 compatibility in the Perl API during the life cycle of one major release.
773 However, this cannot be guaranteed at all times. Watch out for warnings
774 dispatched by the perl plugin after upgrades.
784 Currently, it is not possible to flush a single Perl plugin only. You can
785 either flush all Perl plugins or none at all and you have to use C<perl> as
786 plugin name when doing so.
798 L<threads::shared(3perl)>,
803 The C<perl plugin> has been written by Sebastian Harl
804 E<lt>shE<nbsp>atE<nbsp>tokkee.orgE<gt>.
806 This manpage has been written by Florian Forster
807 E<lt>octoE<nbsp>atE<nbsp>verplant.orgE<gt> and Sebastian Harl
808 E<lt>shE<nbsp>atE<nbsp>tokkee.orgE<gt>.