5 collectd.conf - Configuration for the system statistics collection daemon B<collectd>
9 BaseDir "/var/lib/collectd"
10 PIDFile "/run/collectd.pid"
31 This config file controls how the system statistics collection daemon
32 B<collectd> behaves. The most significant option is B<LoadPlugin>, which
33 controls which plugins to load. These plugins ultimately define collectd's
34 behavior. If the B<AutoLoadPlugin> option has been enabled, the explicit
35 B<LoadPlugin> lines may be omitted for all plugins with a configuration block,
36 i.e. a C<E<lt>PluginE<nbsp>...E<gt>> block.
38 The syntax of this config file is similar to the config file of the famous
39 I<Apache> webserver. Each line contains either an option (a key and a list of
40 one or more values) or a section-start or -end. Empty lines and everything
41 after a non-quoted hash-symbol (C<#>) is ignored. I<Keys> are unquoted
42 strings, consisting only of alphanumeric characters and the underscore (C<_>)
43 character. Keys are handled case insensitive by I<collectd> itself and all
44 plugins included with it. I<Values> can either be an I<unquoted string>, a
45 I<quoted string> (enclosed in double-quotes) a I<number> or a I<boolean>
46 expression. I<Unquoted strings> consist of only alphanumeric characters and
47 underscores (C<_>) and do not need to be quoted. I<Quoted strings> are
48 enclosed in double quotes (C<">). You can use the backslash character (C<\>)
49 to include double quotes as part of the string. I<Numbers> can be specified in
50 decimal and floating point format (using a dot C<.> as decimal separator),
51 hexadecimal when using the C<0x> prefix and octal with a leading zero (C<0>).
52 I<Boolean> values are either B<true> or B<false>.
54 Lines may be wrapped by using C<\> as the last character before the newline.
55 This allows long lines to be split into multiple lines. Quoted strings may be
56 wrapped as well. However, those are treated special in that whitespace at the
57 beginning of the following lines will be ignored, which allows for nicely
58 indenting the wrapped lines.
60 The configuration is read and processed in order, i.e. from top to bottom. So
61 the plugins are loaded in the order listed in this config file. It is a good
62 idea to load any logging plugins first in order to catch messages from plugins
63 during configuration. Also, unless B<AutoLoadPlugin> is enabled, the
64 B<LoadPlugin> option I<must> occur I<before> the appropriate
65 C<E<lt>B<Plugin> ...E<gt>> block.
71 =item B<BaseDir> I<Directory>
73 Sets the base directory. This is the directory beneath all RRD-files are
74 created. Possibly more subdirectories are created. This is also the working
75 directory for the daemon.
77 =item B<LoadPlugin> I<Plugin>
79 Loads the plugin I<Plugin>. This is required to load plugins, unless the
80 B<AutoLoadPlugin> option is enabled (see below). Without any loaded plugins,
81 I<collectd> will be mostly useless.
83 Only the first B<LoadPlugin> statement or block for a given plugin name has any
84 effect. This is useful when you want to split up the configuration into smaller
85 files and want each file to be "self contained", i.e. it contains a B<Plugin>
86 block I<and> then appropriate B<LoadPlugin> statement. The downside is that if
87 you have multiple conflicting B<LoadPlugin> blocks, e.g. when they specify
88 different intervals, only one of them (the first one encountered) will take
89 effect and all others will be silently ignored.
91 B<LoadPlugin> may either be a simple configuration I<statement> or a I<block>
92 with additional options, affecting the behavior of B<LoadPlugin>. A simple
93 statement looks like this:
97 Options inside a B<LoadPlugin> block can override default settings and
98 influence the way plugins are loaded, e.g.:
104 The following options are valid inside B<LoadPlugin> blocks:
108 =item B<Globals> B<true|false>
110 If enabled, collectd will export all global symbols of the plugin (and of all
111 libraries loaded as dependencies of the plugin) and, thus, makes those symbols
112 available for resolving unresolved symbols in subsequently loaded plugins if
113 that is supported by your system.
115 This is useful (or possibly even required), e.g., when loading a plugin that
116 embeds some scripting language into the daemon (e.g. the I<Perl> and
117 I<Python plugins>). Scripting languages usually provide means to load
118 extensions written in C. Those extensions require symbols provided by the
119 interpreter, which is loaded as a dependency of the respective collectd plugin.
120 See the documentation of those plugins (e.g., L<collectd-perl(5)> or
121 L<collectd-python(5)>) for details.
123 By default, this is disabled. As a special exception, if the plugin name is
124 either C<perl> or C<python>, the default is changed to enabled in order to keep
125 the average user from ever having to deal with this low level linking stuff.
127 =item B<Interval> I<Seconds>
129 Sets a plugin-specific interval for collecting metrics. This overrides the
130 global B<Interval> setting. If a plugin provides own support for specifying an
131 interval, that setting will take precedence.
133 =item B<FlushInterval> I<Seconds>
135 Specifies the the interval, in seconds, to call the flush callback if it's
136 defined in this plugin. By default, this is disabled
138 =item B<FlushTimeout> I<Seconds>
140 Specifies the value of the timeout argument of the flush callback.
144 =item B<AutoLoadPlugin> B<false>|B<true>
146 When set to B<false> (the default), each plugin needs to be loaded explicitly,
147 using the B<LoadPlugin> statement documented above. If a
148 B<E<lt>PluginE<nbsp>...E<gt>> block is encountered and no configuration
149 handling callback for this plugin has been registered, a warning is logged and
150 the block is ignored.
152 When set to B<true>, explicit B<LoadPlugin> statements are not required. Each
153 B<E<lt>PluginE<nbsp>...E<gt>> block acts as if it was immediately preceded by a
154 B<LoadPlugin> statement. B<LoadPlugin> statements are still required for
155 plugins that don't provide any configuration, e.g. the I<Load plugin>.
157 =item B<CollectInternalStats> B<false>|B<true>
159 When set to B<true>, various statistics about the I<collectd> daemon will be
160 collected, with "collectd" as the I<plugin name>. Defaults to B<false>.
162 The "write_queue" I<plugin instance> reports the number of elements currently
163 queued and the number of elements dropped off the queue by the
164 B<WriteQueueLimitLow>/B<WriteQueueLimitHigh> mechanism.
166 The "cache" I<plugin instance> reports the number of elements in the value list
167 cache (the cache you can interact with using L<collectd-unixsock(5)>).
169 =item B<Include> I<Path> [I<pattern>]
171 If I<Path> points to a file, includes that file. If I<Path> points to a
172 directory, recursively includes all files within that directory and its
173 subdirectories. If the C<wordexp> function is available on your system,
174 shell-like wildcards are expanded before files are included. This means you can
175 use statements like the following:
177 Include "/etc/collectd.d/*.conf"
179 Starting with version 5.3, this may also be a block in which further options
180 affecting the behavior of B<Include> may be specified. The following option is
183 <Include "/etc/collectd.d">
189 =item B<Filter> I<pattern>
191 If the C<fnmatch> function is available on your system, a shell-like wildcard
192 I<pattern> may be specified to filter which files to include. This may be used
193 in combination with recursively including a directory to easily be able to
194 arbitrarily mix configuration files and other documents (e.g. README files).
195 The given example is similar to the first example above but includes all files
196 matching C<*.conf> in any subdirectory of C</etc/collectd.d>:
198 Include "/etc/collectd.d" "*.conf"
202 If more than one files are included by a single B<Include> option, the files
203 will be included in lexicographical order (as defined by the C<strcmp>
204 function). Thus, you can e.E<nbsp>g. use numbered prefixes to specify the
205 order in which the files are loaded.
207 To prevent loops and shooting yourself in the foot in interesting ways the
208 nesting is limited to a depth of 8E<nbsp>levels, which should be sufficient for
209 most uses. Since symlinks are followed it is still possible to crash the daemon
210 by looping symlinks. In our opinion significant stupidity should result in an
211 appropriate amount of pain.
213 It is no problem to have a block like C<E<lt>Plugin fooE<gt>> in more than one
214 file, but you cannot include files from within blocks.
216 =item B<PIDFile> I<File>
218 Sets where to write the PID file to. This file is overwritten when it exists
219 and deleted when the program is stopped. Some init-scripts might override this
220 setting using the B<-P> command-line option.
222 =item B<PluginDir> I<Directory>
224 Path to the plugins (shared objects) of collectd.
226 =item B<TypesDB> I<File> [I<File> ...]
228 Set one or more files that contain the data-set descriptions. See
229 L<types.db(5)> for a description of the format of this file.
231 =item B<Interval> I<Seconds>
233 Configures the interval in which to query the read plugins. Obviously smaller
234 values lead to a higher system load produced by collectd, while higher values
235 lead to more coarse statistics.
237 B<Warning:> You should set this once and then never touch it again. If you do,
238 I<you will have to delete all your RRD files> or know some serious RRDtool
239 magic! (Assuming you're using the I<RRDtool> or I<RRDCacheD> plugin.)
241 =item B<MaxReadInterval> I<Seconds>
243 Read plugin doubles interval between queries after each failed attempt
246 This options limits the maximum value of the interval. The default value is
249 =item B<Timeout> I<Iterations>
251 Consider a value list "missing" when no update has been read or received for
252 I<Iterations> iterations. By default, I<collectd> considers a value list
253 missing when no update has been received for twice the update interval. Since
254 this setting uses iterations, the maximum allowed time without update depends
255 on the I<Interval> information contained in each value list. This is used in
256 the I<Threshold> configuration to dispatch notifications about missing values,
257 see L<collectd-threshold(5)> for details.
259 =item B<ReadThreads> I<Num>
261 Number of threads to start for reading plugins. The default value is B<5>, but
262 you may want to increase this if you have more than five plugins that take a
263 long time to read. Mostly those are plugins that do network-IO. Setting this to
264 a value higher than the number of registered read callbacks is not recommended.
266 =item B<WriteThreads> I<Num>
268 Number of threads to start for dispatching value lists to write plugins. The
269 default value is B<5>, but you may want to increase this if you have more than
270 five plugins that may take relatively long to write to.
272 =item B<WriteQueueLimitHigh> I<HighNum>
274 =item B<WriteQueueLimitLow> I<LowNum>
276 Metrics are read by the I<read threads> and then put into a queue to be handled
277 by the I<write threads>. If one of the I<write plugins> is slow (e.g. network
278 timeouts, I/O saturation of the disk) this queue will grow. In order to avoid
279 running into memory issues in such a case, you can limit the size of this
282 By default, there is no limit and memory may grow indefinitely. This is most
283 likely not an issue for clients, i.e. instances that only handle the local
284 metrics. For servers it is recommended to set this to a non-zero value, though.
286 You can set the limits using B<WriteQueueLimitHigh> and B<WriteQueueLimitLow>.
287 Each of them takes a numerical argument which is the number of metrics in the
288 queue. If there are I<HighNum> metrics in the queue, any new metrics I<will> be
289 dropped. If there are less than I<LowNum> metrics in the queue, all new metrics
290 I<will> be enqueued. If the number of metrics currently in the queue is between
291 I<LowNum> and I<HighNum>, the metric is dropped with a probability that is
292 proportional to the number of metrics in the queue (i.e. it increases linearly
293 until it reaches 100%.)
295 If B<WriteQueueLimitHigh> is set to non-zero and B<WriteQueueLimitLow> is
296 unset, the latter will default to half of B<WriteQueueLimitHigh>.
298 If you do not want to randomly drop values when the queue size is between
299 I<LowNum> and I<HighNum>, set B<WriteQueueLimitHigh> and B<WriteQueueLimitLow>
302 Enabling the B<CollectInternalStats> option is of great help to figure out the
303 values to set B<WriteQueueLimitHigh> and B<WriteQueueLimitLow> to.
305 =item B<Hostname> I<Name>
307 Sets the hostname that identifies a host. If you omit this setting, the
308 hostname will be determined using the L<gethostname(2)> system call.
310 =item B<FQDNLookup> B<true|false>
312 If B<Hostname> is determined automatically this setting controls whether or not
313 the daemon should try to figure out the "fully qualified domain name", FQDN.
314 This is done using a lookup of the name returned by C<gethostname>. This option
315 is enabled by default.
317 =item B<PreCacheChain> I<ChainName>
319 =item B<PostCacheChain> I<ChainName>
321 Configure the name of the "pre-cache chain" and the "post-cache chain". Please
322 see L<FILTER CONFIGURATION> below on information on chains and how these
323 setting change the daemon's behavior.
327 =head1 PLUGIN OPTIONS
329 Some plugins may register own options. These options must be enclosed in a
330 C<Plugin>-Section. Which options exist depends on the plugin used. Some plugins
331 require external configuration, too. The C<apache plugin>, for example,
332 required C<mod_status> to be configured in the webserver you're going to
333 collect data from. These plugins are listed below as well, even if they don't
334 require any configuration within collectd's configuration file.
336 A list of all plugins and a short summary for each plugin can be found in the
337 F<README> file shipped with the sourcecode and hopefully binary packets as
340 =head2 Plugin C<aggregation>
342 The I<Aggregation plugin> makes it possible to aggregate several values into
343 one using aggregation functions such as I<sum>, I<average>, I<min> and I<max>.
344 This can be put to a wide variety of uses, e.g. average and total CPU
345 statistics for your entire fleet.
347 The grouping is powerful but, as with many powerful tools, may be a bit
348 difficult to wrap your head around. The grouping will therefore be
349 demonstrated using an example: The average and sum of the CPU usage across
350 all CPUs of each host is to be calculated.
352 To select all the affected values for our example, set C<Plugin cpu> and
353 C<Type cpu>. The other values are left unspecified, meaning "all values". The
354 I<Host>, I<Plugin>, I<PluginInstance>, I<Type> and I<TypeInstance> options
355 work as if they were specified in the C<WHERE> clause of an C<SELECT> SQL
361 Although the I<Host>, I<PluginInstance> (CPU number, i.e. 0, 1, 2, ...) and
362 I<TypeInstance> (idle, user, system, ...) fields are left unspecified in the
363 example, the intention is to have a new value for each host / type instance
364 pair. This is achieved by "grouping" the values using the C<GroupBy> option.
365 It can be specified multiple times to group by more than one field.
368 GroupBy "TypeInstance"
370 We do neither specify nor group by I<plugin instance> (the CPU number), so all
371 metrics that differ in the CPU number only will be aggregated. Each
372 aggregation needs I<at least one> such field, otherwise no aggregation would
375 The full example configuration looks like this:
377 <Plugin "aggregation">
383 GroupBy "TypeInstance"
386 CalculateAverage true
390 There are a couple of limitations you should be aware of:
396 The I<Type> cannot be left unspecified, because it is not reasonable to add
397 apples to oranges. Also, the internal lookup structure won't work if you try
402 There must be at least one unspecified, ungrouped field. Otherwise nothing
407 As you can see in the example above, each aggregation has its own
408 B<Aggregation> block. You can have multiple aggregation blocks and aggregation
409 blocks may match the same values, i.e. one value list can update multiple
410 aggregations. The following options are valid inside B<Aggregation> blocks:
414 =item B<Host> I<Host>
416 =item B<Plugin> I<Plugin>
418 =item B<PluginInstance> I<PluginInstance>
420 =item B<Type> I<Type>
422 =item B<TypeInstance> I<TypeInstance>
424 Selects the value lists to be added to this aggregation. B<Type> must be a
425 valid data set name, see L<types.db(5)> for details.
427 If the string starts with and ends with a slash (C</>), the string is
428 interpreted as a I<regular expression>. The regex flavor used are POSIX
429 extended regular expressions as described in L<regex(7)>. Example usage:
431 Host "/^db[0-9]\\.example\\.com$/"
433 =item B<GroupBy> B<Host>|B<Plugin>|B<PluginInstance>|B<TypeInstance>
435 Group valued by the specified field. The B<GroupBy> option may be repeated to
436 group by multiple fields.
438 =item B<SetHost> I<Host>
440 =item B<SetPlugin> I<Plugin>
442 =item B<SetPluginInstance> I<PluginInstance>
444 =item B<SetTypeInstance> I<TypeInstance>
446 Sets the appropriate part of the identifier to the provided string.
448 The I<PluginInstance> should include the placeholder C<%{aggregation}> which
449 will be replaced with the aggregation function, e.g. "average". Not including
450 the placeholder will result in duplication warnings and/or messed up values if
451 more than one aggregation function are enabled.
453 The following example calculates the average usage of all "even" CPUs:
455 <Plugin "aggregation">
458 PluginInstance "/[0,2,4,6,8]$/"
462 SetPluginInstance "even-%{aggregation}"
465 GroupBy "TypeInstance"
467 CalculateAverage true
471 This will create the files:
477 foo.example.com/cpu-even-average/cpu-idle
481 foo.example.com/cpu-even-average/cpu-system
485 foo.example.com/cpu-even-average/cpu-user
493 =item B<CalculateNum> B<true>|B<false>
495 =item B<CalculateSum> B<true>|B<false>
497 =item B<CalculateAverage> B<true>|B<false>
499 =item B<CalculateMinimum> B<true>|B<false>
501 =item B<CalculateMaximum> B<true>|B<false>
503 =item B<CalculateStddev> B<true>|B<false>
505 Boolean options for enabling calculation of the number of value lists, their
506 sum, average, minimum, maximum andE<nbsp>/ or standard deviation. All options
507 are disabled by default.
511 =head2 Plugin C<amqp>
513 The I<AMQMP plugin> can be used to communicate with other instances of
514 I<collectd> or third party applications using an AMQP message broker. Values
515 are sent to or received from the broker, which handles routing, queueing and
516 possibly filtering or messages.
519 # Send values to an AMQP broker
520 <Publish "some_name">
526 Exchange "amq.fanout"
527 # ExchangeType "fanout"
528 # RoutingKey "collectd"
530 # ConnectionRetryDelay 0
533 # GraphitePrefix "collectd."
534 # GraphiteEscapeChar "_"
535 # GraphiteSeparateInstances false
536 # GraphiteAlwaysAppendDS false
539 # Receive values from an AMQP broker
540 <Subscribe "some_name">
546 Exchange "amq.fanout"
547 # ExchangeType "fanout"
550 # QueueAutoDelete true
551 # RoutingKey "collectd.#"
552 # ConnectionRetryDelay 0
556 The plugin's configuration consists of a number of I<Publish> and I<Subscribe>
557 blocks, which configure sending and receiving of values respectively. The two
558 blocks are very similar, so unless otherwise noted, an option can be used in
559 either block. The name given in the blocks starting tag is only used for
560 reporting messages, but may be used to support I<flushing> of certain
561 I<Publish> blocks in the future.
565 =item B<Host> I<Host>
567 Hostname or IP-address of the AMQP broker. Defaults to the default behavior of
568 the underlying communications library, I<rabbitmq-c>, which is "localhost".
570 =item B<Port> I<Port>
572 Service name or port number on which the AMQP broker accepts connections. This
573 argument must be a string, even if the numeric form is used. Defaults to
576 =item B<VHost> I<VHost>
578 Name of the I<virtual host> on the AMQP broker to use. Defaults to "/".
580 =item B<User> I<User>
582 =item B<Password> I<Password>
584 Credentials used to authenticate to the AMQP broker. By default "guest"/"guest"
587 =item B<Exchange> I<Exchange>
589 In I<Publish> blocks, this option specifies the I<exchange> to send values to.
590 By default, "amq.fanout" will be used.
592 In I<Subscribe> blocks this option is optional. If given, a I<binding> between
593 the given exchange and the I<queue> is created, using the I<routing key> if
594 configured. See the B<Queue> and B<RoutingKey> options below.
596 =item B<ExchangeType> I<Type>
598 If given, the plugin will try to create the configured I<exchange> with this
599 I<type> after connecting. When in a I<Subscribe> block, the I<queue> will then
600 be bound to this exchange.
602 =item B<Queue> I<Queue> (Subscribe only)
604 Configures the I<queue> name to subscribe to. If no queue name was configured
605 explicitly, a unique queue name will be created by the broker.
607 =item B<QueueDurable> B<true>|B<false> (Subscribe only)
609 Defines if the I<queue> subscribed to is durable (saved to persistent storage)
610 or transient (will disappear if the AMQP broker is restarted). Defaults to
613 This option should be used in conjunction with the I<Persistent> option on the
616 =item B<QueueAutoDelete> B<true>|B<false> (Subscribe only)
618 Defines if the I<queue> subscribed to will be deleted once the last consumer
619 unsubscribes. Defaults to "true".
621 =item B<RoutingKey> I<Key>
623 In I<Publish> blocks, this configures the routing key to set on all outgoing
624 messages. If not given, the routing key will be computed from the I<identifier>
625 of the value. The host, plugin, type and the two instances are concatenated
626 together using dots as the separator and all containing dots replaced with
627 slashes. For example "collectd.host/example/com.cpu.0.cpu.user". This makes it
628 possible to receive only specific values using a "topic" exchange.
630 In I<Subscribe> blocks, configures the I<routing key> used when creating a
631 I<binding> between an I<exchange> and the I<queue>. The usual wildcards can be
632 used to filter messages when using a "topic" exchange. If you're only
633 interested in CPU statistics, you could use the routing key "collectd.*.cpu.#"
636 =item B<Persistent> B<true>|B<false> (Publish only)
638 Selects the I<delivery method> to use. If set to B<true>, the I<persistent>
639 mode will be used, i.e. delivery is guaranteed. If set to B<false> (the
640 default), the I<transient> delivery mode will be used, i.e. messages may be
641 lost due to high load, overflowing queues or similar issues.
643 =item B<ConnectionRetryDelay> I<Delay>
645 When the connection to the AMQP broker is lost, defines the time in seconds to
646 wait before attempting to reconnect. Defaults to 0, which implies collectd will
647 attempt to reconnect at each read interval (in Subscribe mode) or each time
648 values are ready for submission (in Publish mode).
650 =item B<Format> B<Command>|B<JSON>|B<Graphite> (Publish only)
652 Selects the format in which messages are sent to the broker. If set to
653 B<Command> (the default), values are sent as C<PUTVAL> commands which are
654 identical to the syntax used by the I<Exec> and I<UnixSock plugins>. In this
655 case, the C<Content-Type> header field will be set to C<text/collectd>.
657 If set to B<JSON>, the values are encoded in the I<JavaScript Object Notation>,
658 an easy and straight forward exchange format. The C<Content-Type> header field
659 will be set to C<application/json>.
661 If set to B<Graphite>, values are encoded in the I<Graphite> format, which is
662 "<metric> <value> <timestamp>\n". The C<Content-Type> header field will be set to
665 A subscribing client I<should> use the C<Content-Type> header field to
666 determine how to decode the values. Currently, the I<AMQP plugin> itself can
667 only decode the B<Command> format.
669 =item B<StoreRates> B<true>|B<false> (Publish only)
671 Determines whether or not C<COUNTER>, C<DERIVE> and C<ABSOLUTE> data sources
672 are converted to a I<rate> (i.e. a C<GAUGE> value). If set to B<false> (the
673 default), no conversion is performed. Otherwise the conversion is performed
674 using the internal value cache.
676 Please note that currently this option is only used if the B<Format> option has
679 =item B<GraphitePrefix> (Publish and B<Format>=I<Graphite> only)
681 A prefix can be added in the metric name when outputting in the I<Graphite> format.
682 It's added before the I<Host> name.
683 Metric name will be "<prefix><host><postfix><plugin><type><name>"
685 =item B<GraphitePostfix> (Publish and B<Format>=I<Graphite> only)
687 A postfix can be added in the metric name when outputting in the I<Graphite> format.
688 It's added after the I<Host> name.
689 Metric name will be "<prefix><host><postfix><plugin><type><name>"
691 =item B<GraphiteEscapeChar> (Publish and B<Format>=I<Graphite> only)
693 Specify a character to replace dots (.) in the host part of the metric name.
694 In I<Graphite> metric name, dots are used as separators between different
695 metric parts (host, plugin, type).
696 Default is "_" (I<Underscore>).
698 =item B<GraphiteSeparateInstances> B<true>|B<false>
700 If set to B<true>, the plugin instance and type instance will be in their own
701 path component, for example C<host.cpu.0.cpu.idle>. If set to B<false> (the
702 default), the plugin and plugin instance (and likewise the type and type
703 instance) are put into one component, for example C<host.cpu-0.cpu-idle>.
705 =item B<GraphiteAlwaysAppendDS> B<true>|B<false>
707 If set to B<true>, append the name of the I<Data Source> (DS) to the "metric"
708 identifier. If set to B<false> (the default), this is only done when there is
713 =head2 Plugin C<apache>
715 To configure the C<apache>-plugin you first need to configure the Apache
716 webserver correctly. The Apache-plugin C<mod_status> needs to be loaded and
717 working and the C<ExtendedStatus> directive needs to be B<enabled>. You can use
718 the following snipped to base your Apache config upon:
721 <IfModule mod_status.c>
722 <Location /mod_status>
723 SetHandler server-status
727 Since its C<mod_status> module is very similar to Apache's, B<lighttpd> is
728 also supported. It introduces a new field, called C<BusyServers>, to count the
729 number of currently connected clients. This field is also supported.
731 The configuration of the I<Apache> plugin consists of one or more
732 C<E<lt>InstanceE<nbsp>/E<gt>> blocks. Each block requires one string argument
733 as the instance name. For example:
737 URL "http://www1.example.com/mod_status?auto"
740 URL "http://www2.example.com/mod_status?auto"
744 The instance name will be used as the I<plugin instance>. To emulate the old
745 (versionE<nbsp>4) behavior, you can use an empty string (""). In order for the
746 plugin to work correctly, each instance name must be unique. This is not
747 enforced by the plugin and it is your responsibility to ensure it.
749 The following options are accepted within each I<Instance> block:
753 =item B<URL> I<http://host/mod_status?auto>
755 Sets the URL of the C<mod_status> output. This needs to be the output generated
756 by C<ExtendedStatus on> and it needs to be the machine readable output
757 generated by appending the C<?auto> argument. This option is I<mandatory>.
759 =item B<User> I<Username>
761 Optional user name needed for authentication.
763 =item B<Password> I<Password>
765 Optional password needed for authentication.
767 =item B<VerifyPeer> B<true|false>
769 Enable or disable peer SSL certificate verification. See
770 L<http://curl.haxx.se/docs/sslcerts.html> for details. Enabled by default.
772 =item B<VerifyHost> B<true|false>
774 Enable or disable peer host name verification. If enabled, the plugin checks
775 if the C<Common Name> or a C<Subject Alternate Name> field of the SSL
776 certificate matches the host name provided by the B<URL> option. If this
777 identity check fails, the connection is aborted. Obviously, only works when
778 connecting to a SSL enabled server. Enabled by default.
780 =item B<CACert> I<File>
782 File that holds one or more SSL certificates. If you want to use HTTPS you will
783 possibly need this option. What CA certificates come bundled with C<libcurl>
784 and are checked by default depends on the distribution you use.
786 =item B<SSLCiphers> I<list of ciphers>
788 Specifies which ciphers to use in the connection. The list of ciphers
789 must specify valid ciphers. See
790 L<http://www.openssl.org/docs/apps/ciphers.html> for details.
792 =item B<Timeout> I<Milliseconds>
794 The B<Timeout> option sets the overall timeout for HTTP requests to B<URL>, in
795 milliseconds. By default, the configured B<Interval> is used to set the
800 =head2 Plugin C<apcups>
804 =item B<Host> I<Hostname>
806 Hostname of the host running B<apcupsd>. Defaults to B<localhost>. Please note
807 that IPv6 support has been disabled unless someone can confirm or decline that
808 B<apcupsd> can handle it.
810 =item B<Port> I<Port>
812 TCP-Port to connect to. Defaults to B<3551>.
814 =item B<ReportSeconds> B<true>|B<false>
816 If set to B<true>, the time reported in the C<timeleft> metric will be
817 converted to seconds. This is the recommended setting. If set to B<false>, the
818 default for backwards compatibility, the time will be reported in minutes.
820 =item B<PersistentConnection> B<true>|B<false>
822 By default, the plugin will try to keep the connection to UPS open between
823 reads. Since this appears to be somewhat brittle (I<apcupsd> appears to close
824 the connection due to inactivity quite quickly), the plugin will try to detect
825 this problem and switch to an open-read-close mode in such cases.
827 You can instruct the plugin to close the connection after each read by setting
828 this option to B<false>.
832 =head2 Plugin C<aquaero>
834 This plugin collects the value of the available sensors in an
835 I<AquaeroE<nbsp>5> board. AquaeroE<nbsp>5 is a water-cooling controller board,
836 manufactured by Aqua Computer GmbH L<http://www.aquacomputer.de/>, with a USB2
837 connection for monitoring and configuration. The board can handle multiple
838 temperature sensors, fans, water pumps and water level sensors and adjust the
839 output settings such as fan voltage or power used by the water pump based on
840 the available inputs using a configurable controller included in the board.
841 This plugin collects all the available inputs as well as some of the output
842 values chosen by this controller. The plugin is based on the I<libaquaero5>
843 library provided by I<aquatools-ng>.
847 =item B<Device> I<DevicePath>
849 Device path of the AquaeroE<nbsp>5's USB HID (human interface device), usually
850 in the form C</dev/usb/hiddevX>. If this option is no set the plugin will try
851 to auto-detect the Aquaero 5 USB device based on vendor-ID and product-ID.
855 =head2 Plugin C<ascent>
857 This plugin collects information about an Ascent server, a free server for the
858 "World of Warcraft" game. This plugin gathers the information by fetching the
859 XML status page using C<libcurl> and parses it using C<libxml2>.
861 The configuration options are the same as for the C<apache> plugin above:
865 =item B<URL> I<http://localhost/ascent/status/>
867 Sets the URL of the XML status output.
869 =item B<User> I<Username>
871 Optional user name needed for authentication.
873 =item B<Password> I<Password>
875 Optional password needed for authentication.
877 =item B<VerifyPeer> B<true|false>
879 Enable or disable peer SSL certificate verification. See
880 L<http://curl.haxx.se/docs/sslcerts.html> for details. Enabled by default.
882 =item B<VerifyHost> B<true|false>
884 Enable or disable peer host name verification. If enabled, the plugin checks
885 if the C<Common Name> or a C<Subject Alternate Name> field of the SSL
886 certificate matches the host name provided by the B<URL> option. If this
887 identity check fails, the connection is aborted. Obviously, only works when
888 connecting to a SSL enabled server. Enabled by default.
890 =item B<CACert> I<File>
892 File that holds one or more SSL certificates. If you want to use HTTPS you will
893 possibly need this option. What CA certificates come bundled with C<libcurl>
894 and are checked by default depends on the distribution you use.
896 =item B<Timeout> I<Milliseconds>
898 The B<Timeout> option sets the overall timeout for HTTP requests to B<URL>, in
899 milliseconds. By default, the configured B<Interval> is used to set the
904 =head2 Plugin C<barometer>
906 This plugin reads absolute air pressure using digital barometer sensor on a I2C
907 bus. Supported sensors are:
911 =item I<MPL115A2> from Freescale,
912 see L<http://www.freescale.com/webapp/sps/site/prod_summary.jsp?code=MPL115A>.
915 =item I<MPL3115> from Freescale
916 see L<http://www.freescale.com/webapp/sps/site/prod_summary.jsp?code=MPL3115A2>.
919 =item I<BMP085> from Bosch Sensortec
923 The sensor type - one of the above - is detected automatically by the plugin
924 and indicated in the plugin_instance (you will see subdirectory
925 "barometer-mpl115" or "barometer-mpl3115", or "barometer-bmp085"). The order of
926 detection is BMP085 -> MPL3115 -> MPL115A2, the first one found will be used
927 (only one sensor can be used by the plugin).
929 The plugin provides absolute barometric pressure, air pressure reduced to sea
930 level (several possible approximations) and as an auxiliary value also internal
931 sensor temperature. It uses (expects/provides) typical metric units - pressure
932 in [hPa], temperature in [C], altitude in [m].
934 It was developed and tested under Linux only. The only platform dependency is
935 the standard Linux i2c-dev interface (the particular bus driver has to
936 support the SM Bus command subset).
938 The reduction or normalization to mean sea level pressure requires (depending
939 on selected method/approximation) also altitude and reference to temperature
940 sensor(s). When multiple temperature sensors are configured the minumum of
941 their values is always used (expecting that the warmer ones are affected by
942 e.g. direct sun light at that moment).
950 TemperatureOffset 0.0
953 TemperatureSensor "myserver/onewire-F10FCA000800/temperature"
958 =item B<Device> I<device>
960 The only mandatory configuration parameter.
962 Device name of the I2C bus to which the sensor is connected. Note that
963 typically you need to have loaded the i2c-dev module.
964 Using i2c-tools you can check/list i2c buses available on your system by:
968 Then you can scan for devices on given bus. E.g. to scan the whole bus 0 use:
972 This way you should be able to verify that the pressure sensor (either type) is
973 connected and detected on address 0x60.
975 =item B<Oversampling> I<value>
977 Optional parameter controlling the oversampling/accuracy. Default value
978 is 1 providing fastest and least accurate reading.
980 For I<MPL115> this is the size of the averaging window. To filter out sensor
981 noise a simple averaging using floating window of this configurable size is
982 used. The plugin will use average of the last C<value> measurements (value of 1
983 means no averaging). Minimal size is 1, maximal 1024.
985 For I<MPL3115> this is the oversampling value. The actual oversampling is
986 performed by the sensor and the higher value the higher accuracy and longer
987 conversion time (although nothing to worry about in the collectd context).
988 Supported values are: 1, 2, 4, 8, 16, 32, 64 and 128. Any other value is
989 adjusted by the plugin to the closest supported one.
991 For I<BMP085> this is the oversampling value. The actual oversampling is
992 performed by the sensor and the higher value the higher accuracy and longer
993 conversion time (although nothing to worry about in the collectd context).
994 Supported values are: 1, 2, 4, 8. Any other value is adjusted by the plugin to
995 the closest supported one.
997 =item B<PressureOffset> I<offset>
999 Optional parameter for MPL3115 only.
1001 You can further calibrate the sensor by supplying pressure and/or temperature
1002 offsets. This is added to the measured/caclulated value (i.e. if the measured
1003 value is too high then use negative offset).
1004 In hPa, default is 0.0.
1006 =item B<TemperatureOffset> I<offset>
1008 Optional parameter for MPL3115 only.
1010 You can further calibrate the sensor by supplying pressure and/or temperature
1011 offsets. This is added to the measured/caclulated value (i.e. if the measured
1012 value is too high then use negative offset).
1013 In C, default is 0.0.
1015 =item B<Normalization> I<method>
1017 Optional parameter, default value is 0.
1019 Normalization method - what approximation/model is used to compute the mean sea
1020 level pressure from the air absolute pressure.
1022 Supported values of the C<method> (integer between from 0 to 2) are:
1026 =item B<0> - no conversion, absolute pressure is simply copied over. For this method you
1027 do not need to configure C<Altitude> or C<TemperatureSensor>.
1029 =item B<1> - international formula for conversion ,
1031 L<http://en.wikipedia.org/wiki/Atmospheric_pressure#Altitude_atmospheric_pressure_variation>.
1032 For this method you have to configure C<Altitude> but do not need
1033 C<TemperatureSensor> (uses fixed global temperature average instead).
1035 =item B<2> - formula as recommended by the Deutsche Wetterdienst (German
1036 Meteorological Service).
1037 See L<http://de.wikipedia.org/wiki/Barometrische_H%C3%B6henformel#Theorie>
1038 For this method you have to configure both C<Altitude> and
1039 C<TemperatureSensor>.
1044 =item B<Altitude> I<altitude>
1046 The altitude (in meters) of the location where you meassure the pressure.
1048 =item B<TemperatureSensor> I<reference>
1050 Temperature sensor(s) which should be used as a reference when normalizing the
1051 pressure using C<Normalization> method 2.
1052 When specified more sensors a minumum is found and used each time. The
1053 temperature reading directly from this pressure sensor/plugin is typically not
1054 suitable as the pressure sensor will be probably inside while we want outside
1055 temperature. The collectd reference name is something like
1056 <hostname>/<plugin_name>-<plugin_instance>/<type>-<type_instance>
1057 (<type_instance> is usually omitted when there is just single value type). Or
1058 you can figure it out from the path of the output data files.
1062 =head2 Plugin C<battery>
1064 The I<battery plugin> reports the remaining capacity, power and voltage of
1069 =item B<ValuesPercentage> B<false>|B<true>
1071 When enabled, remaining capacity is reported as a percentage, e.g. "42%
1072 capacity remaining". Otherwise the capacity is stored as reported by the
1073 battery, most likely in "Wh". This option does not work with all input methods,
1074 in particular when only C</proc/pmu> is available on an old Linux system.
1075 Defaults to B<false>.
1077 =item B<ReportDegraded> B<false>|B<true>
1079 Typical laptop batteries degrade over time, meaning the capacity decreases with
1080 recharge cycles. The maximum charge of the previous charge cycle is tracked as
1081 "last full capacity" and used to determine that a battery is "fully charged".
1083 When this option is set to B<false>, the default, the I<battery plugin> will
1084 only report the remaining capacity. If the B<ValuesPercentage> option is
1085 enabled, the relative remaining capacity is calculated as the ratio of the
1086 "remaining capacity" and the "last full capacity". This is what most tools,
1087 such as the status bar of desktop environments, also do.
1089 When set to B<true>, the battery plugin will report three values: B<charged>
1090 (remaining capacity), B<discharged> (difference between "last full capacity"
1091 and "remaining capacity") and B<degraded> (difference between "design capacity"
1092 and "last full capacity").
1096 =head2 Plugin C<bind>
1098 Starting with BIND 9.5.0, the most widely used DNS server software provides
1099 extensive statistics about queries, responses and lots of other information.
1100 The bind plugin retrieves this information that's encoded in XML and provided
1101 via HTTP and submits the values to collectd.
1103 To use this plugin, you first need to tell BIND to make this information
1104 available. This is done with the C<statistics-channels> configuration option:
1106 statistics-channels {
1107 inet localhost port 8053;
1110 The configuration follows the grouping that can be seen when looking at the
1111 data with an XSLT compatible viewer, such as a modern web browser. It's
1112 probably a good idea to make yourself familiar with the provided values, so you
1113 can understand what the collected statistics actually mean.
1118 URL "http://localhost:8053/"
1133 Zone "127.in-addr.arpa/IN"
1137 The bind plugin accepts the following configuration options:
1143 URL from which to retrieve the XML data. If not specified,
1144 C<http://localhost:8053/> will be used.
1146 =item B<ParseTime> B<true>|B<false>
1148 When set to B<true>, the time provided by BIND will be parsed and used to
1149 dispatch the values. When set to B<false>, the local time source is queried.
1151 This setting is set to B<true> by default for backwards compatibility; setting
1152 this to B<false> is I<recommended> to avoid problems with timezones and
1155 =item B<OpCodes> B<true>|B<false>
1157 When enabled, statistics about the I<"OpCodes">, for example the number of
1158 C<QUERY> packets, are collected.
1162 =item B<QTypes> B<true>|B<false>
1164 When enabled, the number of I<incoming> queries by query types (for example
1165 C<A>, C<MX>, C<AAAA>) is collected.
1169 =item B<ServerStats> B<true>|B<false>
1171 Collect global server statistics, such as requests received over IPv4 and IPv6,
1172 successful queries, and failed updates.
1176 =item B<ZoneMaintStats> B<true>|B<false>
1178 Collect zone maintenance statistics, mostly information about notifications
1179 (zone updates) and zone transfers.
1183 =item B<ResolverStats> B<true>|B<false>
1185 Collect resolver statistics, i.E<nbsp>e. statistics about outgoing requests
1186 (e.E<nbsp>g. queries over IPv4, lame servers). Since the global resolver
1187 counters apparently were removed in BIND 9.5.1 and 9.6.0, this is disabled by
1188 default. Use the B<ResolverStats> option within a B<View "_default"> block
1189 instead for the same functionality.
1193 =item B<MemoryStats>
1195 Collect global memory statistics.
1199 =item B<Timeout> I<Milliseconds>
1201 The B<Timeout> option sets the overall timeout for HTTP requests to B<URL>, in
1202 milliseconds. By default, the configured B<Interval> is used to set the
1205 =item B<View> I<Name>
1207 Collect statistics about a specific I<"view">. BIND can behave different,
1208 mostly depending on the source IP-address of the request. These different
1209 configurations are called "views". If you don't use this feature, you most
1210 likely are only interested in the C<_default> view.
1212 Within a E<lt>B<View>E<nbsp>I<name>E<gt> block, you can specify which
1213 information you want to collect about a view. If no B<View> block is
1214 configured, no detailed view statistics will be collected.
1218 =item B<QTypes> B<true>|B<false>
1220 If enabled, the number of I<outgoing> queries by query type (e.E<nbsp>g. C<A>,
1221 C<MX>) is collected.
1225 =item B<ResolverStats> B<true>|B<false>
1227 Collect resolver statistics, i.E<nbsp>e. statistics about outgoing requests
1228 (e.E<nbsp>g. queries over IPv4, lame servers).
1232 =item B<CacheRRSets> B<true>|B<false>
1234 If enabled, the number of entries (I<"RR sets">) in the view's cache by query
1235 type is collected. Negative entries (queries which resulted in an error, for
1236 example names that do not exist) are reported with a leading exclamation mark,
1241 =item B<Zone> I<Name>
1243 When given, collect detailed information about the given zone in the view. The
1244 information collected if very similar to the global B<ServerStats> information
1247 You can repeat this option to collect detailed information about multiple
1250 By default no detailed zone information is collected.
1256 =head2 Plugin C<ceph>
1258 The ceph plugin collects values from JSON data to be parsed by B<libyajl>
1259 (L<https://lloyd.github.io/yajl/>) retrieved from ceph daemon admin sockets.
1261 A separate B<Daemon> block must be configured for each ceph daemon to be
1262 monitored. The following example will read daemon statistics from four
1263 separate ceph daemons running on the same device (two OSDs, one MON, one MDS) :
1266 LongRunAvgLatency false
1267 ConvertSpecialMetricTypes true
1269 SocketPath "/var/run/ceph/ceph-osd.0.asok"
1272 SocketPath "/var/run/ceph/ceph-osd.1.asok"
1275 SocketPath "/var/run/ceph/ceph-mon.ceph1.asok"
1278 SocketPath "/var/run/ceph/ceph-mds.ceph1.asok"
1282 The ceph plugin accepts the following configuration options:
1286 =item B<LongRunAvgLatency> B<true>|B<false>
1288 If enabled, latency values(sum,count pairs) are calculated as the long run
1289 average - average since the ceph daemon was started = (sum / count).
1290 When disabled, latency values are calculated as the average since the last
1291 collection = (sum_now - sum_last) / (count_now - count_last).
1295 =item B<ConvertSpecialMetricTypes> B<true>|B<false>
1297 If enabled, special metrics (metrics that differ in type from similar counters)
1298 are converted to the type of those similar counters. This currently only
1299 applies to filestore.journal_wr_bytes which is a counter for OSD daemons. The
1300 ceph schema reports this metric type as a sum,count pair while similar counters
1301 are treated as derive types. When converted, the sum is used as the counter
1302 value and is treated as a derive type.
1303 When disabled, all metrics are treated as the types received from the ceph schema.
1309 Each B<Daemon> block must have a string argument for the plugin instance name.
1310 A B<SocketPath> is also required for each B<Daemon> block:
1314 =item B<Daemon> I<DaemonName>
1316 Name to be used as the instance name for this daemon.
1318 =item B<SocketPath> I<SocketPath>
1320 Specifies the path to the UNIX admin socket of the ceph daemon.
1324 =head2 Plugin C<cgroups>
1326 This plugin collects the CPU user/system time for each I<cgroup> by reading the
1327 F<cpuacct.stat> files in the first cpuacct-mountpoint (typically
1328 F</sys/fs/cgroup/cpu.cpuacct> on machines using systemd).
1332 =item B<CGroup> I<Directory>
1334 Select I<cgroup> based on the name. Whether only matching I<cgroups> are
1335 collected or if they are ignored is controlled by the B<IgnoreSelected> option;
1338 =item B<IgnoreSelected> B<true>|B<false>
1340 Invert the selection: If set to true, all cgroups I<except> the ones that
1341 match any one of the criteria are collected. By default only selected
1342 cgroups are collected if a selection is made. If no selection is configured
1343 at all, B<all> cgroups are selected.
1347 =head2 Plugin C<conntrack>
1349 This plugin collects IP conntrack statistics.
1355 Assume the B<conntrack_count> and B<conntrack_max> files to be found in
1356 F</proc/sys/net/ipv4/netfilter> instead of F</proc/sys/net/netfilter/>.
1360 =head2 Plugin C<cpu>
1362 The I<CPU plugin> collects CPU usage metrics. By default, CPU usage is reported
1363 as Jiffies, using the C<cpu> type. Two aggregations are available:
1369 Sum, per-state, over all CPUs installed in the system; and
1373 Sum, per-CPU, over all non-idle states of a CPU, creating an "active" state.
1377 The two aggregations can be combined, leading to I<collectd> only emitting a
1378 single "active" metric for the entire system. As soon as one of these
1379 aggregations (or both) is enabled, the I<cpu plugin> will report a percentage,
1380 rather than Jiffies. In addition, you can request individual, per-state,
1381 per-CPU metrics to be reported as percentage.
1383 The following configuration options are available:
1387 =item B<ReportByState> B<true>|B<false>
1389 When set to B<true>, the default, reports per-state metrics, e.g. "system",
1391 When set to B<false>, aggregates (sums) all I<non-idle> states into one
1394 =item B<ReportByCpu> B<true>|B<false>
1396 When set to B<true>, the default, reports per-CPU (per-core) metrics.
1397 When set to B<false>, instead of reporting metrics for individual CPUs, only a
1398 global sum of CPU states is emitted.
1400 =item B<ValuesPercentage> B<false>|B<true>
1402 This option is only considered when both, B<ReportByCpu> and B<ReportByState>
1403 are set to B<true>. In this case, by default, metrics will be reported as
1404 Jiffies. By setting this option to B<true>, you can request percentage values
1405 in the un-aggregated (per-CPU, per-state) mode as well.
1409 =head2 Plugin C<cpufreq>
1411 This plugin doesn't have any options. It reads
1412 F</sys/devices/system/cpu/cpu0/cpufreq/scaling_cur_freq> (for the first CPU
1413 installed) to get the current CPU frequency. If this file does not exist make
1414 sure B<cpufreqd> (L<http://cpufreqd.sourceforge.net/>) or a similar tool is
1415 installed and an "cpu governor" (that's a kernel module) is loaded.
1417 =head2 Plugin C<csv>
1421 =item B<DataDir> I<Directory>
1423 Set the directory to store CSV-files under. Per default CSV-files are generated
1424 beneath the daemon's working directory, i.E<nbsp>e. the B<BaseDir>.
1425 The special strings B<stdout> and B<stderr> can be used to write to the standard
1426 output and standard error channels, respectively. This, of course, only makes
1427 much sense when collectd is running in foreground- or non-daemon-mode.
1429 =item B<StoreRates> B<true|false>
1431 If set to B<true>, convert counter values to rates. If set to B<false> (the
1432 default) counter values are stored as is, i.E<nbsp>e. as an increasing integer
1437 =head2 Plugin C<curl>
1439 The curl plugin uses the B<libcurl> (L<http://curl.haxx.se/>) to read web pages
1440 and the match infrastructure (the same code used by the tail plugin) to use
1441 regular expressions with the received data.
1443 The following example will read the current value of AMD stock from Google's
1444 finance page and dispatch the value to collectd.
1447 <Page "stock_quotes">
1448 URL "http://finance.google.com/finance?q=NYSE%3AAMD"
1454 CACert "/path/to/ca.crt"
1455 Header "X-Custom-Header: foobar"
1458 MeasureResponseTime false
1459 MeasureResponseCode false
1462 Regex "<span +class=\"pr\"[^>]*> *([0-9]*\\.[0-9]+) *</span>"
1463 DSType "GaugeAverage"
1464 # Note: `stock_value' is not a standard type.
1471 In the B<Plugin> block, there may be one or more B<Page> blocks, each defining
1472 a web page and one or more "matches" to be performed on the returned data. The
1473 string argument to the B<Page> block is used as plugin instance.
1475 The following options are valid within B<Page> blocks:
1481 URL of the web site to retrieve. Since a regular expression will be used to
1482 extract information from this data, non-binary data is a big plus here ;)
1484 =item B<User> I<Name>
1486 Username to use if authorization is required to read the page.
1488 =item B<Password> I<Password>
1490 Password to use if authorization is required to read the page.
1492 =item B<Digest> B<true>|B<false>
1494 Enable HTTP digest authentication.
1496 =item B<VerifyPeer> B<true>|B<false>
1498 Enable or disable peer SSL certificate verification. See
1499 L<http://curl.haxx.se/docs/sslcerts.html> for details. Enabled by default.
1501 =item B<VerifyHost> B<true>|B<false>
1503 Enable or disable peer host name verification. If enabled, the plugin checks if
1504 the C<Common Name> or a C<Subject Alternate Name> field of the SSL certificate
1505 matches the host name provided by the B<URL> option. If this identity check
1506 fails, the connection is aborted. Obviously, only works when connecting to a
1507 SSL enabled server. Enabled by default.
1509 =item B<CACert> I<file>
1511 File that holds one or more SSL certificates. If you want to use HTTPS you will
1512 possibly need this option. What CA certificates come bundled with C<libcurl>
1513 and are checked by default depends on the distribution you use.
1515 =item B<Header> I<Header>
1517 A HTTP header to add to the request. Multiple headers are added if this option
1518 is specified more than once.
1520 =item B<Post> I<Body>
1522 Specifies that the HTTP operation should be a POST instead of a GET. The
1523 complete data to be posted is given as the argument. This option will usually
1524 need to be accompanied by a B<Header> option to set an appropriate
1525 C<Content-Type> for the post body (e.g. to
1526 C<application/x-www-form-urlencoded>).
1528 =item B<MeasureResponseTime> B<true>|B<false>
1530 Measure response time for the request. If this setting is enabled, B<Match>
1531 blocks (see below) are optional. Disabled by default.
1533 Beware that requests will get aborted if they take too long to complete. Adjust
1534 B<Timeout> accordingly if you expect B<MeasureResponseTime> to report such slow
1537 =item B<MeasureResponseCode> B<true>|B<false>
1539 Measure response code for the request. If this setting is enabled, B<Match>
1540 blocks (see below) are optional. Disabled by default.
1542 =item B<E<lt>MatchE<gt>>
1544 One or more B<Match> blocks that define how to match information in the data
1545 returned by C<libcurl>. The C<curl> plugin uses the same infrastructure that's
1546 used by the C<tail> plugin, so please see the documentation of the C<tail>
1547 plugin below on how matches are defined. If the B<MeasureResponseTime> or
1548 B<MeasureResponseCode> options are set to B<true>, B<Match> blocks are
1551 =item B<Timeout> I<Milliseconds>
1553 The B<Timeout> option sets the overall timeout for HTTP requests to B<URL>, in
1554 milliseconds. By default, the configured B<Interval> is used to set the
1555 timeout. Prior to version 5.5.0, there was no timeout and requests could hang
1556 indefinitely. This legacy behaviour can be achieved by setting the value of
1559 If B<Timeout> is 0 or bigger than the B<Interval>, keep in mind that each slow
1560 network connection will stall one read thread. Adjust the B<ReadThreads> global
1561 setting accordingly to prevent this from blocking other plugins.
1565 =head2 Plugin C<curl_json>
1567 The B<curl_json plugin> collects values from JSON data to be parsed by
1568 B<libyajl> (L<https://lloyd.github.io/yajl/>) retrieved via
1569 either B<libcurl> (L<http://curl.haxx.se/>) or read directly from a
1570 unix socket. The former can be used, for example, to collect values
1571 from CouchDB documents (which are stored JSON notation), and the
1572 latter to collect values from a uWSGI stats socket.
1574 The following example will collect several values from the built-in
1575 C<_stats> runtime statistics module of I<CouchDB>
1576 (L<http://wiki.apache.org/couchdb/Runtime_Statistics>).
1579 <URL "http://localhost:5984/_stats">
1581 <Key "httpd/requests/count">
1582 Type "http_requests"
1585 <Key "httpd_request_methods/*/count">
1586 Type "http_request_methods"
1589 <Key "httpd_status_codes/*/count">
1590 Type "http_response_codes"
1595 This example will collect data directly from a I<uWSGI> "Stats Server" socket.
1598 <Sock "/var/run/uwsgi.stats.sock">
1600 <Key "workers/*/requests">
1601 Type "http_requests"
1604 <Key "workers/*/apps/*/requests">
1605 Type "http_requests"
1610 In the B<Plugin> block, there may be one or more B<URL> blocks, each
1611 defining a URL to be fetched via HTTP (using libcurl) or B<Sock>
1612 blocks defining a unix socket to read JSON from directly. Each of
1613 these blocks may have one or more B<Key> blocks.
1615 The B<Key> string argument must be in a path format. Each component is
1616 used to match the key from a JSON map or the index of an JSON
1617 array. If a path component of a B<Key> is a I<*>E<nbsp>wildcard, the
1618 values for all map keys or array indices will be collectd.
1620 The following options are valid within B<URL> blocks:
1624 =item B<Instance> I<Instance>
1626 Sets the plugin instance to I<Instance>.
1628 =item B<Interval> I<Interval>
1630 Sets the interval (in seconds) in which the values will be collected from this
1631 URL. By default the global B<Interval> setting will be used.
1633 =item B<User> I<Name>
1635 =item B<Password> I<Password>
1637 =item B<Digest> B<true>|B<false>
1639 =item B<VerifyPeer> B<true>|B<false>
1641 =item B<VerifyHost> B<true>|B<false>
1643 =item B<CACert> I<file>
1645 =item B<Header> I<Header>
1647 =item B<Post> I<Body>
1649 =item B<Timeout> I<Milliseconds>
1651 These options behave exactly equivalent to the appropriate options of the
1652 I<cURL> plugin. Please see there for a detailed description.
1656 The following options are valid within B<Key> blocks:
1660 =item B<Type> I<Type>
1662 Sets the type used to dispatch the values to the daemon. Detailed information
1663 about types and their configuration can be found in L<types.db(5)>. This
1664 option is mandatory.
1666 =item B<Instance> I<Instance>
1668 Type-instance to use. Defaults to the current map key or current string array element value.
1672 =head2 Plugin C<curl_xml>
1674 The B<curl_xml plugin> uses B<libcurl> (L<http://curl.haxx.se/>) and B<libxml2>
1675 (L<http://xmlsoft.org/>) to retrieve XML data via cURL.
1678 <URL "http://localhost/stats.xml">
1680 Instance "some_instance"
1685 CACert "/path/to/ca.crt"
1686 Header "X-Custom-Header: foobar"
1689 <XPath "table[@id=\"magic_level\"]/tr">
1691 #InstancePrefix "prefix-"
1692 InstanceFrom "td[1]"
1693 ValuesFrom "td[2]/span[@class=\"level\"]"
1698 In the B<Plugin> block, there may be one or more B<URL> blocks, each defining a
1699 URL to be fetched using libcurl. Within each B<URL> block there are
1700 options which specify the connection parameters, for example authentication
1701 information, and one or more B<XPath> blocks.
1703 Each B<XPath> block specifies how to get one type of information. The
1704 string argument must be a valid XPath expression which returns a list
1705 of "base elements". One value is dispatched for each "base element". The
1706 I<type instance> and values are looked up using further I<XPath> expressions
1707 that should be relative to the base element.
1709 Within the B<URL> block the following options are accepted:
1713 =item B<Host> I<Name>
1715 Use I<Name> as the host name when submitting values. Defaults to the global
1718 =item B<Instance> I<Instance>
1720 Use I<Instance> as the plugin instance when submitting values. Defaults to an
1721 empty string (no plugin instance).
1723 =item B<Namespace> I<Prefix> I<URL>
1725 If an XPath expression references namespaces, they must be specified
1726 with this option. I<Prefix> is the "namespace prefix" used in the XML document.
1727 I<URL> is the "namespace name", an URI reference uniquely identifying the
1728 namespace. The option can be repeated to register multiple namespaces.
1732 Namespace "s" "http://schemas.xmlsoap.org/soap/envelope/"
1733 Namespace "m" "http://www.w3.org/1998/Math/MathML"
1735 =item B<User> I<User>
1737 =item B<Password> I<Password>
1739 =item B<Digest> B<true>|B<false>
1741 =item B<VerifyPeer> B<true>|B<false>
1743 =item B<VerifyHost> B<true>|B<false>
1745 =item B<CACert> I<CA Cert File>
1747 =item B<Header> I<Header>
1749 =item B<Post> I<Body>
1751 =item B<Timeout> I<Milliseconds>
1753 These options behave exactly equivalent to the appropriate options of the
1754 I<cURL plugin>. Please see there for a detailed description.
1756 =item E<lt>B<XPath> I<XPath-expression>E<gt>
1758 Within each B<URL> block, there must be one or more B<XPath> blocks. Each
1759 B<XPath> block specifies how to get one type of information. The string
1760 argument must be a valid XPath expression which returns a list of "base
1761 elements". One value is dispatched for each "base element".
1763 Within the B<XPath> block the following options are accepted:
1767 =item B<Type> I<Type>
1769 Specifies the I<Type> used for submitting patches. This determines the number
1770 of values that are required / expected and whether the strings are parsed as
1771 signed or unsigned integer or as double values. See L<types.db(5)> for details.
1772 This option is required.
1774 =item B<InstancePrefix> I<InstancePrefix>
1776 Prefix the I<type instance> with I<InstancePrefix>. The values are simply
1777 concatenated together without any separator.
1778 This option is optional.
1780 =item B<InstanceFrom> I<InstanceFrom>
1782 Specifies a XPath expression to use for determining the I<type instance>. The
1783 XPath expression must return exactly one element. The element's value is then
1784 used as I<type instance>, possibly prefixed with I<InstancePrefix> (see above).
1786 This value is required. As a special exception, if the "base XPath expression"
1787 (the argument to the B<XPath> block) returns exactly one argument, then this
1788 option may be omitted.
1790 =item B<ValuesFrom> I<ValuesFrom> [I<ValuesFrom> ...]
1792 Specifies one or more XPath expression to use for reading the values. The
1793 number of XPath expressions must match the number of data sources in the
1794 I<type> specified with B<Type> (see above). Each XPath expression must return
1795 exactly one element. The element's value is then parsed as a number and used as
1796 value for the appropriate value in the value list dispatched to the daemon.
1802 =head2 Plugin C<dbi>
1804 This plugin uses the B<dbi> library (L<http://libdbi.sourceforge.net/>) to
1805 connect to various databases, execute I<SQL> statements and read back the
1806 results. I<dbi> is an acronym for "database interface" in case you were
1807 wondering about the name. You can configure how each column is to be
1808 interpreted and the plugin will generate one or more data sets from each row
1809 returned according to these rules.
1811 Because the plugin is very generic, the configuration is a little more complex
1812 than those of other plugins. It usually looks something like this:
1815 <Query "out_of_stock">
1816 Statement "SELECT category, COUNT(*) AS value FROM products WHERE in_stock = 0 GROUP BY category"
1817 # Use with MySQL 5.0.0 or later
1821 InstancePrefix "out_of_stock"
1822 InstancesFrom "category"
1826 <Database "product_information">
1828 DriverOption "host" "localhost"
1829 DriverOption "username" "collectd"
1830 DriverOption "password" "aZo6daiw"
1831 DriverOption "dbname" "prod_info"
1832 SelectDB "prod_info"
1833 Query "out_of_stock"
1837 The configuration above defines one query with one result and one database. The
1838 query is then linked to the database with the B<Query> option I<within> the
1839 B<E<lt>DatabaseE<gt>> block. You can have any number of queries and databases
1840 and you can also use the B<Include> statement to split up the configuration
1841 file in multiple, smaller files. However, the B<E<lt>QueryE<gt>> block I<must>
1842 precede the B<E<lt>DatabaseE<gt>> blocks, because the file is interpreted from
1845 The following is a complete list of options:
1847 =head3 B<Query> blocks
1849 Query blocks define I<SQL> statements and how the returned data should be
1850 interpreted. They are identified by the name that is given in the opening line
1851 of the block. Thus the name needs to be unique. Other than that, the name is
1852 not used in collectd.
1854 In each B<Query> block, there is one or more B<Result> blocks. B<Result> blocks
1855 define which column holds which value or instance information. You can use
1856 multiple B<Result> blocks to create multiple values from one returned row. This
1857 is especially useful, when queries take a long time and sending almost the same
1858 query again and again is not desirable.
1862 <Query "environment">
1863 Statement "select station, temperature, humidity from environment"
1866 # InstancePrefix "foo"
1867 InstancesFrom "station"
1868 ValuesFrom "temperature"
1872 InstancesFrom "station"
1873 ValuesFrom "humidity"
1877 The following options are accepted:
1881 =item B<Statement> I<SQL>
1883 Sets the statement that should be executed on the server. This is B<not>
1884 interpreted by collectd, but simply passed to the database server. Therefore,
1885 the SQL dialect that's used depends on the server collectd is connected to.
1887 The query has to return at least two columns, one for the instance and one
1888 value. You cannot omit the instance, even if the statement is guaranteed to
1889 always return exactly one line. In that case, you can usually specify something
1892 Statement "SELECT \"instance\", COUNT(*) AS value FROM table"
1894 (That works with MySQL but may not be valid SQL according to the spec. If you
1895 use a more strict database server, you may have to select from a dummy table or
1898 Please note that some databases, for example B<Oracle>, will fail if you
1899 include a semicolon at the end of the statement.
1901 =item B<MinVersion> I<Version>
1903 =item B<MaxVersion> I<Value>
1905 Only use this query for the specified database version. You can use these
1906 options to provide multiple queries with the same name but with a slightly
1907 different syntax. The plugin will use only those queries, where the specified
1908 minimum and maximum versions fit the version of the database in use.
1910 The database version is determined by C<dbi_conn_get_engine_version>, see the
1911 L<libdbi documentation|http://libdbi.sourceforge.net/docs/programmers-guide/reference-conn.html#DBI-CONN-GET-ENGINE-VERSION>
1912 for details. Basically, each part of the version is assumed to be in the range
1913 from B<00> to B<99> and all dots are removed. So version "4.1.2" becomes
1914 "40102", version "5.0.42" becomes "50042".
1916 B<Warning:> The plugin will use B<all> matching queries, so if you specify
1917 multiple queries with the same name and B<overlapping> ranges, weird stuff will
1918 happen. Don't to it! A valid example would be something along these lines:
1929 In the above example, there are three ranges that don't overlap. The last one
1930 goes from version "5.1.0" to infinity, meaning "all later versions". Versions
1931 before "4.0.0" are not specified.
1933 =item B<Type> I<Type>
1935 The B<type> that's used for each line returned. See L<types.db(5)> for more
1936 details on how types are defined. In short: A type is a predefined layout of
1937 data and the number of values and type of values has to match the type
1940 If you specify "temperature" here, you need exactly one gauge column. If you
1941 specify "if_octets", you will need two counter columns. See the B<ValuesFrom>
1944 There must be exactly one B<Type> option inside each B<Result> block.
1946 =item B<InstancePrefix> I<prefix>
1948 Prepends I<prefix> to the type instance. If B<InstancesFrom> (see below) is not
1949 given, the string is simply copied. If B<InstancesFrom> is given, I<prefix> and
1950 all strings returned in the appropriate columns are concatenated together,
1951 separated by dashes I<("-")>.
1953 =item B<InstancesFrom> I<column0> [I<column1> ...]
1955 Specifies the columns whose values will be used to create the "type-instance"
1956 for each row. If you specify more than one column, the value of all columns
1957 will be joined together with dashes I<("-")> as separation characters.
1959 The plugin itself does not check whether or not all built instances are
1960 different. It's your responsibility to assure that each is unique. This is
1961 especially true, if you do not specify B<InstancesFrom>: B<You> have to make
1962 sure that only one row is returned in this case.
1964 If neither B<InstancePrefix> nor B<InstancesFrom> is given, the type-instance
1967 =item B<ValuesFrom> I<column0> [I<column1> ...]
1969 Names the columns whose content is used as the actual data for the data sets
1970 that are dispatched to the daemon. How many such columns you need is determined
1971 by the B<Type> setting above. If you specify too many or not enough columns,
1972 the plugin will complain about that and no data will be submitted to the
1975 The actual data type in the columns is not that important. The plugin will
1976 automatically cast the values to the right type if it know how to do that. So
1977 it should be able to handle integer an floating point types, as well as strings
1978 (if they include a number at the beginning).
1980 There must be at least one B<ValuesFrom> option inside each B<Result> block.
1982 =item B<MetadataFrom> [I<column0> I<column1> ...]
1984 Names the columns whose content is used as metadata for the data sets
1985 that are dispatched to the daemon.
1987 The actual data type in the columns is not that important. The plugin will
1988 automatically cast the values to the right type if it know how to do that. So
1989 it should be able to handle integer an floating point types, as well as strings
1990 (if they include a number at the beginning).
1994 =head3 B<Database> blocks
1996 Database blocks define a connection to a database and which queries should be
1997 sent to that database. Since the used "dbi" library can handle a wide variety
1998 of databases, the configuration is very generic. If in doubt, refer to libdbi's
1999 documentationE<nbsp>- we stick as close to the terminology used there.
2001 Each database needs a "name" as string argument in the starting tag of the
2002 block. This name will be used as "PluginInstance" in the values submitted to
2003 the daemon. Other than that, that name is not used.
2007 =item B<Driver> I<Driver>
2009 Specifies the driver to use to connect to the database. In many cases those
2010 drivers are named after the database they can connect to, but this is not a
2011 technical necessity. These drivers are sometimes referred to as "DBD",
2012 B<D>ataB<B>ase B<D>river, and some distributions ship them in separate
2013 packages. Drivers for the "dbi" library are developed by the B<libdbi-drivers>
2014 project at L<http://libdbi-drivers.sourceforge.net/>.
2016 You need to give the driver name as expected by the "dbi" library here. You
2017 should be able to find that in the documentation for each driver. If you
2018 mistype the driver name, the plugin will dump a list of all known driver names
2021 =item B<DriverOption> I<Key> I<Value>
2023 Sets driver-specific options. What option a driver supports can be found in the
2024 documentation for each driver, somewhere at
2025 L<http://libdbi-drivers.sourceforge.net/>. However, the options "host",
2026 "username", "password", and "dbname" seem to be deE<nbsp>facto standards.
2028 DBDs can register two types of options: String options and numeric options. The
2029 plugin will use the C<dbi_conn_set_option> function when the configuration
2030 provides a string and the C<dbi_conn_require_option_numeric> function when the
2031 configuration provides a number. So these two lines will actually result in
2032 different calls being used:
2034 DriverOption "Port" 1234 # numeric
2035 DriverOption "Port" "1234" # string
2037 Unfortunately, drivers are not too keen to report errors when an unknown option
2038 is passed to them, so invalid settings here may go unnoticed. This is not the
2039 plugin's fault, it will report errors if it gets them from the libraryE<nbsp>/
2040 the driver. If a driver complains about an option, the plugin will dump a
2041 complete list of all options understood by that driver to the log. There is no
2042 way to programatically find out if an option expects a string or a numeric
2043 argument, so you will have to refer to the appropriate DBD's documentation to
2044 find this out. Sorry.
2046 =item B<SelectDB> I<Database>
2048 In some cases, the database name you connect with is not the database name you
2049 want to use for querying data. If this option is set, the plugin will "select"
2050 (switch to) that database after the connection is established.
2052 =item B<Query> I<QueryName>
2054 Associates the query named I<QueryName> with this database connection. The
2055 query needs to be defined I<before> this statement, i.E<nbsp>e. all query
2056 blocks you want to refer to must be placed above the database block you want to
2059 =item B<Host> I<Hostname>
2061 Sets the B<host> field of I<value lists> to I<Hostname> when dispatching
2062 values. Defaults to the global hostname setting.
2070 =item B<Device> I<Device>
2072 Select partitions based on the devicename.
2074 =item B<MountPoint> I<Directory>
2076 Select partitions based on the mountpoint.
2078 =item B<FSType> I<FSType>
2080 Select partitions based on the filesystem type.
2082 =item B<IgnoreSelected> B<true>|B<false>
2084 Invert the selection: If set to true, all partitions B<except> the ones that
2085 match any one of the criteria are collected. By default only selected
2086 partitions are collected if a selection is made. If no selection is configured
2087 at all, B<all> partitions are selected.
2089 =item B<ReportByDevice> B<true>|B<false>
2091 Report using the device name rather than the mountpoint. i.e. with this I<false>,
2092 (the default), it will report a disk as "root", but with it I<true>, it will be
2093 "sda1" (or whichever).
2095 =item B<ReportInodes> B<true>|B<false>
2097 Enables or disables reporting of free, reserved and used inodes. Defaults to
2098 inode collection being disabled.
2100 Enable this option if inodes are a scarce resource for you, usually because
2101 many small files are stored on the disk. This is a usual scenario for mail
2102 transfer agents and web caches.
2104 =item B<ValuesAbsolute> B<true>|B<false>
2106 Enables or disables reporting of free and used disk space in 1K-blocks.
2107 Defaults to B<true>.
2109 =item B<ValuesPercentage> B<false>|B<true>
2111 Enables or disables reporting of free and used disk space in percentage.
2112 Defaults to B<false>.
2114 This is useful for deploying I<collectd> on the cloud, where machines with
2115 different disk size may exist. Then it is more practical to configure
2116 thresholds based on relative disk size.
2120 =head2 Plugin C<disk>
2122 The C<disk> plugin collects information about the usage of physical disks and
2123 logical disks (partitions). Values collected are the number of octets written
2124 to and read from a disk or partition, the number of read/write operations
2125 issued to the disk and a rather complex "time" it took for these commands to be
2128 Using the following two options you can ignore some disks or configure the
2129 collection only of specific disks.
2133 =item B<Disk> I<Name>
2135 Select the disk I<Name>. Whether it is collected or ignored depends on the
2136 B<IgnoreSelected> setting, see below. As with other plugins that use the
2137 daemon's ignorelist functionality, a string that starts and ends with a slash
2138 is interpreted as a regular expression. Examples:
2143 =item B<IgnoreSelected> B<true>|B<false>
2145 Sets whether selected disks, i.E<nbsp>e. the ones matches by any of the B<Disk>
2146 statements, are ignored or if all other disks are ignored. The behavior
2147 (hopefully) is intuitive: If no B<Disk> option is configured, all disks are
2148 collected. If at least one B<Disk> option is given and no B<IgnoreSelected> or
2149 set to B<false>, B<only> matching disks will be collected. If B<IgnoreSelected>
2150 is set to B<true>, all disks are collected B<except> the ones matched.
2152 =item B<UseBSDName> B<true>|B<false>
2154 Whether to use the device's "BSD Name", on MacE<nbsp>OSE<nbsp>X, instead of the
2155 default major/minor numbers. Requires collectd to be built with Apple's
2158 =item B<UdevNameAttr> I<Attribute>
2160 Attempt to override disk instance name with the value of a specified udev
2161 attribute when built with B<libudev>. If the attribute is not defined for the
2162 given device, the default name is used. Example:
2164 UdevNameAttr "DM_NAME"
2168 =head2 Plugin C<dns>
2172 =item B<Interface> I<Interface>
2174 The dns plugin uses B<libpcap> to capture dns traffic and analyzes it. This
2175 option sets the interface that should be used. If this option is not set, or
2176 set to "any", the plugin will try to get packets from B<all> interfaces. This
2177 may not work on certain platforms, such as MacE<nbsp>OSE<nbsp>X.
2179 =item B<IgnoreSource> I<IP-address>
2181 Ignore packets that originate from this address.
2183 =item B<SelectNumericQueryTypes> B<true>|B<false>
2185 Enabled by default, collects unknown (and thus presented as numeric only) query types.
2189 =head2 Plugin C<email>
2193 =item B<SocketFile> I<Path>
2195 Sets the socket-file which is to be created.
2197 =item B<SocketGroup> I<Group>
2199 If running as root change the group of the UNIX-socket after it has been
2200 created. Defaults to B<collectd>.
2202 =item B<SocketPerms> I<Permissions>
2204 Change the file permissions of the UNIX-socket after it has been created. The
2205 permissions must be given as a numeric, octal value as you would pass to
2206 L<chmod(1)>. Defaults to B<0770>.
2208 =item B<MaxConns> I<Number>
2210 Sets the maximum number of connections that can be handled in parallel. Since
2211 this many threads will be started immediately setting this to a very high
2212 value will waste valuable resources. Defaults to B<5> and will be forced to be
2213 at most B<16384> to prevent typos and dumb mistakes.
2217 =head2 Plugin C<ethstat>
2219 The I<ethstat plugin> collects information about network interface cards (NICs)
2220 by talking directly with the underlying kernel driver using L<ioctl(2)>.
2226 Map "rx_csum_offload_errors" "if_rx_errors" "checksum_offload"
2227 Map "multicast" "if_multicast"
2234 =item B<Interface> I<Name>
2236 Collect statistical information about interface I<Name>.
2238 =item B<Map> I<Name> I<Type> [I<TypeInstance>]
2240 By default, the plugin will submit values as type C<derive> and I<type
2241 instance> set to I<Name>, the name of the metric as reported by the driver. If
2242 an appropriate B<Map> option exists, the given I<Type> and, optionally,
2243 I<TypeInstance> will be used.
2245 =item B<MappedOnly> B<true>|B<false>
2247 When set to B<true>, only metrics that can be mapped to to a I<type> will be
2248 collected, all other metrics will be ignored. Defaults to B<false>.
2252 =head2 Plugin C<exec>
2254 Please make sure to read L<collectd-exec(5)> before using this plugin. It
2255 contains valuable information on when the executable is executed and the
2256 output that is expected from it.
2260 =item B<Exec> I<User>[:[I<Group>]] I<Executable> [I<E<lt>argE<gt>> [I<E<lt>argE<gt>> ...]]
2262 =item B<NotificationExec> I<User>[:[I<Group>]] I<Executable> [I<E<lt>argE<gt>> [I<E<lt>argE<gt>> ...]]
2264 Execute the executable I<Executable> as user I<User>. If the user name is
2265 followed by a colon and a group name, the effective group is set to that group.
2266 The real group and saved-set group will be set to the default group of that
2267 user. If no group is given the effective group ID will be the same as the real
2270 Please note that in order to change the user and/or group the daemon needs
2271 superuser privileges. If the daemon is run as an unprivileged user you must
2272 specify the same user/group here. If the daemon is run with superuser
2273 privileges, you must supply a non-root user here.
2275 The executable may be followed by optional arguments that are passed to the
2276 program. Please note that due to the configuration parsing numbers and boolean
2277 values may be changed. If you want to be absolutely sure that something is
2278 passed as-is please enclose it in quotes.
2280 The B<Exec> and B<NotificationExec> statements change the semantics of the
2281 programs executed, i.E<nbsp>e. the data passed to them and the response
2282 expected from them. This is documented in great detail in L<collectd-exec(5)>.
2286 =head2 Plugin C<fhcount>
2288 The C<fhcount> plugin provides statistics about used, unused and total number of
2291 The I<fhcount plugin> provides the following configuration options:
2295 =item B<ValuesAbsolute> B<true>|B<false>
2297 Enables or disables reporting of file handles usage in absolute numbers,
2298 e.g. file handles used. Defaults to B<true>.
2300 =item B<ValuesPercentage> B<false>|B<true>
2302 Enables or disables reporting of file handles usage in percentages, e.g.
2303 percent of file handles used. Defaults to B<false>.
2307 =head2 Plugin C<filecount>
2309 The C<filecount> plugin counts the number of files in a certain directory (and
2310 its subdirectories) and their combined size. The configuration is very straight
2313 <Plugin "filecount">
2314 <Directory "/var/qmail/queue/mess">
2315 Instance "qmail-message"
2317 <Directory "/var/qmail/queue/todo">
2318 Instance "qmail-todo"
2320 <Directory "/var/lib/php5">
2321 Instance "php5-sessions"
2326 The example above counts the number of files in QMail's queue directories and
2327 the number of PHP5 sessions. Jfiy: The "todo" queue holds the messages that
2328 QMail has not yet looked at, the "message" queue holds the messages that were
2329 classified into "local" and "remote".
2331 As you can see, the configuration consists of one or more C<Directory> blocks,
2332 each of which specifies a directory in which to count the files. Within those
2333 blocks, the following options are recognized:
2337 =item B<Instance> I<Instance>
2339 Sets the plugin instance to I<Instance>. That instance name must be unique, but
2340 it's your responsibility, the plugin doesn't check for that. If not given, the
2341 instance is set to the directory name with all slashes replaced by underscores
2342 and all leading underscores removed.
2344 =item B<Name> I<Pattern>
2346 Only count files that match I<Pattern>, where I<Pattern> is a shell-like
2347 wildcard as understood by L<fnmatch(3)>. Only the B<filename> is checked
2348 against the pattern, not the entire path. In case this makes it easier for you:
2349 This option has been named after the B<-name> parameter to L<find(1)>.
2351 =item B<MTime> I<Age>
2353 Count only files of a specific age: If I<Age> is greater than zero, only files
2354 that haven't been touched in the last I<Age> seconds are counted. If I<Age> is
2355 a negative number, this is inversed. For example, if B<-60> is specified, only
2356 files that have been modified in the last minute will be counted.
2358 The number can also be followed by a "multiplier" to easily specify a larger
2359 timespan. When given in this notation, the argument must in quoted, i.E<nbsp>e.
2360 must be passed as string. So the B<-60> could also be written as B<"-1m"> (one
2361 minute). Valid multipliers are C<s> (second), C<m> (minute), C<h> (hour), C<d>
2362 (day), C<w> (week), and C<y> (year). There is no "month" multiplier. You can
2363 also specify fractional numbers, e.E<nbsp>g. B<"0.5d"> is identical to
2366 =item B<Size> I<Size>
2368 Count only files of a specific size. When I<Size> is a positive number, only
2369 files that are at least this big are counted. If I<Size> is a negative number,
2370 this is inversed, i.E<nbsp>e. only files smaller than the absolute value of
2371 I<Size> are counted.
2373 As with the B<MTime> option, a "multiplier" may be added. For a detailed
2374 description see above. Valid multipliers here are C<b> (byte), C<k> (kilobyte),
2375 C<m> (megabyte), C<g> (gigabyte), C<t> (terabyte), and C<p> (petabyte). Please
2376 note that there are 1000 bytes in a kilobyte, not 1024.
2378 =item B<Recursive> I<true>|I<false>
2380 Controls whether or not to recurse into subdirectories. Enabled by default.
2382 =item B<IncludeHidden> I<true>|I<false>
2384 Controls whether or not to include "hidden" files and directories in the count.
2385 "Hidden" files and directories are those, whose name begins with a dot.
2386 Defaults to I<false>, i.e. by default hidden files and directories are ignored.
2390 =head2 Plugin C<GenericJMX>
2392 The I<GenericJMX plugin> is written in I<Java> and therefore documented in
2393 L<collectd-java(5)>.
2395 =head2 Plugin C<gmond>
2397 The I<gmond> plugin received the multicast traffic sent by B<gmond>, the
2398 statistics collection daemon of Ganglia. Mappings for the standard "metrics"
2399 are built-in, custom mappings may be added via B<Metric> blocks, see below.
2404 MCReceiveFrom "239.2.11.71" "8649"
2405 <Metric "swap_total">
2407 TypeInstance "total"
2410 <Metric "swap_free">
2417 The following metrics are built-in:
2423 load_one, load_five, load_fifteen
2427 cpu_user, cpu_system, cpu_idle, cpu_nice, cpu_wio
2431 mem_free, mem_shared, mem_buffers, mem_cached, mem_total
2443 Available configuration options:
2447 =item B<MCReceiveFrom> I<MCGroup> [I<Port>]
2449 Sets sets the multicast group and UDP port to which to subscribe.
2451 Default: B<239.2.11.71>E<nbsp>/E<nbsp>B<8649>
2453 =item E<lt>B<Metric> I<Name>E<gt>
2455 These blocks add a new metric conversion to the internal table. I<Name>, the
2456 string argument to the B<Metric> block, is the metric name as used by Ganglia.
2460 =item B<Type> I<Type>
2462 Type to map this metric to. Required.
2464 =item B<TypeInstance> I<Instance>
2466 Type-instance to use. Optional.
2468 =item B<DataSource> I<Name>
2470 Data source to map this metric to. If the configured type has exactly one data
2471 source, this is optional. Otherwise the option is required.
2477 =head2 Plugin C<hddtemp>
2479 To get values from B<hddtemp> collectd connects to B<localhost> (127.0.0.1),
2480 port B<7634/tcp>. The B<Host> and B<Port> options can be used to change these
2481 default values, see below. C<hddtemp> has to be running to work correctly. If
2482 C<hddtemp> is not running timeouts may appear which may interfere with other
2485 The B<hddtemp> homepage can be found at
2486 L<http://www.guzu.net/linux/hddtemp.php>.
2490 =item B<Host> I<Hostname>
2492 Hostname to connect to. Defaults to B<127.0.0.1>.
2494 =item B<Port> I<Port>
2496 TCP-Port to connect to. Defaults to B<7634>.
2500 =head2 Plugin C<interface>
2504 =item B<Interface> I<Interface>
2506 Select this interface. By default these interfaces will then be collected. For
2507 a more detailed description see B<IgnoreSelected> below.
2509 =item B<IgnoreSelected> I<true>|I<false>
2511 If no configuration if given, the B<traffic>-plugin will collect data from
2512 all interfaces. This may not be practical, especially for loopback- and
2513 similar interfaces. Thus, you can use the B<Interface>-option to pick the
2514 interfaces you're interested in. Sometimes, however, it's easier/preferred
2515 to collect all interfaces I<except> a few ones. This option enables you to
2516 do that: By setting B<IgnoreSelected> to I<true> the effect of
2517 B<Interface> is inverted: All selected interfaces are ignored and all
2518 other interfaces are collected.
2522 =head2 Plugin C<ipmi>
2526 =item B<Sensor> I<Sensor>
2528 Selects sensors to collect or to ignore, depending on B<IgnoreSelected>.
2530 =item B<IgnoreSelected> I<true>|I<false>
2532 If no configuration if given, the B<ipmi> plugin will collect data from all
2533 sensors found of type "temperature", "voltage", "current" and "fanspeed".
2534 This option enables you to do that: By setting B<IgnoreSelected> to I<true>
2535 the effect of B<Sensor> is inverted: All selected sensors are ignored and
2536 all other sensors are collected.
2538 =item B<NotifySensorAdd> I<true>|I<false>
2540 If a sensor appears after initialization time of a minute a notification
2543 =item B<NotifySensorRemove> I<true>|I<false>
2545 If a sensor disappears a notification is sent.
2547 =item B<NotifySensorNotPresent> I<true>|I<false>
2549 If you have for example dual power supply and one of them is (un)plugged then
2550 a notification is sent.
2554 =head2 Plugin C<iptables>
2558 =item B<Chain> I<Table> I<Chain> [I<Comment|Number> [I<Name>]]
2560 Select the rules to count. If only I<Table> and I<Chain> are given, this plugin
2561 will collect the counters of all rules which have a comment-match. The comment
2562 is then used as type-instance.
2564 If I<Comment> or I<Number> is given, only the rule with the matching comment or
2565 the I<n>th rule will be collected. Again, the comment (or the number) will be
2566 used as the type-instance.
2568 If I<Name> is supplied, it will be used as the type-instance instead of the
2569 comment or the number.
2573 =head2 Plugin C<irq>
2579 Select this irq. By default these irqs will then be collected. For a more
2580 detailed description see B<IgnoreSelected> below.
2582 =item B<IgnoreSelected> I<true>|I<false>
2584 If no configuration if given, the B<irq>-plugin will collect data from all
2585 irqs. This may not be practical, especially if no interrupts happen. Thus, you
2586 can use the B<Irq>-option to pick the interrupt you're interested in.
2587 Sometimes, however, it's easier/preferred to collect all interrupts I<except> a
2588 few ones. This option enables you to do that: By setting B<IgnoreSelected> to
2589 I<true> the effect of B<Irq> is inverted: All selected interrupts are ignored
2590 and all other interrupts are collected.
2594 =head2 Plugin C<java>
2596 The I<Java> plugin makes it possible to write extensions for collectd in Java.
2597 This section only discusses the syntax and semantic of the configuration
2598 options. For more in-depth information on the I<Java> plugin, please read
2599 L<collectd-java(5)>.
2604 JVMArg "-verbose:jni"
2605 JVMArg "-Djava.class.path=/opt/collectd/lib/collectd/bindings/java"
2606 LoadPlugin "org.collectd.java.Foobar"
2607 <Plugin "org.collectd.java.Foobar">
2608 # To be parsed by the plugin
2612 Available configuration options:
2616 =item B<JVMArg> I<Argument>
2618 Argument that is to be passed to the I<Java Virtual Machine> (JVM). This works
2619 exactly the way the arguments to the I<java> binary on the command line work.
2620 Execute C<javaE<nbsp>--help> for details.
2622 Please note that B<all> these options must appear B<before> (i.E<nbsp>e. above)
2623 any other options! When another option is found, the JVM will be started and
2624 later options will have to be ignored!
2626 =item B<LoadPlugin> I<JavaClass>
2628 Instantiates a new I<JavaClass> object. The constructor of this object very
2629 likely then registers one or more callback methods with the server.
2631 See L<collectd-java(5)> for details.
2633 When the first such option is found, the virtual machine (JVM) is created. This
2634 means that all B<JVMArg> options must appear before (i.E<nbsp>e. above) all
2635 B<LoadPlugin> options!
2637 =item B<Plugin> I<Name>
2639 The entire block is passed to the Java plugin as an
2640 I<org.collectd.api.OConfigItem> object.
2642 For this to work, the plugin has to register a configuration callback first,
2643 see L<collectd-java(5)/"config callback">. This means, that the B<Plugin> block
2644 must appear after the appropriate B<LoadPlugin> block. Also note, that I<Name>
2645 depends on the (Java) plugin registering the callback and is completely
2646 independent from the I<JavaClass> argument passed to B<LoadPlugin>.
2650 =head2 Plugin C<load>
2652 The I<Load plugin> collects the system load. These numbers give a rough overview
2653 over the utilization of a machine. The system load is defined as the number of
2654 runnable tasks in the run-queue and is provided by many operating systems as a
2655 one, five or fifteen minute average.
2657 The following configuration options are available:
2661 =item B<ReportRelative> B<false>|B<true>
2663 When enabled, system load divided by number of available CPU cores is reported
2664 for intervals 1 min, 5 min and 15 min. Defaults to false.
2669 =head2 Plugin C<logfile>
2673 =item B<LogLevel> B<debug|info|notice|warning|err>
2675 Sets the log-level. If, for example, set to B<notice>, then all events with
2676 severity B<notice>, B<warning>, or B<err> will be written to the logfile.
2678 Please note that B<debug> is only available if collectd has been compiled with
2681 =item B<File> I<File>
2683 Sets the file to write log messages to. The special strings B<stdout> and
2684 B<stderr> can be used to write to the standard output and standard error
2685 channels, respectively. This, of course, only makes much sense when I<collectd>
2686 is running in foreground- or non-daemon-mode.
2688 =item B<Timestamp> B<true>|B<false>
2690 Prefix all lines printed by the current time. Defaults to B<true>.
2692 =item B<PrintSeverity> B<true>|B<false>
2694 When enabled, all lines are prefixed by the severity of the log message, for
2695 example "warning". Defaults to B<false>.
2699 B<Note>: There is no need to notify the daemon after moving or removing the
2700 log file (e.E<nbsp>g. when rotating the logs). The plugin reopens the file
2701 for each line it writes.
2703 =head2 Plugin C<log_logstash>
2705 The I<log logstash plugin> behaves like the logfile plugin but formats
2706 messages as JSON events for logstash to parse and input.
2710 =item B<LogLevel> B<debug|info|notice|warning|err>
2712 Sets the log-level. If, for example, set to B<notice>, then all events with
2713 severity B<notice>, B<warning>, or B<err> will be written to the logfile.
2715 Please note that B<debug> is only available if collectd has been compiled with
2718 =item B<File> I<File>
2720 Sets the file to write log messages to. The special strings B<stdout> and
2721 B<stderr> can be used to write to the standard output and standard error
2722 channels, respectively. This, of course, only makes much sense when I<collectd>
2723 is running in foreground- or non-daemon-mode.
2727 B<Note>: There is no need to notify the daemon after moving or removing the
2728 log file (e.E<nbsp>g. when rotating the logs). The plugin reopens the file
2729 for each line it writes.
2731 =head2 Plugin C<lpar>
2733 The I<LPAR plugin> reads CPU statistics of I<Logical Partitions>, a
2734 virtualization technique for IBM POWER processors. It takes into account CPU
2735 time stolen from or donated to a partition, in addition to the usual user,
2736 system, I/O statistics.
2738 The following configuration options are available:
2742 =item B<CpuPoolStats> B<false>|B<true>
2744 When enabled, statistics about the processor pool are read, too. The partition
2745 needs to have pool authority in order to be able to acquire this information.
2748 =item B<ReportBySerial> B<false>|B<true>
2750 If enabled, the serial of the physical machine the partition is currently
2751 running on is reported as I<hostname> and the logical hostname of the machine
2752 is reported in the I<plugin instance>. Otherwise, the logical hostname will be
2753 used (just like other plugins) and the I<plugin instance> will be empty.
2758 =head2 Plugin C<mbmon>
2760 The C<mbmon plugin> uses mbmon to retrieve temperature, voltage, etc.
2762 Be default collectd connects to B<localhost> (127.0.0.1), port B<411/tcp>. The
2763 B<Host> and B<Port> options can be used to change these values, see below.
2764 C<mbmon> has to be running to work correctly. If C<mbmon> is not running
2765 timeouts may appear which may interfere with other statistics..
2767 C<mbmon> must be run with the -r option ("print TAG and Value format");
2768 Debian's F</etc/init.d/mbmon> script already does this, other people
2769 will need to ensure that this is the case.
2773 =item B<Host> I<Hostname>
2775 Hostname to connect to. Defaults to B<127.0.0.1>.
2777 =item B<Port> I<Port>
2779 TCP-Port to connect to. Defaults to B<411>.
2785 The C<md plugin> collects information from Linux Software-RAID devices (md).
2787 All reported values are of the type C<md_disks>. Reported type instances are
2788 I<active>, I<failed> (present but not operational), I<spare> (hot stand-by) and
2789 I<missing> (physically absent) disks.
2793 =item B<Device> I<Device>
2795 Select md devices based on device name. The I<device name> is the basename of
2796 the device, i.e. the name of the block device without the leading C</dev/>.
2797 See B<IgnoreSelected> for more details.
2799 =item B<IgnoreSelected> B<true>|B<false>
2801 Invert device selection: If set to B<true>, all md devices B<except> those
2802 listed using B<Device> are collected. If B<false> (the default), only those
2803 listed are collected. If no configuration is given, the B<md> plugin will
2804 collect data from all md devices.
2808 =head2 Plugin C<memcachec>
2810 The C<memcachec plugin> connects to a memcached server, queries one or more
2811 given I<pages> and parses the returned data according to user specification.
2812 The I<matches> used are the same as the matches used in the C<curl> and C<tail>
2815 In order to talk to the memcached server, this plugin uses the I<libmemcached>
2816 library. Please note that there is another library with a very similar name,
2817 libmemcache (notice the missing `d'), which is not applicable.
2819 Synopsis of the configuration:
2821 <Plugin "memcachec">
2822 <Page "plugin_instance">
2826 Regex "(\\d+) bytes sent"
2829 Instance "type_instance"
2834 The configuration options are:
2838 =item E<lt>B<Page> I<Name>E<gt>
2840 Each B<Page> block defines one I<page> to be queried from the memcached server.
2841 The block requires one string argument which is used as I<plugin instance>.
2843 =item B<Server> I<Address>
2845 Sets the server address to connect to when querying the page. Must be inside a
2850 When connected to the memcached server, asks for the page I<Key>.
2852 =item E<lt>B<Match>E<gt>
2854 Match blocks define which strings to look for and how matches substrings are
2855 interpreted. For a description of match blocks, please see L<"Plugin tail">.
2859 =head2 Plugin C<memcached>
2861 The B<memcached plugin> connects to a memcached server and queries statistics
2862 about cache utilization, memory and bandwidth used.
2863 L<http://www.danga.com/memcached/>
2865 <Plugin "memcached">
2867 Host "memcache.example.com"
2872 The plugin configuration consists of one or more B<Instance> blocks which
2873 specify one I<memcached> connection each. Within the B<Instance> blocks, the
2874 following options are allowed:
2878 =item B<Host> I<Hostname>
2880 Hostname to connect to. Defaults to B<127.0.0.1>.
2882 =item B<Port> I<Port>
2884 TCP-Port to connect to. Defaults to B<11211>.
2886 =item B<Socket> I<Path>
2888 Connect to I<memcached> using the UNIX domain socket at I<Path>. If this
2889 setting is given, the B<Host> and B<Port> settings are ignored.
2893 =head2 Plugin C<mic>
2895 The B<mic plugin> gathers CPU statistics, memory usage and temperatures from
2896 Intel's Many Integrated Core (MIC) systems.
2905 ShowTemperatures true
2908 IgnoreSelectedTemperature true
2913 IgnoreSelectedPower true
2916 The following options are valid inside the B<PluginE<nbsp>mic> block:
2920 =item B<ShowCPU> B<true>|B<false>
2922 If enabled (the default) a sum of the CPU usage across all cores is reported.
2924 =item B<ShowCPUCores> B<true>|B<false>
2926 If enabled (the default) per-core CPU usage is reported.
2928 =item B<ShowMemory> B<true>|B<false>
2930 If enabled (the default) the physical memory usage of the MIC system is
2933 =item B<ShowTemperatures> B<true>|B<false>
2935 If enabled (the default) various temperatures of the MIC system are reported.
2937 =item B<Temperature> I<Name>
2939 This option controls which temperatures are being reported. Whether matching
2940 temperatures are being ignored or I<only> matching temperatures are reported
2941 depends on the B<IgnoreSelectedTemperature> setting below. By default I<all>
2942 temperatures are reported.
2944 =item B<IgnoreSelectedTemperature> B<false>|B<true>
2946 Controls the behavior of the B<Temperature> setting above. If set to B<false>
2947 (the default) only temperatures matching a B<Temperature> option are reported
2948 or, if no B<Temperature> option is specified, all temperatures are reported. If
2949 set to B<true>, matching temperatures are I<ignored> and all other temperatures
2952 Known temperature names are:
2986 =item B<ShowPower> B<true>|B<false>
2988 If enabled (the default) various temperatures of the MIC system are reported.
2990 =item B<Power> I<Name>
2992 This option controls which power readings are being reported. Whether matching
2993 power readings are being ignored or I<only> matching power readings are reported
2994 depends on the B<IgnoreSelectedPower> setting below. By default I<all>
2995 power readings are reported.
2997 =item B<IgnoreSelectedPower> B<false>|B<true>
2999 Controls the behavior of the B<Power> setting above. If set to B<false>
3000 (the default) only power readings matching a B<Power> option are reported
3001 or, if no B<Power> option is specified, all power readings are reported. If
3002 set to B<true>, matching power readings are I<ignored> and all other power readings
3005 Known power names are:
3011 Total power utilization averaged over Time Window 0 (uWatts).
3015 Total power utilization averaged over Time Window 0 (uWatts).
3019 Instantaneous power (uWatts).
3023 Max instantaneous power (uWatts).
3027 PCI-E connector power (uWatts).
3031 2x3 connector power (uWatts).
3035 2x4 connector power (uWatts).
3043 Uncore rail (uVolts).
3047 Memory subsystem rail (uVolts).
3053 =head2 Plugin C<memory>
3055 The I<memory plugin> provides the following configuration options:
3059 =item B<ValuesAbsolute> B<true>|B<false>
3061 Enables or disables reporting of physical memory usage in absolute numbers,
3062 i.e. bytes. Defaults to B<true>.
3064 =item B<ValuesPercentage> B<false>|B<true>
3066 Enables or disables reporting of physical memory usage in percentages, e.g.
3067 percent of physical memory used. Defaults to B<false>.
3069 This is useful for deploying I<collectd> in a heterogeneous environment in
3070 which the sizes of physical memory vary.
3074 =head2 Plugin C<modbus>
3076 The B<modbus plugin> connects to a Modbus "slave" via Modbus/TCP or Modbus/RTU and
3077 reads register values. It supports reading single registers (unsigned 16E<nbsp>bit
3078 values), large integer values (unsigned 32E<nbsp>bit values) and floating point
3079 values (two registers interpreted as IEEE floats in big endian notation).
3083 <Data "voltage-input-1">
3086 RegisterCmd ReadHolding
3091 <Data "voltage-input-2">
3094 RegisterCmd ReadHolding
3099 <Data "supply-temperature-1">
3102 RegisterCmd ReadHolding
3107 <Host "modbus.example.com">
3108 Address "192.168.0.42"
3113 Instance "power-supply"
3114 Collect "voltage-input-1"
3115 Collect "voltage-input-2"
3120 Device "/dev/ttyUSB0"
3125 Instance "temperature"
3126 Collect "supply-temperature-1"
3132 =item E<lt>B<Data> I<Name>E<gt> blocks
3134 Data blocks define a mapping between register numbers and the "types" used by
3137 Within E<lt>DataE<nbsp>/E<gt> blocks, the following options are allowed:
3141 =item B<RegisterBase> I<Number>
3143 Configures the base register to read from the device. If the option
3144 B<RegisterType> has been set to B<Uint32> or B<Float>, this and the next
3145 register will be read (the register number is increased by one).
3147 =item B<RegisterType> B<Int16>|B<Int32>|B<Uint16>|B<Uint32>|B<Float>
3149 Specifies what kind of data is returned by the device. If the type is B<Int32>,
3150 B<Uint32> or B<Float>, two 16E<nbsp>bit registers will be read and the data is
3151 combined into one value. Defaults to B<Uint16>.
3153 =item B<RegisterCmd> B<ReadHolding>|B<ReadInput>
3155 Specifies register type to be collected from device. Works only with libmodbus
3156 2.9.2 or higher. Defaults to B<ReadHolding>.
3158 =item B<Type> I<Type>
3160 Specifies the "type" (data set) to use when dispatching the value to
3161 I<collectd>. Currently, only data sets with exactly one data source are
3164 =item B<Instance> I<Instance>
3166 Sets the type instance to use when dispatching the value to I<collectd>. If
3167 unset, an empty string (no type instance) is used.
3171 =item E<lt>B<Host> I<Name>E<gt> blocks
3173 Host blocks are used to specify to which hosts to connect and what data to read
3174 from their "slaves". The string argument I<Name> is used as hostname when
3175 dispatching the values to I<collectd>.
3177 Within E<lt>HostE<nbsp>/E<gt> blocks, the following options are allowed:
3181 =item B<Address> I<Hostname>
3183 For Modbus/TCP, specifies the node name (the actual network address) used to
3184 connect to the host. This may be an IP address or a hostname. Please note that
3185 the used I<libmodbus> library only supports IPv4 at the moment.
3187 =item B<Port> I<Service>
3189 for Modbus/TCP, specifies the port used to connect to the host. The port can
3190 either be given as a number or as a service name. Please note that the
3191 I<Service> argument must be a string, even if ports are given in their numerical
3192 form. Defaults to "502".
3194 =item B<Device> I<Devicenode>
3196 For Modbus/RTU, specifies the path to the serial device being used.
3198 =item B<Baudrate> I<Baudrate>
3200 For Modbus/RTU, specifies the baud rate of the serial device.
3201 Note, connections currently support only 8/N/1.
3203 =item B<Interval> I<Interval>
3205 Sets the interval (in seconds) in which the values will be collected from this
3206 host. By default the global B<Interval> setting will be used.
3208 =item E<lt>B<Slave> I<ID>E<gt>
3210 Over each connection, multiple Modbus devices may be reached. The slave ID
3211 is used to specify which device should be addressed. For each device you want
3212 to query, one B<Slave> block must be given.
3214 Within E<lt>SlaveE<nbsp>/E<gt> blocks, the following options are allowed:
3218 =item B<Instance> I<Instance>
3220 Specify the plugin instance to use when dispatching the values to I<collectd>.
3221 By default "slave_I<ID>" is used.
3223 =item B<Collect> I<DataName>
3225 Specifies which data to retrieve from the device. I<DataName> must be the same
3226 string as the I<Name> argument passed to a B<Data> block. You can specify this
3227 option multiple times to collect more than one value from a slave. At least one
3228 B<Collect> option is mandatory.
3236 =head2 Plugin C<mysql>
3238 The C<mysql plugin> requires B<mysqlclient> to be installed. It connects to
3239 one or more databases when started and keeps the connection up as long as
3240 possible. When the connection is interrupted for whatever reason it will try
3241 to re-connect. The plugin will complain loudly in case anything goes wrong.
3243 This plugin issues the MySQL C<SHOW STATUS> / C<SHOW GLOBAL STATUS> command
3244 and collects information about MySQL network traffic, executed statements,
3245 requests, the query cache and threads by evaluating the
3246 C<Bytes_{received,sent}>, C<Com_*>, C<Handler_*>, C<Qcache_*> and C<Threads_*>
3247 return values. Please refer to the B<MySQL reference manual>, I<5.1.6. Server
3248 Status Variables> for an explanation of these values.
3250 Optionally, master and slave statistics may be collected in a MySQL
3251 replication setup. In that case, information about the synchronization state
3252 of the nodes are collected by evaluating the C<Position> return value of the
3253 C<SHOW MASTER STATUS> command and the C<Seconds_Behind_Master>,
3254 C<Read_Master_Log_Pos> and C<Exec_Master_Log_Pos> return values of the
3255 C<SHOW SLAVE STATUS> command. See the B<MySQL reference manual>,
3256 I<12.5.5.21 SHOW MASTER STATUS Syntax> and
3257 I<12.5.5.31 SHOW SLAVE STATUS Syntax> for details.
3274 Socket "/var/run/mysql/mysqld.sock"
3276 SlaveNotifications true
3280 A B<Database> block defines one connection to a MySQL database. It accepts a
3281 single argument which specifies the name of the database. None of the other
3282 options are required. MySQL will use default values as documented in the
3283 section "mysql_real_connect()" in the B<MySQL reference manual>.
3287 =item B<Alias> I<Alias>
3289 Alias to use as sender instead of hostname when reporting. This may be useful
3290 when having cryptic hostnames.
3292 =item B<Host> I<Hostname>
3294 Hostname of the database server. Defaults to B<localhost>.
3296 =item B<User> I<Username>
3298 Username to use when connecting to the database. The user does not have to be
3299 granted any privileges (which is synonym to granting the C<USAGE> privilege),
3300 unless you want to collectd replication statistics (see B<MasterStats> and
3301 B<SlaveStats> below). In this case, the user needs the C<REPLICATION CLIENT>
3302 (or C<SUPER>) privileges. Else, any existing MySQL user will do.
3304 =item B<Password> I<Password>
3306 Password needed to log into the database.
3308 =item B<Database> I<Database>
3310 Select this database. Defaults to I<no database> which is a perfectly reasonable
3311 option for what this plugin does.
3313 =item B<Port> I<Port>
3315 TCP-port to connect to. The port must be specified in its numeric form, but it
3316 must be passed as a string nonetheless. For example:
3320 If B<Host> is set to B<localhost> (the default), this setting has no effect.
3321 See the documentation for the C<mysql_real_connect> function for details.
3323 =item B<Socket> I<Socket>
3325 Specifies the path to the UNIX domain socket of the MySQL server. This option
3326 only has any effect, if B<Host> is set to B<localhost> (the default).
3327 Otherwise, use the B<Port> option above. See the documentation for the
3328 C<mysql_real_connect> function for details.
3330 =item B<InnodbStats> I<true|false>
3332 If enabled, metrics about the InnoDB storage engine are collected.
3333 Disabled by default.
3335 =item B<MasterStats> I<true|false>
3337 =item B<SlaveStats> I<true|false>
3339 Enable the collection of master / slave statistics in a replication setup. In
3340 order to be able to get access to these statistics, the user needs special
3341 privileges. See the B<User> documentation above. Defaults to B<false>.
3343 =item B<SlaveNotifications> I<true|false>
3345 If enabled, the plugin sends a notification if the replication slave I/O and /
3346 or SQL threads are not running. Defaults to B<false>.
3348 =item B<ConnectTimeout> I<Seconds>
3350 Sets the connect timeout for the MySQL client.
3354 =head2 Plugin C<netapp>
3356 The netapp plugin can collect various performance and capacity information
3357 from a NetApp filer using the NetApp API.
3359 Please note that NetApp has a wide line of products and a lot of different
3360 software versions for each of these products. This plugin was developed for a
3361 NetApp FAS3040 running OnTap 7.2.3P8 and tested on FAS2050 7.3.1.1L1,
3362 FAS3140 7.2.5.1 and FAS3020 7.2.4P9. It I<should> work for most combinations of
3363 model and software version but it is very hard to test this.
3364 If you have used this plugin with other models and/or software version, feel
3365 free to send us a mail to tell us about the results, even if it's just a short
3368 To collect these data collectd will log in to the NetApp via HTTP(S) and HTTP
3369 basic authentication.
3371 B<Do not use a regular user for this!> Create a special collectd user with just
3372 the minimum of capabilities needed. The user only needs the "login-http-admin"
3373 capability as well as a few more depending on which data will be collected.
3374 Required capabilities are documented below.
3379 <Host "netapp1.example.com">
3403 IgnoreSelectedIO false
3405 IgnoreSelectedOps false
3406 GetLatency "volume0"
3407 IgnoreSelectedLatency false
3414 IgnoreSelectedCapacity false
3417 IgnoreSelectedSnapshot false
3445 The netapp plugin accepts the following configuration options:
3449 =item B<Host> I<Name>
3451 A host block defines one NetApp filer. It will appear in collectd with the name
3452 you specify here which does not have to be its real name nor its hostname (see
3453 the B<Address> option below).
3455 =item B<VFiler> I<Name>
3457 A B<VFiler> block may only be used inside a host block. It accepts all the
3458 same options as the B<Host> block (except for cascaded B<VFiler> blocks) and
3459 will execute all NetApp API commands in the context of the specified
3460 VFiler(R). It will appear in collectd with the name you specify here which
3461 does not have to be its real name. The VFiler name may be specified using the
3462 B<VFilerName> option. If this is not specified, it will default to the name
3465 The VFiler block inherits all connection related settings from the surrounding
3466 B<Host> block (which appear before the B<VFiler> block) but they may be
3467 overwritten inside the B<VFiler> block.
3469 This feature is useful, for example, when using a VFiler as SnapVault target
3470 (supported since OnTap 8.1). In that case, the SnapVault statistics are not
3471 available in the host filer (vfiler0) but only in the respective VFiler
3474 =item B<Protocol> B<httpd>|B<http>
3476 The protocol collectd will use to query this host.
3484 Valid options: http, https
3486 =item B<Address> I<Address>
3488 The hostname or IP address of the host.
3494 Default: The "host" block's name.
3496 =item B<Port> I<Port>
3498 The TCP port to connect to on the host.
3504 Default: 80 for protocol "http", 443 for protocol "https"
3506 =item B<User> I<User>
3508 =item B<Password> I<Password>
3510 The username and password to use to login to the NetApp.
3516 =item B<VFilerName> I<Name>
3518 The name of the VFiler in which context to execute API commands. If not
3519 specified, the name provided to the B<VFiler> block will be used instead.
3525 Default: name of the B<VFiler> block
3527 B<Note:> This option may only be used inside B<VFiler> blocks.
3529 =item B<Interval> I<Interval>
3535 The following options decide what kind of data will be collected. You can
3536 either use them as a block and fine tune various parameters inside this block,
3537 use them as a single statement to just accept all default values, or omit it to
3538 not collect any data.
3540 The following options are valid inside all blocks:
3544 =item B<Interval> I<Seconds>
3546 Collect the respective statistics every I<Seconds> seconds. Defaults to the
3547 host specific setting.
3551 =head3 The System block
3553 This will collect various performance data about the whole system.
3555 B<Note:> To get this data the collectd user needs the
3556 "api-perf-object-get-instances" capability.
3560 =item B<Interval> I<Seconds>
3562 Collect disk statistics every I<Seconds> seconds.
3564 =item B<GetCPULoad> B<true>|B<false>
3566 If you set this option to true the current CPU usage will be read. This will be
3567 the average usage between all CPUs in your NetApp without any information about
3570 B<Note:> These are the same values that the NetApp CLI command "sysstat"
3571 returns in the "CPU" field.
3579 Result: Two value lists of type "cpu", and type instances "idle" and "system".
3581 =item B<GetInterfaces> B<true>|B<false>
3583 If you set this option to true the current traffic of the network interfaces
3584 will be read. This will be the total traffic over all interfaces of your NetApp
3585 without any information about individual interfaces.
3587 B<Note:> This is the same values that the NetApp CLI command "sysstat" returns
3588 in the "Net kB/s" field.
3598 Result: One value list of type "if_octects".
3600 =item B<GetDiskIO> B<true>|B<false>
3602 If you set this option to true the current IO throughput will be read. This
3603 will be the total IO of your NetApp without any information about individual
3604 disks, volumes or aggregates.
3606 B<Note:> This is the same values that the NetApp CLI command "sysstat" returns
3607 in the "DiskE<nbsp>kB/s" field.
3615 Result: One value list of type "disk_octets".
3617 =item B<GetDiskOps> B<true>|B<false>
3619 If you set this option to true the current number of HTTP, NFS, CIFS, FCP,
3620 iSCSI, etc. operations will be read. This will be the total number of
3621 operations on your NetApp without any information about individual volumes or
3624 B<Note:> These are the same values that the NetApp CLI command "sysstat"
3625 returns in the "NFS", "CIFS", "HTTP", "FCP" and "iSCSI" fields.
3633 Result: A variable number of value lists of type "disk_ops_complex". Each type
3634 of operation will result in one value list with the name of the operation as
3639 =head3 The WAFL block
3641 This will collect various performance data about the WAFL file system. At the
3642 moment this just means cache performance.
3644 B<Note:> To get this data the collectd user needs the
3645 "api-perf-object-get-instances" capability.
3647 B<Note:> The interface to get these values is classified as "Diagnostics" by
3648 NetApp. This means that it is not guaranteed to be stable even between minor
3653 =item B<Interval> I<Seconds>
3655 Collect disk statistics every I<Seconds> seconds.
3657 =item B<GetNameCache> B<true>|B<false>
3665 Result: One value list of type "cache_ratio" and type instance
3668 =item B<GetDirCache> B<true>|B<false>
3676 Result: One value list of type "cache_ratio" and type instance "find_dir_hit".
3678 =item B<GetInodeCache> B<true>|B<false>
3686 Result: One value list of type "cache_ratio" and type instance
3689 =item B<GetBufferCache> B<true>|B<false>
3691 B<Note:> This is the same value that the NetApp CLI command "sysstat" returns
3692 in the "Cache hit" field.
3700 Result: One value list of type "cache_ratio" and type instance "buf_hash_hit".
3704 =head3 The Disks block
3706 This will collect performance data about the individual disks in the NetApp.
3708 B<Note:> To get this data the collectd user needs the
3709 "api-perf-object-get-instances" capability.
3713 =item B<Interval> I<Seconds>
3715 Collect disk statistics every I<Seconds> seconds.
3717 =item B<GetBusy> B<true>|B<false>
3719 If you set this option to true the busy time of all disks will be calculated
3720 and the value of the busiest disk in the system will be written.
3722 B<Note:> This is the same values that the NetApp CLI command "sysstat" returns
3723 in the "Disk util" field. Probably.
3731 Result: One value list of type "percent" and type instance "disk_busy".
3735 =head3 The VolumePerf block
3737 This will collect various performance data about the individual volumes.
3739 You can select which data to collect about which volume using the following
3740 options. They follow the standard ignorelist semantic.
3742 B<Note:> To get this data the collectd user needs the
3743 I<api-perf-object-get-instances> capability.
3747 =item B<Interval> I<Seconds>
3749 Collect volume performance data every I<Seconds> seconds.
3751 =item B<GetIO> I<Volume>
3753 =item B<GetOps> I<Volume>
3755 =item B<GetLatency> I<Volume>
3757 Select the given volume for IO, operations or latency statistics collection.
3758 The argument is the name of the volume without the C</vol/> prefix.
3760 Since the standard ignorelist functionality is used here, you can use a string
3761 starting and ending with a slash to specify regular expression matching: To
3762 match the volumes "vol0", "vol2" and "vol7", you can use this regular
3765 GetIO "/^vol[027]$/"
3767 If no regular expression is specified, an exact match is required. Both,
3768 regular and exact matching are case sensitive.
3770 If no volume was specified at all for either of the three options, that data
3771 will be collected for all available volumes.
3773 =item B<IgnoreSelectedIO> B<true>|B<false>
3775 =item B<IgnoreSelectedOps> B<true>|B<false>
3777 =item B<IgnoreSelectedLatency> B<true>|B<false>
3779 When set to B<true>, the volumes selected for IO, operations or latency
3780 statistics collection will be ignored and the data will be collected for all
3783 When set to B<false>, data will only be collected for the specified volumes and
3784 all other volumes will be ignored.
3786 If no volumes have been specified with the above B<Get*> options, all volumes
3787 will be collected regardless of the B<IgnoreSelected*> option.
3789 Defaults to B<false>
3793 =head3 The VolumeUsage block
3795 This will collect capacity data about the individual volumes.
3797 B<Note:> To get this data the collectd user needs the I<api-volume-list-info>
3802 =item B<Interval> I<Seconds>
3804 Collect volume usage statistics every I<Seconds> seconds.
3806 =item B<GetCapacity> I<VolumeName>
3808 The current capacity of the volume will be collected. This will result in two
3809 to four value lists, depending on the configuration of the volume. All data
3810 sources are of type "df_complex" with the name of the volume as
3813 There will be type_instances "used" and "free" for the number of used and
3814 available bytes on the volume. If the volume has some space reserved for
3815 snapshots, a type_instance "snap_reserved" will be available. If the volume
3816 has SIS enabled, a type_instance "sis_saved" will be available. This is the
3817 number of bytes saved by the SIS feature.
3819 B<Note:> The current NetApp API has a bug that results in this value being
3820 reported as a 32E<nbsp>bit number. This plugin tries to guess the correct
3821 number which works most of the time. If you see strange values here, bug
3822 NetApp support to fix this.
3824 Repeat this option to specify multiple volumes.
3826 =item B<IgnoreSelectedCapacity> B<true>|B<false>
3828 Specify whether to collect only the volumes selected by the B<GetCapacity>
3829 option or to ignore those volumes. B<IgnoreSelectedCapacity> defaults to
3830 B<false>. However, if no B<GetCapacity> option is specified at all, all
3831 capacities will be selected anyway.
3833 =item B<GetSnapshot> I<VolumeName>
3835 Select volumes from which to collect snapshot information.
3837 Usually, the space used for snapshots is included in the space reported as
3838 "used". If snapshot information is collected as well, the space used for
3839 snapshots is subtracted from the used space.
3841 To make things even more interesting, it is possible to reserve space to be
3842 used for snapshots. If the space required for snapshots is less than that
3843 reserved space, there is "reserved free" and "reserved used" space in addition
3844 to "free" and "used". If the space required for snapshots exceeds the reserved
3845 space, that part allocated in the normal space is subtracted from the "used"
3848 Repeat this option to specify multiple volumes.
3850 =item B<IgnoreSelectedSnapshot>
3852 Specify whether to collect only the volumes selected by the B<GetSnapshot>
3853 option or to ignore those volumes. B<IgnoreSelectedSnapshot> defaults to
3854 B<false>. However, if no B<GetSnapshot> option is specified at all, all
3855 capacities will be selected anyway.
3859 =head3 The Quota block
3861 This will collect (tree) quota statistics (used disk space and number of used
3862 files). This mechanism is useful to get usage information for single qtrees.
3863 In case the quotas are not used for any other purpose, an entry similar to the
3864 following in C</etc/quotas> would be sufficient:
3866 /vol/volA/some_qtree tree - - - - -
3868 After adding the entry, issue C<quota on -w volA> on the NetApp filer.
3872 =item B<Interval> I<Seconds>
3874 Collect SnapVault(R) statistics every I<Seconds> seconds.
3878 =head3 The SnapVault block
3880 This will collect statistics about the time and traffic of SnapVault(R)
3885 =item B<Interval> I<Seconds>
3887 Collect SnapVault(R) statistics every I<Seconds> seconds.
3891 =head2 Plugin C<netlink>
3893 The C<netlink> plugin uses a netlink socket to query the Linux kernel about
3894 statistics of various interface and routing aspects.
3898 =item B<Interface> I<Interface>
3900 =item B<VerboseInterface> I<Interface>
3902 Instruct the plugin to collect interface statistics. This is basically the same
3903 as the statistics provided by the C<interface> plugin (see above) but
3904 potentially much more detailed.
3906 When configuring with B<Interface> only the basic statistics will be collected,
3907 namely octets, packets, and errors. These statistics are collected by
3908 the C<interface> plugin, too, so using both at the same time is no benefit.
3910 When configured with B<VerboseInterface> all counters B<except> the basic ones,
3911 so that no data needs to be collected twice if you use the C<interface> plugin.
3912 This includes dropped packets, received multicast packets, collisions and a
3913 whole zoo of differentiated RX and TX errors. You can try the following command
3914 to get an idea of what awaits you:
3918 If I<Interface> is B<All>, all interfaces will be selected.
3920 =item B<QDisc> I<Interface> [I<QDisc>]
3922 =item B<Class> I<Interface> [I<Class>]
3924 =item B<Filter> I<Interface> [I<Filter>]
3926 Collect the octets and packets that pass a certain qdisc, class or filter.
3928 QDiscs and classes are identified by their type and handle (or classid).
3929 Filters don't necessarily have a handle, therefore the parent's handle is used.
3930 The notation used in collectd differs from that used in tc(1) in that it
3931 doesn't skip the major or minor number if it's zero and doesn't print special
3932 ids by their name. So, for example, a qdisc may be identified by
3933 C<pfifo_fast-1:0> even though the minor number of B<all> qdiscs is zero and
3934 thus not displayed by tc(1).
3936 If B<QDisc>, B<Class>, or B<Filter> is given without the second argument,
3937 i.E<nbsp>.e. without an identifier, all qdiscs, classes, or filters that are
3938 associated with that interface will be collected.
3940 Since a filter itself doesn't necessarily have a handle, the parent's handle is
3941 used. This may lead to problems when more than one filter is attached to a
3942 qdisc or class. This isn't nice, but we don't know how this could be done any
3943 better. If you have a idea, please don't hesitate to tell us.
3945 As with the B<Interface> option you can specify B<All> as the interface,
3946 meaning all interfaces.
3948 Here are some examples to help you understand the above text more easily:
3951 VerboseInterface "All"
3952 QDisc "eth0" "pfifo_fast-1:0"
3954 Class "ppp0" "htb-1:10"
3955 Filter "ppp0" "u32-1:0"
3958 =item B<IgnoreSelected>
3960 The behavior is the same as with all other similar plugins: If nothing is
3961 selected at all, everything is collected. If some things are selected using the
3962 options described above, only these statistics are collected. If you set
3963 B<IgnoreSelected> to B<true>, this behavior is inverted, i.E<nbsp>e. the
3964 specified statistics will not be collected.
3968 =head2 Plugin C<network>
3970 The Network plugin sends data to a remote instance of collectd, receives data
3971 from a remote instance, or both at the same time. Data which has been received
3972 from the network is usually not transmitted again, but this can be activated, see
3973 the B<Forward> option below.
3975 The default IPv6 multicast group is C<ff18::efc0:4a42>. The default IPv4
3976 multicast group is C<239.192.74.66>. The default I<UDP> port is B<25826>.
3978 Both, B<Server> and B<Listen> can be used as single option or as block. When
3979 used as block, given options are valid for this socket only. The following
3980 example will export the metrics twice: Once to an "internal" server (without
3981 encryption and signing) and one to an external server (with cryptographic
3985 # Export to an internal server
3986 # (demonstrates usage without additional options)
3987 Server "collectd.internal.tld"
3989 # Export to an external server
3990 # (demonstrates usage with signature options)
3991 <Server "collectd.external.tld">
3992 SecurityLevel "sign"
3993 Username "myhostname"
4000 =item B<E<lt>Server> I<Host> [I<Port>]B<E<gt>>
4002 The B<Server> statement/block sets the server to send datagrams to. The
4003 statement may occur multiple times to send each datagram to multiple
4006 The argument I<Host> may be a hostname, an IPv4 address or an IPv6 address. The
4007 optional second argument specifies a port number or a service name. If not
4008 given, the default, B<25826>, is used.
4010 The following options are recognized within B<Server> blocks:
4014 =item B<SecurityLevel> B<Encrypt>|B<Sign>|B<None>
4016 Set the security you require for network communication. When the security level
4017 has been set to B<Encrypt>, data sent over the network will be encrypted using
4018 I<AES-256>. The integrity of encrypted packets is ensured using I<SHA-1>. When
4019 set to B<Sign>, transmitted data is signed using the I<HMAC-SHA-256> message
4020 authentication code. When set to B<None>, data is sent without any security.
4022 This feature is only available if the I<network> plugin was linked with
4025 =item B<Username> I<Username>
4027 Sets the username to transmit. This is used by the server to lookup the
4028 password. See B<AuthFile> below. All security levels except B<None> require
4031 This feature is only available if the I<network> plugin was linked with
4034 =item B<Password> I<Password>
4036 Sets a password (shared secret) for this socket. All security levels except
4037 B<None> require this setting.
4039 This feature is only available if the I<network> plugin was linked with
4042 =item B<Interface> I<Interface name>
4044 Set the outgoing interface for IP packets. This applies at least
4045 to IPv6 packets and if possible to IPv4. If this option is not applicable,
4046 undefined or a non-existent interface name is specified, the default
4047 behavior is to let the kernel choose the appropriate interface. Be warned
4048 that the manual selection of an interface for unicast traffic is only
4049 necessary in rare cases.
4051 =item B<ResolveInterval> I<Seconds>
4053 Sets the interval at which to re-resolve the DNS for the I<Host>. This is
4054 useful to force a regular DNS lookup to support a high availability setup. If
4055 not specified, re-resolves are never attempted.
4059 =item B<E<lt>Listen> I<Host> [I<Port>]B<E<gt>>
4061 The B<Listen> statement sets the interfaces to bind to. When multiple
4062 statements are found the daemon will bind to multiple interfaces.
4064 The argument I<Host> may be a hostname, an IPv4 address or an IPv6 address. If
4065 the argument is a multicast address the daemon will join that multicast group.
4066 The optional second argument specifies a port number or a service name. If not
4067 given, the default, B<25826>, is used.
4069 The following options are recognized within C<E<lt>ListenE<gt>> blocks:
4073 =item B<SecurityLevel> B<Encrypt>|B<Sign>|B<None>
4075 Set the security you require for network communication. When the security level
4076 has been set to B<Encrypt>, only encrypted data will be accepted. The integrity
4077 of encrypted packets is ensured using I<SHA-1>. When set to B<Sign>, only
4078 signed and encrypted data is accepted. When set to B<None>, all data will be
4079 accepted. If an B<AuthFile> option was given (see below), encrypted data is
4080 decrypted if possible.
4082 This feature is only available if the I<network> plugin was linked with
4085 =item B<AuthFile> I<Filename>
4087 Sets a file in which usernames are mapped to passwords. These passwords are
4088 used to verify signatures and to decrypt encrypted network packets. If
4089 B<SecurityLevel> is set to B<None>, this is optional. If given, signed data is
4090 verified and encrypted packets are decrypted. Otherwise, signed data is
4091 accepted without checking the signature and encrypted data cannot be decrypted.
4092 For the other security levels this option is mandatory.
4094 The file format is very simple: Each line consists of a username followed by a
4095 colon and any number of spaces followed by the password. To demonstrate, an
4096 example file could look like this:
4101 Each time a packet is received, the modification time of the file is checked
4102 using L<stat(2)>. If the file has been changed, the contents is re-read. While
4103 the file is being read, it is locked using L<fcntl(2)>.
4105 =item B<Interface> I<Interface name>
4107 Set the incoming interface for IP packets explicitly. This applies at least
4108 to IPv6 packets and if possible to IPv4. If this option is not applicable,
4109 undefined or a non-existent interface name is specified, the default
4110 behavior is, to let the kernel choose the appropriate interface. Thus incoming
4111 traffic gets only accepted, if it arrives on the given interface.
4115 =item B<TimeToLive> I<1-255>
4117 Set the time-to-live of sent packets. This applies to all, unicast and
4118 multicast, and IPv4 and IPv6 packets. The default is to not change this value.
4119 That means that multicast packets will be sent with a TTL of C<1> (one) on most
4122 =item B<MaxPacketSize> I<1024-65535>
4124 Set the maximum size for datagrams received over the network. Packets larger
4125 than this will be truncated. Defaults to 1452E<nbsp>bytes, which is the maximum
4126 payload size that can be transmitted in one Ethernet frame using IPv6E<nbsp>/
4129 On the server side, this limit should be set to the largest value used on
4130 I<any> client. Likewise, the value on the client must not be larger than the
4131 value on the server, or data will be lost.
4133 B<Compatibility:> Versions prior to I<versionE<nbsp>4.8> used a fixed sized
4134 buffer of 1024E<nbsp>bytes. Versions I<4.8>, I<4.9> and I<4.10> used a default
4135 value of 1024E<nbsp>bytes to avoid problems when sending data to an older
4138 =item B<Forward> I<true|false>
4140 If set to I<true>, write packets that were received via the network plugin to
4141 the sending sockets. This should only be activated when the B<Listen>- and
4142 B<Server>-statements differ. Otherwise packets may be send multiple times to
4143 the same multicast group. While this results in more network traffic than
4144 necessary it's not a huge problem since the plugin has a duplicate detection,
4145 so the values will not loop.
4147 =item B<ReportStats> B<true>|B<false>
4149 The network plugin cannot only receive and send statistics, it can also create
4150 statistics about itself. Collected data included the number of received and
4151 sent octets and packets, the length of the receive queue and the number of
4152 values handled. When set to B<true>, the I<Network plugin> will make these
4153 statistics available. Defaults to B<false>.
4157 =head2 Plugin C<nginx>
4159 This plugin collects the number of connections and requests handled by the
4160 C<nginx daemon> (speak: engineE<nbsp>X), a HTTP and mail server/proxy. It
4161 queries the page provided by the C<ngx_http_stub_status_module> module, which
4162 isn't compiled by default. Please refer to
4163 L<http://wiki.codemongers.com/NginxStubStatusModule> for more information on
4164 how to compile and configure nginx and this module.
4166 The following options are accepted by the C<nginx plugin>:
4170 =item B<URL> I<http://host/nginx_status>
4172 Sets the URL of the C<ngx_http_stub_status_module> output.
4174 =item B<User> I<Username>
4176 Optional user name needed for authentication.
4178 =item B<Password> I<Password>
4180 Optional password needed for authentication.
4182 =item B<VerifyPeer> B<true|false>
4184 Enable or disable peer SSL certificate verification. See
4185 L<http://curl.haxx.se/docs/sslcerts.html> for details. Enabled by default.
4187 =item B<VerifyHost> B<true|false>
4189 Enable or disable peer host name verification. If enabled, the plugin checks
4190 if the C<Common Name> or a C<Subject Alternate Name> field of the SSL
4191 certificate matches the host name provided by the B<URL> option. If this
4192 identity check fails, the connection is aborted. Obviously, only works when
4193 connecting to a SSL enabled server. Enabled by default.
4195 =item B<CACert> I<File>
4197 File that holds one or more SSL certificates. If you want to use HTTPS you will
4198 possibly need this option. What CA certificates come bundled with C<libcurl>
4199 and are checked by default depends on the distribution you use.
4201 =item B<Timeout> I<Milliseconds>
4203 The B<Timeout> option sets the overall timeout for HTTP requests to B<URL>, in
4204 milliseconds. By default, the configured B<Interval> is used to set the
4209 =head2 Plugin C<notify_desktop>
4211 This plugin sends a desktop notification to a notification daemon, as defined
4212 in the Desktop Notification Specification. To actually display the
4213 notifications, B<notification-daemon> is required and B<collectd> has to be
4214 able to access the X server (i.E<nbsp>e., the C<DISPLAY> and C<XAUTHORITY>
4215 environment variables have to be set correctly) and the D-Bus message bus.
4217 The Desktop Notification Specification can be found at
4218 L<http://www.galago-project.org/specs/notification/>.
4222 =item B<OkayTimeout> I<timeout>
4224 =item B<WarningTimeout> I<timeout>
4226 =item B<FailureTimeout> I<timeout>
4228 Set the I<timeout>, in milliseconds, after which to expire the notification
4229 for C<OKAY>, C<WARNING> and C<FAILURE> severities respectively. If zero has
4230 been specified, the displayed notification will not be closed at all - the
4231 user has to do so herself. These options default to 5000. If a negative number
4232 has been specified, the default is used as well.
4236 =head2 Plugin C<notify_email>
4238 The I<notify_email> plugin uses the I<ESMTP> library to send notifications to a
4239 configured email address.
4241 I<libESMTP> is available from L<http://www.stafford.uklinux.net/libesmtp/>.
4243 Available configuration options:
4247 =item B<From> I<Address>
4249 Email address from which the emails should appear to come from.
4251 Default: C<root@localhost>
4253 =item B<Recipient> I<Address>
4255 Configures the email address(es) to which the notifications should be mailed.
4256 May be repeated to send notifications to multiple addresses.
4258 At least one B<Recipient> must be present for the plugin to work correctly.
4260 =item B<SMTPServer> I<Hostname>
4262 Hostname of the SMTP server to connect to.
4264 Default: C<localhost>
4266 =item B<SMTPPort> I<Port>
4268 TCP port to connect to.
4272 =item B<SMTPUser> I<Username>
4274 Username for ASMTP authentication. Optional.
4276 =item B<SMTPPassword> I<Password>
4278 Password for ASMTP authentication. Optional.
4280 =item B<Subject> I<Subject>
4282 Subject-template to use when sending emails. There must be exactly two
4283 string-placeholders in the subject, given in the standard I<printf(3)> syntax,
4284 i.E<nbsp>e. C<%s>. The first will be replaced with the severity, the second
4287 Default: C<Collectd notify: %s@%s>
4291 =head2 Plugin C<ntpd>
4295 =item B<Host> I<Hostname>
4297 Hostname of the host running B<ntpd>. Defaults to B<localhost>.
4299 =item B<Port> I<Port>
4301 UDP-Port to connect to. Defaults to B<123>.
4303 =item B<ReverseLookups> B<true>|B<false>
4305 Sets whether or not to perform reverse lookups on peers. Since the name or
4306 IP-address may be used in a filename it is recommended to disable reverse
4307 lookups. The default is to do reverse lookups to preserve backwards
4308 compatibility, though.
4310 =item B<IncludeUnitID> B<true>|B<false>
4312 When a peer is a refclock, include the unit ID in the I<type instance>.
4313 Defaults to B<false> for backward compatibility.
4315 If two refclock peers use the same driver and this is B<false>, the plugin will
4316 try to write simultaneous measurements from both to the same type instance.
4317 This will result in error messages in the log and only one set of measurements
4322 =head2 Plugin C<nut>
4326 =item B<UPS> I<upsname>B<@>I<hostname>[B<:>I<port>]
4328 Add a UPS to collect data from. The format is identical to the one accepted by
4333 =head2 Plugin C<olsrd>
4335 The I<olsrd> plugin connects to the TCP port opened by the I<txtinfo> plugin of
4336 the Optimized Link State Routing daemon and reads information about the current
4337 state of the meshed network.
4339 The following configuration options are understood:
4343 =item B<Host> I<Host>
4345 Connect to I<Host>. Defaults to B<"localhost">.
4347 =item B<Port> I<Port>
4349 Specifies the port to connect to. This must be a string, even if you give the
4350 port as a number rather than a service name. Defaults to B<"2006">.
4352 =item B<CollectLinks> B<No>|B<Summary>|B<Detail>
4354 Specifies what information to collect about links, i.E<nbsp>e. direct
4355 connections of the daemon queried. If set to B<No>, no information is
4356 collected. If set to B<Summary>, the number of links and the average of all
4357 I<link quality> (LQ) and I<neighbor link quality> (NLQ) values is calculated.
4358 If set to B<Detail> LQ and NLQ are collected per link.
4360 Defaults to B<Detail>.
4362 =item B<CollectRoutes> B<No>|B<Summary>|B<Detail>
4364 Specifies what information to collect about routes of the daemon queried. If
4365 set to B<No>, no information is collected. If set to B<Summary>, the number of
4366 routes and the average I<metric> and I<ETX> is calculated. If set to B<Detail>
4367 metric and ETX are collected per route.
4369 Defaults to B<Summary>.
4371 =item B<CollectTopology> B<No>|B<Summary>|B<Detail>
4373 Specifies what information to collect about the global topology. If set to
4374 B<No>, no information is collected. If set to B<Summary>, the number of links
4375 in the entire topology and the average I<link quality> (LQ) is calculated.
4376 If set to B<Detail> LQ and NLQ are collected for each link in the entire topology.
4378 Defaults to B<Summary>.
4382 =head2 Plugin C<onewire>
4384 B<EXPERIMENTAL!> See notes below.
4386 The C<onewire> plugin uses the B<owcapi> library from the B<owfs> project
4387 L<http://owfs.org/> to read sensors connected via the onewire bus.
4389 It can be used in two possible modes - standard or advanced.
4391 In the standard mode only temperature sensors (sensors with the family code
4392 C<10>, C<22> and C<28> - e.g. DS1820, DS18S20, DS1920) can be read. If you have
4393 other sensors you would like to have included, please send a sort request to
4394 the mailing list. You can select sensors to be read or to be ignored depending
4395 on the option B<IgnoreSelected>). When no list is provided the whole bus is
4396 walked and all sensors are read.
4398 Hubs (the DS2409 chips) are working, but read the note, why this plugin is
4399 experimental, below.
4401 In the advanced mode you can configure any sensor to be read (only numerical
4402 value) using full OWFS path (e.g. "/uncached/10.F10FCA000800/temperature").
4403 In this mode you have to list all the sensors. Neither default bus walk nor
4404 B<IgnoreSelected> are used here. Address and type (file) is extracted from
4405 the path automatically and should produce compatible structure with the "standard"
4406 mode (basically the path is expected as for example
4407 "/uncached/10.F10FCA000800/temperature" where it would extract address part
4408 "F10FCA000800" and the rest after the slash is considered the type - here
4410 There are two advantages to this mode - you can access virtually any sensor
4411 (not just temperature), select whether to use cached or directly read values
4412 and it is slighlty faster. The downside is more complex configuration.
4414 The two modes are distinguished automatically by the format of the address.
4415 It is not possible to mix the two modes. Once a full path is detected in any
4416 B<Sensor> then the whole addressing (all sensors) is considered to be this way
4417 (and as standard addresses will fail parsing they will be ignored).
4421 =item B<Device> I<Device>
4423 Sets the device to read the values from. This can either be a "real" hardware
4424 device, such as a serial port or an USB port, or the address of the
4425 L<owserver(1)> socket, usually B<localhost:4304>.
4427 Though the documentation claims to automatically recognize the given address
4428 format, with versionE<nbsp>2.7p4 we had to specify the type explicitly. So
4429 with that version, the following configuration worked for us:
4432 Device "-s localhost:4304"
4435 This directive is B<required> and does not have a default value.
4437 =item B<Sensor> I<Sensor>
4439 In the standard mode selects sensors to collect or to ignore
4440 (depending on B<IgnoreSelected>, see below). Sensors are specified without
4441 the family byte at the beginning, so you have to use for example C<F10FCA000800>,
4442 and B<not> include the leading C<10.> family byte and point.
4443 When no B<Sensor> is configured the whole Onewire bus is walked and all supported
4444 sensors (see above) are read.
4446 In the advanced mode the B<Sensor> specifies full OWFS path - e.g.
4447 C</uncached/10.F10FCA000800/temperature> (or when cached values are OK
4448 C</10.F10FCA000800/temperature>). B<IgnoreSelected> is not used.
4450 As there can be multiple devices on the bus you can list multiple sensor (use
4451 multiple B<Sensor> elements).
4453 =item B<IgnoreSelected> I<true>|I<false>
4455 If no configuration is given, the B<onewire> plugin will collect data from all
4456 sensors found. This may not be practical, especially if sensors are added and
4457 removed regularly. Sometimes, however, it's easier/preferred to collect only
4458 specific sensors or all sensors I<except> a few specified ones. This option
4459 enables you to do that: By setting B<IgnoreSelected> to I<true> the effect of
4460 B<Sensor> is inverted: All selected interfaces are ignored and all other
4461 interfaces are collected.
4463 Used only in the standard mode - see above.
4465 =item B<Interval> I<Seconds>
4467 Sets the interval in which all sensors should be read. If not specified, the
4468 global B<Interval> setting is used.
4472 B<EXPERIMENTAL!> The C<onewire> plugin is experimental, because it doesn't yet
4473 work with big setups. It works with one sensor being attached to one
4474 controller, but as soon as you throw in a couple more senors and maybe a hub
4475 or two, reading all values will take more than ten seconds (the default
4476 interval). We will probably add some separate thread for reading the sensors
4477 and some cache or something like that, but it's not done yet. We will try to
4478 maintain backwards compatibility in the future, but we can't promise. So in
4479 short: If it works for you: Great! But keep in mind that the config I<might>
4480 change, though this is unlikely. Oh, and if you want to help improving this
4481 plugin, just send a short notice to the mailing list. ThanksE<nbsp>:)
4483 =head2 Plugin C<openldap>
4485 To use the C<openldap> plugin you first need to configure the I<OpenLDAP>
4486 server correctly. The backend database C<monitor> needs to be loaded and
4487 working. See slapd-monitor(5) for the details.
4489 The configuration of the C<openldap> plugin consists of one or more B<Instance>
4490 blocks. Each block requires one string argument as the instance name. For
4495 URL "ldap://localhost/"
4498 URL "ldaps://localhost/"
4502 The instance name will be used as the I<plugin instance>. To emulate the old
4503 (versionE<nbsp>4) behavior, you can use an empty string (""). In order for the
4504 plugin to work correctly, each instance name must be unique. This is not
4505 enforced by the plugin and it is your responsibility to ensure it is.
4507 The following options are accepted within each B<Instance> block:
4511 =item B<URL> I<ldap://host/binddn>
4513 Sets the URL to use to connect to the I<OpenLDAP> server. This option is
4516 =item B<BindDN> I<BindDN>
4518 Name in the form of an LDAP distinguished name intended to be used for
4519 authentication. Defaults to empty string to establish an anonymous authorization.
4521 =item B<Password> I<Password>
4523 Password for simple bind authentication. If this option is not set,
4524 unauthenticated bind operation is used.
4526 =item B<StartTLS> B<true|false>
4528 Defines whether TLS must be used when connecting to the I<OpenLDAP> server.
4529 Disabled by default.
4531 =item B<VerifyHost> B<true|false>
4533 Enables or disables peer host name verification. If enabled, the plugin checks
4534 if the C<Common Name> or a C<Subject Alternate Name> field of the SSL
4535 certificate matches the host name provided by the B<URL> option. If this
4536 identity check fails, the connection is aborted. Enabled by default.
4538 =item B<CACert> I<File>
4540 File that holds one or more SSL certificates. If you want to use TLS/SSL you
4541 may possibly need this option. What CA certificates are checked by default
4542 depends on the distribution you use and can be changed with the usual ldap
4543 client configuration mechanisms. See ldap.conf(5) for the details.
4545 =item B<Timeout> I<Seconds>
4547 Sets the timeout value for ldap operations. Defaults to B<-1> which results in
4548 an infinite timeout.
4550 =item B<Version> I<Version>
4552 An integer which sets the LDAP protocol version number to use when connecting
4553 to the I<OpenLDAP> server. Defaults to B<3> for using I<LDAPv3>.
4557 =head2 Plugin C<openvpn>
4559 The OpenVPN plugin reads a status file maintained by OpenVPN and gathers
4560 traffic statistics about connected clients.
4562 To set up OpenVPN to write to the status file periodically, use the
4563 B<--status> option of OpenVPN. Since OpenVPN can write two different formats,
4564 you need to set the required format, too. This is done by setting
4565 B<--status-version> to B<2>.
4567 So, in a nutshell you need:
4569 openvpn $OTHER_OPTIONS \
4570 --status "/var/run/openvpn-status" 10 \
4577 =item B<StatusFile> I<File>
4579 Specifies the location of the status file.
4581 =item B<ImprovedNamingSchema> B<true>|B<false>
4583 When enabled, the filename of the status file will be used as plugin instance
4584 and the client's "common name" will be used as type instance. This is required
4585 when reading multiple status files. Enabling this option is recommended, but to
4586 maintain backwards compatibility this option is disabled by default.
4588 =item B<CollectCompression> B<true>|B<false>
4590 Sets whether or not statistics about the compression used by OpenVPN should be
4591 collected. This information is only available in I<single> mode. Enabled by
4594 =item B<CollectIndividualUsers> B<true>|B<false>
4596 Sets whether or not traffic information is collected for each connected client
4597 individually. If set to false, currently no traffic data is collected at all
4598 because aggregating this data in a save manner is tricky. Defaults to B<true>.
4600 =item B<CollectUserCount> B<true>|B<false>
4602 When enabled, the number of currently connected clients or users is collected.
4603 This is especially interesting when B<CollectIndividualUsers> is disabled, but
4604 can be configured independently from that option. Defaults to B<false>.
4608 =head2 Plugin C<oracle>
4610 The "oracle" plugin uses the Oracle® Call Interface I<(OCI)> to connect to an
4611 Oracle® Database and lets you execute SQL statements there. It is very similar
4612 to the "dbi" plugin, because it was written around the same time. See the "dbi"
4613 plugin's documentation above for details.
4616 <Query "out_of_stock">
4617 Statement "SELECT category, COUNT(*) AS value FROM products WHERE in_stock = 0 GROUP BY category"
4620 # InstancePrefix "foo"
4621 InstancesFrom "category"
4625 <Database "product_information">
4629 Query "out_of_stock"
4633 =head3 B<Query> blocks
4635 The Query blocks are handled identically to the Query blocks of the "dbi"
4636 plugin. Please see its documentation above for details on how to specify
4639 =head3 B<Database> blocks
4641 Database blocks define a connection to a database and which queries should be
4642 sent to that database. Each database needs a "name" as string argument in the
4643 starting tag of the block. This name will be used as "PluginInstance" in the
4644 values submitted to the daemon. Other than that, that name is not used.
4648 =item B<ConnectID> I<ID>
4650 Defines the "database alias" or "service name" to connect to. Usually, these
4651 names are defined in the file named C<$ORACLE_HOME/network/admin/tnsnames.ora>.
4653 =item B<Host> I<Host>
4655 Hostname to use when dispatching values for this database. Defaults to using
4656 the global hostname of the I<collectd> instance.
4658 =item B<Username> I<Username>
4660 Username used for authentication.
4662 =item B<Password> I<Password>
4664 Password used for authentication.
4666 =item B<Query> I<QueryName>
4668 Associates the query named I<QueryName> with this database connection. The
4669 query needs to be defined I<before> this statement, i.E<nbsp>e. all query
4670 blocks you want to refer to must be placed above the database block you want to
4675 =head2 Plugin C<perl>
4677 This plugin embeds a Perl-interpreter into collectd and provides an interface
4678 to collectd's plugin system. See L<collectd-perl(5)> for its documentation.
4680 =head2 Plugin C<pinba>
4682 The I<Pinba plugin> receives profiling information from I<Pinba>, an extension
4683 for the I<PHP> interpreter. At the end of executing a script, i.e. after a
4684 PHP-based webpage has been delivered, the extension will send a UDP packet
4685 containing timing information, peak memory usage and so on. The plugin will
4686 wait for such packets, parse them and account the provided information, which
4687 is then dispatched to the daemon once per interval.
4694 # Overall statistics for the website.
4696 Server "www.example.com"
4698 # Statistics for www-a only
4700 Host "www-a.example.com"
4701 Server "www.example.com"
4703 # Statistics for www-b only
4705 Host "www-b.example.com"
4706 Server "www.example.com"
4710 The plugin provides the following configuration options:
4714 =item B<Address> I<Node>
4716 Configures the address used to open a listening socket. By default, plugin will
4717 bind to the I<any> address C<::0>.
4719 =item B<Port> I<Service>
4721 Configures the port (service) to bind to. By default the default Pinba port
4722 "30002" will be used. The option accepts service names in addition to port
4723 numbers and thus requires a I<string> argument.
4725 =item E<lt>B<View> I<Name>E<gt> block
4727 The packets sent by the Pinba extension include the hostname of the server, the
4728 server name (the name of the virtual host) and the script that was executed.
4729 Using B<View> blocks it is possible to separate the data into multiple groups
4730 to get more meaningful statistics. Each packet is added to all matching groups,
4731 so that a packet may be accounted for more than once.
4735 =item B<Host> I<Host>
4737 Matches the hostname of the system the webserver / script is running on. This
4738 will contain the result of the L<gethostname(2)> system call. If not
4739 configured, all hostnames will be accepted.
4741 =item B<Server> I<Server>
4743 Matches the name of the I<virtual host>, i.e. the contents of the
4744 C<$_SERVER["SERVER_NAME"]> variable when within PHP. If not configured, all
4745 server names will be accepted.
4747 =item B<Script> I<Script>
4749 Matches the name of the I<script name>, i.e. the contents of the
4750 C<$_SERVER["SCRIPT_NAME"]> variable when within PHP. If not configured, all
4751 script names will be accepted.
4757 =head2 Plugin C<ping>
4759 The I<Ping> plugin starts a new thread which sends ICMP "ping" packets to the
4760 configured hosts periodically and measures the network latency. Whenever the
4761 C<read> function of the plugin is called, it submits the average latency, the
4762 standard deviation and the drop rate for each host.
4764 Available configuration options:
4768 =item B<Host> I<IP-address>
4770 Host to ping periodically. This option may be repeated several times to ping
4773 =item B<Interval> I<Seconds>
4775 Sets the interval in which to send ICMP echo packets to the configured hosts.
4776 This is B<not> the interval in which statistics are queries from the plugin but
4777 the interval in which the hosts are "pinged". Therefore, the setting here
4778 should be smaller than or equal to the global B<Interval> setting. Fractional
4779 times, such as "1.24" are allowed.
4783 =item B<Timeout> I<Seconds>
4785 Time to wait for a response from the host to which an ICMP packet had been
4786 sent. If a reply was not received after I<Seconds> seconds, the host is assumed
4787 to be down or the packet to be dropped. This setting must be smaller than the
4788 B<Interval> setting above for the plugin to work correctly. Fractional
4789 arguments are accepted.
4793 =item B<TTL> I<0-255>
4795 Sets the Time-To-Live of generated ICMP packets.
4797 =item B<SourceAddress> I<host>
4799 Sets the source address to use. I<host> may either be a numerical network
4800 address or a network hostname.
4802 =item B<Device> I<name>
4804 Sets the outgoing network device to be used. I<name> has to specify an
4805 interface name (e.E<nbsp>g. C<eth0>). This might not be supported by all
4808 =item B<MaxMissed> I<Packets>
4810 Trigger a DNS resolve after the host has not replied to I<Packets> packets. This
4811 enables the use of dynamic DNS services (like dyndns.org) with the ping plugin.
4813 Default: B<-1> (disabled)
4817 =head2 Plugin C<postgresql>
4819 The C<postgresql> plugin queries statistics from PostgreSQL databases. It
4820 keeps a persistent connection to all configured databases and tries to
4821 reconnect if the connection has been interrupted. A database is configured by
4822 specifying a B<Database> block as described below. The default statistics are
4823 collected from PostgreSQL's B<statistics collector> which thus has to be
4824 enabled for this plugin to work correctly. This should usually be the case by
4825 default. See the section "The Statistics Collector" of the B<PostgreSQL
4826 Documentation> for details.
4828 By specifying custom database queries using a B<Query> block as described
4829 below, you may collect any data that is available from some PostgreSQL
4830 database. This way, you are able to access statistics of external daemons
4831 which are available in a PostgreSQL database or use future or special
4832 statistics provided by PostgreSQL without the need to upgrade your collectd
4835 Starting with version 5.2, the C<postgresql> plugin supports writing data to
4836 PostgreSQL databases as well. This has been implemented in a generic way. You
4837 need to specify an SQL statement which will then be executed by collectd in
4838 order to write the data (see below for details). The benefit of that approach
4839 is that there is no fixed database layout. Rather, the layout may be optimized
4840 for the current setup.
4842 The B<PostgreSQL Documentation> manual can be found at
4843 L<http://www.postgresql.org/docs/manuals/>.
4847 Statement "SELECT magic FROM wizard WHERE host = $1;"
4851 InstancePrefix "magic"
4856 <Query rt36_tickets>
4857 Statement "SELECT COUNT(type) AS count, type \
4859 WHEN resolved = 'epoch' THEN 'open' \
4860 ELSE 'resolved' END AS type \
4861 FROM tickets) type \
4865 InstancePrefix "rt36_tickets"
4866 InstancesFrom "type"
4872 Statement "SELECT collectd_insert($1, $2, $3, $4, $5, $6, $7, $8, $9);"
4882 KRBSrvName "kerberos_service_name"
4888 Service "service_name"
4889 Query backend # predefined
4900 The B<Query> block defines one database query which may later be used by a
4901 database definition. It accepts a single mandatory argument which specifies
4902 the name of the query. The names of all queries have to be unique (see the
4903 B<MinVersion> and B<MaxVersion> options below for an exception to this
4904 rule). The following configuration options are available to define the query:
4906 In each B<Query> block, there is one or more B<Result> blocks. B<Result>
4907 blocks define how to handle the values returned from the query. They define
4908 which column holds which value and how to dispatch that value to the daemon.
4909 Multiple B<Result> blocks may be used to extract multiple values from a single
4914 =item B<Statement> I<sql query statement>
4916 Specify the I<sql query statement> which the plugin should execute. The string
4917 may contain the tokens B<$1>, B<$2>, etc. which are used to reference the
4918 first, second, etc. parameter. The value of the parameters is specified by the
4919 B<Param> configuration option - see below for details. To include a literal
4920 B<$> character followed by a number, surround it with single quotes (B<'>).
4922 Any SQL command which may return data (such as C<SELECT> or C<SHOW>) is
4923 allowed. Note, however, that only a single command may be used. Semicolons are
4924 allowed as long as a single non-empty command has been specified only.
4926 The returned lines will be handled separately one after another.
4928 =item B<Param> I<hostname>|I<database>|I<username>|I<interval>
4930 Specify the parameters which should be passed to the SQL query. The parameters
4931 are referred to in the SQL query as B<$1>, B<$2>, etc. in the same order as
4932 they appear in the configuration file. The value of the parameter is
4933 determined depending on the value of the B<Param> option as follows:
4939 The configured hostname of the database connection. If a UNIX domain socket is
4940 used, the parameter expands to "localhost".
4944 The name of the database of the current connection.
4948 The name of the database plugin instance. See the B<Instance> option of the
4949 database specification below for details.
4953 The username used to connect to the database.
4957 The interval with which this database is queried (as specified by the database
4958 specific or global B<Interval> options).
4962 Please note that parameters are only supported by PostgreSQL's protocol
4963 version 3 and above which was introduced in version 7.4 of PostgreSQL.
4965 =item B<Type> I<type>
4967 The I<type> name to be used when dispatching the values. The type describes
4968 how to handle the data and where to store it. See L<types.db(5)> for more
4969 details on types and their configuration. The number and type of values (as
4970 selected by the B<ValuesFrom> option) has to match the type of the given name.
4972 This option is required inside a B<Result> block.
4974 =item B<InstancePrefix> I<prefix>
4976 =item B<InstancesFrom> I<column0> [I<column1> ...]
4978 Specify how to create the "TypeInstance" for each data set (i.E<nbsp>e. line).
4979 B<InstancePrefix> defines a static prefix that will be prepended to all type
4980 instances. B<InstancesFrom> defines the column names whose values will be used
4981 to create the type instance. Multiple values will be joined together using the
4982 hyphen (C<->) as separation character.
4984 The plugin itself does not check whether or not all built instances are
4985 different. It is your responsibility to assure that each is unique.
4987 Both options are optional. If none is specified, the type instance will be
4990 =item B<ValuesFrom> I<column0> [I<column1> ...]
4992 Names the columns whose content is used as the actual data for the data sets
4993 that are dispatched to the daemon. How many such columns you need is
4994 determined by the B<Type> setting as explained above. If you specify too many
4995 or not enough columns, the plugin will complain about that and no data will be
4996 submitted to the daemon.
4998 The actual data type, as seen by PostgreSQL, is not that important as long as
4999 it represents numbers. The plugin will automatically cast the values to the
5000 right type if it know how to do that. For that, it uses the L<strtoll(3)> and
5001 L<strtod(3)> functions, so anything supported by those functions is supported
5002 by the plugin as well.
5004 This option is required inside a B<Result> block and may be specified multiple
5005 times. If multiple B<ValuesFrom> options are specified, the columns are read
5008 =item B<MinVersion> I<version>
5010 =item B<MaxVersion> I<version>
5012 Specify the minimum or maximum version of PostgreSQL that this query should be
5013 used with. Some statistics might only be available with certain versions of
5014 PostgreSQL. This allows you to specify multiple queries with the same name but
5015 which apply to different versions, thus allowing you to use the same
5016 configuration in a heterogeneous environment.
5018 The I<version> has to be specified as the concatenation of the major, minor
5019 and patch-level versions, each represented as two-decimal-digit numbers. For
5020 example, version 8.2.3 will become 80203.
5024 The following predefined queries are available (the definitions can be found
5025 in the F<postgresql_default.conf> file which, by default, is available at
5026 C<I<prefix>/share/collectd/>):
5032 This query collects the number of backends, i.E<nbsp>e. the number of
5035 =item B<transactions>
5037 This query collects the numbers of committed and rolled-back transactions of
5042 This query collects the numbers of various table modifications (i.E<nbsp>e.
5043 insertions, updates, deletions) of the user tables.
5045 =item B<query_plans>
5047 This query collects the numbers of various table scans and returned tuples of
5050 =item B<table_states>
5052 This query collects the numbers of live and dead rows in the user tables.
5056 This query collects disk block access counts for user tables.
5060 This query collects the on-disk size of the database in bytes.
5064 In addition, the following detailed queries are available by default. Please
5065 note that each of those queries collects information B<by table>, thus,
5066 potentially producing B<a lot> of data. For details see the description of the
5067 non-by_table queries above.
5071 =item B<queries_by_table>
5073 =item B<query_plans_by_table>
5075 =item B<table_states_by_table>
5077 =item B<disk_io_by_table>
5081 The B<Writer> block defines a PostgreSQL writer backend. It accepts a single
5082 mandatory argument specifying the name of the writer. This will then be used
5083 in the B<Database> specification in order to activate the writer instance. The
5084 names of all writers have to be unique. The following options may be
5089 =item B<Statement> I<sql statement>
5091 This mandatory option specifies the SQL statement that will be executed for
5092 each submitted value. A single SQL statement is allowed only. Anything after
5093 the first semicolon will be ignored.
5095 Nine parameters will be passed to the statement and should be specified as
5096 tokens B<$1>, B<$2>, through B<$9> in the statement string. The following
5097 values are made available through those parameters:
5103 The timestamp of the queried value as a floating point number.
5107 The hostname of the queried value.
5111 The plugin name of the queried value.
5115 The plugin instance of the queried value. This value may be B<NULL> if there
5116 is no plugin instance.
5120 The type of the queried value (cf. L<types.db(5)>).
5124 The type instance of the queried value. This value may be B<NULL> if there is
5129 An array of names for the submitted values (i.E<nbsp>e., the name of the data
5130 sources of the submitted value-list).
5134 An array of types for the submitted values (i.E<nbsp>e., the type of the data
5135 sources of the submitted value-list; C<counter>, C<gauge>, ...). Note, that if
5136 B<StoreRates> is enabled (which is the default, see below), all types will be
5141 An array of the submitted values. The dimensions of the value name and value
5146 In general, it is advisable to create and call a custom function in the
5147 PostgreSQL database for this purpose. Any procedural language supported by
5148 PostgreSQL will do (see chapter "Server Programming" in the PostgreSQL manual
5151 =item B<StoreRates> B<false>|B<true>
5153 If set to B<true> (the default), convert counter values to rates. If set to
5154 B<false> counter values are stored as is, i.E<nbsp>e. as an increasing integer
5159 The B<Database> block defines one PostgreSQL database for which to collect
5160 statistics. It accepts a single mandatory argument which specifies the
5161 database name. None of the other options are required. PostgreSQL will use
5162 default values as documented in the section "CONNECTING TO A DATABASE" in the
5163 L<psql(1)> manpage. However, be aware that those defaults may be influenced by
5164 the user collectd is run as and special environment variables. See the manpage
5169 =item B<Interval> I<seconds>
5171 Specify the interval with which the database should be queried. The default is
5172 to use the global B<Interval> setting.
5174 =item B<CommitInterval> I<seconds>
5176 This option may be used for database connections which have "writers" assigned
5177 (see above). If specified, it causes a writer to put several updates into a
5178 single transaction. This transaction will last for the specified amount of
5179 time. By default, each update will be executed in a separate transaction. Each
5180 transaction generates a fair amount of overhead which can, thus, be reduced by
5181 activating this option. The draw-back is, that data covering the specified
5182 amount of time will be lost, for example, if a single statement within the
5183 transaction fails or if the database server crashes.
5185 =item B<Instance> I<name>
5187 Specify the plugin instance name that should be used instead of the database
5188 name (which is the default, if this option has not been specified). This
5189 allows to query multiple databases of the same name on the same host (e.g.
5190 when running multiple database server versions in parallel).
5192 =item B<Host> I<hostname>
5194 Specify the hostname or IP of the PostgreSQL server to connect to. If the
5195 value begins with a slash, it is interpreted as the directory name in which to
5196 look for the UNIX domain socket.
5198 This option is also used to determine the hostname that is associated with a
5199 collected data set. If it has been omitted or either begins with with a slash
5200 or equals B<localhost> it will be replaced with the global hostname definition
5201 of collectd. Any other value will be passed literally to collectd when
5202 dispatching values. Also see the global B<Hostname> and B<FQDNLookup> options.
5204 =item B<Port> I<port>
5206 Specify the TCP port or the local UNIX domain socket file extension of the
5209 =item B<User> I<username>
5211 Specify the username to be used when connecting to the server.
5213 =item B<Password> I<password>
5215 Specify the password to be used when connecting to the server.
5217 =item B<ExpireDelay> I<delay>
5219 Skip expired values in query output.
5221 =item B<SSLMode> I<disable>|I<allow>|I<prefer>|I<require>
5223 Specify whether to use an SSL connection when contacting the server. The
5224 following modes are supported:
5230 Do not use SSL at all.
5234 First, try to connect without using SSL. If that fails, try using SSL.
5236 =item I<prefer> (default)
5238 First, try to connect using SSL. If that fails, try without using SSL.
5246 =item B<Instance> I<name>
5248 Specify the plugin instance name that should be used instead of the database
5249 name (which is the default, if this option has not been specified). This
5250 allows to query multiple databases of the same name on the same host (e.g.
5251 when running multiple database server versions in parallel).
5253 =item B<KRBSrvName> I<kerberos_service_name>
5255 Specify the Kerberos service name to use when authenticating with Kerberos 5
5256 or GSSAPI. See the sections "Kerberos authentication" and "GSSAPI" of the
5257 B<PostgreSQL Documentation> for details.
5259 =item B<Service> I<service_name>
5261 Specify the PostgreSQL service name to use for additional parameters. That
5262 service has to be defined in F<pg_service.conf> and holds additional
5263 connection parameters. See the section "The Connection Service File" in the
5264 B<PostgreSQL Documentation> for details.
5266 =item B<Query> I<query>
5268 Specifies a I<query> which should be executed in the context of the database
5269 connection. This may be any of the predefined or user-defined queries. If no
5270 such option is given, it defaults to "backends", "transactions", "queries",
5271 "query_plans", "table_states", "disk_io" and "disk_usage" (unless a B<Writer>
5272 has been specified). Else, the specified queries are used only.
5274 =item B<Writer> I<writer>
5276 Assigns the specified I<writer> backend to the database connection. This
5277 causes all collected data to be send to the database using the settings
5278 defined in the writer configuration (see the section "FILTER CONFIGURATION"
5279 below for details on how to selectively send data to certain plugins).
5281 Each writer will register a flush callback which may be used when having long
5282 transactions enabled (see the B<CommitInterval> option above). When issuing
5283 the B<FLUSH> command (see L<collectd-unixsock(5)> for details) the current
5284 transaction will be committed right away. Two different kinds of flush
5285 callbacks are available with the C<postgresql> plugin:
5291 Flush all writer backends.
5293 =item B<postgresql->I<database>
5295 Flush all writers of the specified I<database> only.
5301 =head2 Plugin C<powerdns>
5303 The C<powerdns> plugin queries statistics from an authoritative PowerDNS
5304 nameserver and/or a PowerDNS recursor. Since both offer a wide variety of
5305 values, many of which are probably meaningless to most users, but may be useful
5306 for some. So you may chose which values to collect, but if you don't, some
5307 reasonable defaults will be collected.
5310 <Server "server_name">
5312 Collect "udp-answers" "udp-queries"
5313 Socket "/var/run/pdns.controlsocket"
5315 <Recursor "recursor_name">
5317 Collect "cache-hits" "cache-misses"
5318 Socket "/var/run/pdns_recursor.controlsocket"
5320 LocalSocket "/opt/collectd/var/run/collectd-powerdns"
5325 =item B<Server> and B<Recursor> block
5327 The B<Server> block defines one authoritative server to query, the B<Recursor>
5328 does the same for an recursing server. The possible options in both blocks are
5329 the same, though. The argument defines a name for the serverE<nbsp>/ recursor
5334 =item B<Collect> I<Field>
5336 Using the B<Collect> statement you can select which values to collect. Here,
5337 you specify the name of the values as used by the PowerDNS servers, e.E<nbsp>g.
5338 C<dlg-only-drops>, C<answers10-100>.
5340 The method of getting the values differs for B<Server> and B<Recursor> blocks:
5341 When querying the server a C<SHOW *> command is issued in any case, because
5342 that's the only way of getting multiple values out of the server at once.
5343 collectd then picks out the values you have selected. When querying the
5344 recursor, a command is generated to query exactly these values. So if you
5345 specify invalid fields when querying the recursor, a syntax error may be
5346 returned by the daemon and collectd may not collect any values at all.
5348 If no B<Collect> statement is given, the following B<Server> values will be
5355 =item packetcache-hit
5357 =item packetcache-miss
5359 =item packetcache-size
5361 =item query-cache-hit
5363 =item query-cache-miss
5365 =item recursing-answers
5367 =item recursing-questions
5379 The following B<Recursor> values will be collected by default:
5383 =item noerror-answers
5385 =item nxdomain-answers
5387 =item servfail-answers
5405 Please note that up to that point collectd doesn't know what values are
5406 available on the server and values that are added do not need a change of the
5407 mechanism so far. However, the values must be mapped to collectd's naming
5408 scheme, which is done using a lookup table that lists all known values. If
5409 values are added in the future and collectd does not know about them, you will
5410 get an error much like this:
5412 powerdns plugin: submit: Not found in lookup table: foobar = 42
5414 In this case please file a bug report with the collectd team.
5416 =item B<Socket> I<Path>
5418 Configures the path to the UNIX domain socket to be used when connecting to the
5419 daemon. By default C<${localstatedir}/run/pdns.controlsocket> will be used for
5420 an authoritative server and C<${localstatedir}/run/pdns_recursor.controlsocket>
5421 will be used for the recursor.
5425 =item B<LocalSocket> I<Path>
5427 Querying the recursor is done using UDP. When using UDP over UNIX domain
5428 sockets, the client socket needs a name in the file system, too. You can set
5429 this local name to I<Path> using the B<LocalSocket> option. The default is
5430 C<I<prefix>/var/run/collectd-powerdns>.
5434 =head2 Plugin C<processes>
5438 =item B<Process> I<Name>
5440 Select more detailed statistics of processes matching this name. The statistics
5441 collected for these selected processes are size of the resident segment size
5442 (RSS), user- and system-time used, number of processes and number of threads,
5443 io data (where available) and minor and major pagefaults.
5445 =item B<ProcessMatch> I<name> I<regex>
5447 Similar to the B<Process> option this allows to select more detailed
5448 statistics of processes matching the specified I<regex> (see L<regex(7)> for
5449 details). The statistics of all matching processes are summed up and
5450 dispatched to the daemon using the specified I<name> as an identifier. This
5451 allows to "group" several processes together. I<name> must not contain
5454 =item B<CollectContextSwitch> I<Boolean>
5456 Collect context switch of the process.
5460 =head2 Plugin C<protocols>
5462 Collects a lot of information about various network protocols, such as I<IP>,
5463 I<TCP>, I<UDP>, etc.
5465 Available configuration options:
5469 =item B<Value> I<Selector>
5471 Selects whether or not to select a specific value. The string being matched is
5472 of the form "I<Protocol>:I<ValueName>", where I<Protocol> will be used as the
5473 plugin instance and I<ValueName> will be used as type instance. An example of
5474 the string being used would be C<Tcp:RetransSegs>.
5476 You can use regular expressions to match a large number of values with just one
5477 configuration option. To select all "extended" I<TCP> values, you could use the
5478 following statement:
5482 Whether only matched values are selected or all matched values are ignored
5483 depends on the B<IgnoreSelected>. By default, only matched values are selected.
5484 If no value is configured at all, all values will be selected.
5486 =item B<IgnoreSelected> B<true>|B<false>
5488 If set to B<true>, inverts the selection made by B<Value>, i.E<nbsp>e. all
5489 matching values will be ignored.
5493 =head2 Plugin C<python>
5495 This plugin embeds a Python-interpreter into collectd and provides an interface
5496 to collectd's plugin system. See L<collectd-python(5)> for its documentation.
5498 =head2 Plugin C<routeros>
5500 The C<routeros> plugin connects to a device running I<RouterOS>, the
5501 Linux-based operating system for routers by I<MikroTik>. The plugin uses
5502 I<librouteros> to connect and reads information about the interfaces and
5503 wireless connections of the device. The configuration supports querying
5508 Host "router0.example.com"
5511 CollectInterface true
5516 Host "router1.example.com"
5519 CollectInterface true
5520 CollectRegistrationTable true
5526 As you can see above, the configuration of the I<routeros> plugin consists of
5527 one or more B<E<lt>RouterE<gt>> blocks. Within each block, the following
5528 options are understood:
5532 =item B<Host> I<Host>
5534 Hostname or IP-address of the router to connect to.
5536 =item B<Port> I<Port>
5538 Port name or port number used when connecting. If left unspecified, the default
5539 will be chosen by I<librouteros>, currently "8728". This option expects a
5540 string argument, even when a numeric port number is given.
5542 =item B<User> I<User>
5544 Use the user name I<User> to authenticate. Defaults to "admin".
5546 =item B<Password> I<Password>
5548 Set the password used to authenticate.
5550 =item B<CollectInterface> B<true>|B<false>
5552 When set to B<true>, interface statistics will be collected for all interfaces
5553 present on the device. Defaults to B<false>.
5555 =item B<CollectRegistrationTable> B<true>|B<false>
5557 When set to B<true>, information about wireless LAN connections will be
5558 collected. Defaults to B<false>.
5560 =item B<CollectCPULoad> B<true>|B<false>
5562 When set to B<true>, information about the CPU usage will be collected. The
5563 number is a dimensionless value where zero indicates no CPU usage at all.
5564 Defaults to B<false>.
5566 =item B<CollectMemory> B<true>|B<false>
5568 When enabled, the amount of used and free memory will be collected. How used
5569 memory is calculated is unknown, for example whether or not caches are counted
5571 Defaults to B<false>.
5573 =item B<CollectDF> B<true>|B<false>
5575 When enabled, the amount of used and free disk space will be collected.
5576 Defaults to B<false>.
5578 =item B<CollectDisk> B<true>|B<false>
5580 When enabled, the number of sectors written and bad blocks will be collected.
5581 Defaults to B<false>.
5585 =head2 Plugin C<redis>
5587 The I<Redis plugin> connects to one or more Redis servers and gathers
5588 information about each server's state. For each server there is a I<Node> block
5589 which configures the connection parameters for this node.
5596 <Query "LLEN myqueue">
5603 The information shown in the synopsis above is the I<default configuration>
5604 which is used by the plugin if no configuration is present.
5608 =item B<Node> I<Nodename>
5610 The B<Node> block identifies a new Redis node, that is a new Redis instance
5611 running in an specified host and port. The name for node is a canonical
5612 identifier which is used as I<plugin instance>. It is limited to
5613 64E<nbsp>characters in length.
5615 =item B<Host> I<Hostname>
5617 The B<Host> option is the hostname or IP-address where the Redis instance is
5620 =item B<Port> I<Port>
5622 The B<Port> option is the TCP port on which the Redis instance accepts
5623 connections. Either a service name of a port number may be given. Please note
5624 that numerical port numbers must be given as a string, too.
5626 =item B<Password> I<Password>
5628 Use I<Password> to authenticate when connecting to I<Redis>.
5630 =item B<Timeout> I<Milliseconds>
5632 The B<Timeout> option set the socket timeout for node response. Since the Redis
5633 read function is blocking, you should keep this value as low as possible. Keep
5634 in mind that the sum of all B<Timeout> values for all B<Nodes> should be lower
5635 than B<Interval> defined globally.
5637 =item B<Query> I<Querystring>
5639 The B<Query> block identifies a query to execute against the redis server.
5640 There may be an arbitrary number of queries to execute.
5642 =item B<Type> I<Collectd type>
5644 Within a query definition, a valid collectd type to use as when submitting
5645 the result of the query. When not supplied, will default to B<gauge>.
5647 =item B<Instance> I<Type instance>
5649 Within a query definition, an optional type instance to use when submitting
5650 the result of the query. When not supplied will default to the escaped
5651 command, up to 64 chars.
5655 =head2 Plugin C<rrdcached>
5657 The C<rrdcached> plugin uses the RRDtool accelerator daemon, L<rrdcached(1)>,
5658 to store values to RRD files in an efficient manner. The combination of the
5659 C<rrdcached> B<plugin> and the C<rrdcached> B<daemon> is very similar to the
5660 way the C<rrdtool> plugin works (see below). The added abstraction layer
5661 provides a number of benefits, though: Because the cache is not within
5662 C<collectd> anymore, it does not need to be flushed when C<collectd> is to be
5663 restarted. This results in much shorter (if any) gaps in graphs, especially
5664 under heavy load. Also, the C<rrdtool> command line utility is aware of the
5665 daemon so that it can flush values to disk automatically when needed. This
5666 allows to integrate automated flushing of values into graphing solutions much
5669 There are disadvantages, though: The daemon may reside on a different host, so
5670 it may not be possible for C<collectd> to create the appropriate RRD files
5671 anymore. And even if C<rrdcached> runs on the same host, it may run in a
5672 different base directory, so relative paths may do weird stuff if you're not
5675 So the B<recommended configuration> is to let C<collectd> and C<rrdcached> run
5676 on the same host, communicating via a UNIX domain socket. The B<DataDir>
5677 setting should be set to an absolute path, so that a changed base directory
5678 does not result in RRD files being createdE<nbsp>/ expected in the wrong place.
5682 =item B<DaemonAddress> I<Address>
5684 Address of the daemon as understood by the C<rrdc_connect> function of the RRD
5685 library. See L<rrdcached(1)> for details. Example:
5687 <Plugin "rrdcached">
5688 DaemonAddress "unix:/var/run/rrdcached.sock"
5691 =item B<DataDir> I<Directory>
5693 Set the base directory in which the RRD files reside. If this is a relative
5694 path, it is relative to the working base directory of the C<rrdcached> daemon!
5695 Use of an absolute path is recommended.
5697 =item B<CreateFiles> B<true>|B<false>
5699 Enables or disables the creation of RRD files. If the daemon is not running
5700 locally, or B<DataDir> is set to a relative path, this will not work as
5701 expected. Default is B<true>.
5703 =item B<CreateFilesAsync> B<false>|B<true>
5705 When enabled, new RRD files are enabled asynchronously, using a separate thread
5706 that runs in the background. This prevents writes to block, which is a problem
5707 especially when many hundreds of files need to be created at once. However,
5708 since the purpose of creating the files asynchronously is I<not> to block until
5709 the file is available, values before the file is available will be discarded.
5710 When disabled (the default) files are created synchronously, blocking for a
5711 short while, while the file is being written.
5713 =item B<StepSize> I<Seconds>
5715 B<Force> the stepsize of newly created RRD-files. Ideally (and per default)
5716 this setting is unset and the stepsize is set to the interval in which the data
5717 is collected. Do not use this option unless you absolutely have to for some
5718 reason. Setting this option may cause problems with the C<snmp plugin>, the
5719 C<exec plugin> or when the daemon is set up to receive data from other hosts.
5721 =item B<HeartBeat> I<Seconds>
5723 B<Force> the heartbeat of newly created RRD-files. This setting should be unset
5724 in which case the heartbeat is set to twice the B<StepSize> which should equal
5725 the interval in which data is collected. Do not set this option unless you have
5726 a very good reason to do so.
5728 =item B<RRARows> I<NumRows>
5730 The C<rrdtool plugin> calculates the number of PDPs per CDP based on the
5731 B<StepSize>, this setting and a timespan. This plugin creates RRD-files with
5732 three times five RRAs, i. e. five RRAs with the CFs B<MIN>, B<AVERAGE>, and
5733 B<MAX>. The five RRAs are optimized for graphs covering one hour, one day, one
5734 week, one month, and one year.
5736 So for each timespan, it calculates how many PDPs need to be consolidated into
5737 one CDP by calculating:
5738 number of PDPs = timespan / (stepsize * rrarows)
5740 Bottom line is, set this no smaller than the width of you graphs in pixels. The
5743 =item B<RRATimespan> I<Seconds>
5745 Adds an RRA-timespan, given in seconds. Use this option multiple times to have
5746 more then one RRA. If this option is never used, the built-in default of (3600,
5747 86400, 604800, 2678400, 31622400) is used.
5749 For more information on how RRA-sizes are calculated see B<RRARows> above.
5751 =item B<XFF> I<Factor>
5753 Set the "XFiles Factor". The default is 0.1. If unsure, don't set this option.
5754 I<Factor> must be in the range C<[0.0-1.0)>, i.e. between zero (inclusive) and
5757 =item B<CollectStatistics> B<false>|B<true>
5759 When set to B<true>, various statistics about the I<rrdcached> daemon will be
5760 collected, with "rrdcached" as the I<plugin name>. Defaults to B<false>.
5762 Statistics are read via I<rrdcached>s socket using the STATS command.
5763 See L<rrdcached(1)> for details.
5767 =head2 Plugin C<rrdtool>
5769 You can use the settings B<StepSize>, B<HeartBeat>, B<RRARows>, and B<XFF> to
5770 fine-tune your RRD-files. Please read L<rrdcreate(1)> if you encounter problems
5771 using these settings. If you don't want to dive into the depths of RRDtool, you
5772 can safely ignore these settings.
5776 =item B<DataDir> I<Directory>
5778 Set the directory to store RRD files under. By default RRD files are generated
5779 beneath the daemon's working directory, i.e. the B<BaseDir>.
5781 =item B<CreateFilesAsync> B<false>|B<true>
5783 When enabled, new RRD files are enabled asynchronously, using a separate thread
5784 that runs in the background. This prevents writes to block, which is a problem
5785 especially when many hundreds of files need to be created at once. However,
5786 since the purpose of creating the files asynchronously is I<not> to block until
5787 the file is available, values before the file is available will be discarded.
5788 When disabled (the default) files are created synchronously, blocking for a
5789 short while, while the file is being written.
5791 =item B<StepSize> I<Seconds>
5793 B<Force> the stepsize of newly created RRD-files. Ideally (and per default)
5794 this setting is unset and the stepsize is set to the interval in which the data
5795 is collected. Do not use this option unless you absolutely have to for some
5796 reason. Setting this option may cause problems with the C<snmp plugin>, the
5797 C<exec plugin> or when the daemon is set up to receive data from other hosts.
5799 =item B<HeartBeat> I<Seconds>
5801 B<Force> the heartbeat of newly created RRD-files. This setting should be unset
5802 in which case the heartbeat is set to twice the B<StepSize> which should equal
5803 the interval in which data is collected. Do not set this option unless you have
5804 a very good reason to do so.
5806 =item B<RRARows> I<NumRows>
5808 The C<rrdtool plugin> calculates the number of PDPs per CDP based on the
5809 B<StepSize>, this setting and a timespan. This plugin creates RRD-files with
5810 three times five RRAs, i.e. five RRAs with the CFs B<MIN>, B<AVERAGE>, and
5811 B<MAX>. The five RRAs are optimized for graphs covering one hour, one day, one
5812 week, one month, and one year.
5814 So for each timespan, it calculates how many PDPs need to be consolidated into
5815 one CDP by calculating:
5816 number of PDPs = timespan / (stepsize * rrarows)
5818 Bottom line is, set this no smaller than the width of you graphs in pixels. The
5821 =item B<RRATimespan> I<Seconds>
5823 Adds an RRA-timespan, given in seconds. Use this option multiple times to have
5824 more then one RRA. If this option is never used, the built-in default of (3600,
5825 86400, 604800, 2678400, 31622400) is used.
5827 For more information on how RRA-sizes are calculated see B<RRARows> above.
5829 =item B<XFF> I<Factor>
5831 Set the "XFiles Factor". The default is 0.1. If unsure, don't set this option.
5832 I<Factor> must be in the range C<[0.0-1.0)>, i.e. between zero (inclusive) and
5835 =item B<CacheFlush> I<Seconds>
5837 When the C<rrdtool> plugin uses a cache (by setting B<CacheTimeout>, see below)
5838 it writes all values for a certain RRD-file if the oldest value is older than
5839 (or equal to) the number of seconds specified. If some RRD-file is not updated
5840 anymore for some reason (the computer was shut down, the network is broken,
5841 etc.) some values may still be in the cache. If B<CacheFlush> is set, then the
5842 entire cache is searched for entries older than B<CacheTimeout> seconds and
5843 written to disk every I<Seconds> seconds. Since this is kind of expensive and
5844 does nothing under normal circumstances, this value should not be too small.
5845 900 seconds might be a good value, though setting this to 7200 seconds doesn't
5846 normally do much harm either.
5848 =item B<CacheTimeout> I<Seconds>
5850 If this option is set to a value greater than zero, the C<rrdtool plugin> will
5851 save values in a cache, as described above. Writing multiple values at once
5852 reduces IO-operations and thus lessens the load produced by updating the files.
5853 The trade off is that the graphs kind of "drag behind" and that more memory is
5856 =item B<WritesPerSecond> I<Updates>
5858 When collecting many statistics with collectd and the C<rrdtool> plugin, you
5859 will run serious performance problems. The B<CacheFlush> setting and the
5860 internal update queue assert that collectd continues to work just fine even
5861 under heavy load, but the system may become very unresponsive and slow. This is
5862 a problem especially if you create graphs from the RRD files on the same
5863 machine, for example using the C<graph.cgi> script included in the
5864 C<contrib/collection3/> directory.
5866 This setting is designed for very large setups. Setting this option to a value
5867 between 25 and 80 updates per second, depending on your hardware, will leave
5868 the server responsive enough to draw graphs even while all the cached values
5869 are written to disk. Flushed values, i.E<nbsp>e. values that are forced to disk
5870 by the B<FLUSH> command, are B<not> effected by this limit. They are still
5871 written as fast as possible, so that web frontends have up to date data when
5874 For example: If you have 100,000 RRD files and set B<WritesPerSecond> to 30
5875 updates per second, writing all values to disk will take approximately
5876 56E<nbsp>minutes. Together with the flushing ability that's integrated into
5877 "collection3" you'll end up with a responsive and fast system, up to date
5878 graphs and basically a "backup" of your values every hour.
5880 =item B<RandomTimeout> I<Seconds>
5882 When set, the actual timeout for each value is chosen randomly between
5883 I<CacheTimeout>-I<RandomTimeout> and I<CacheTimeout>+I<RandomTimeout>. The
5884 intention is to avoid high load situations that appear when many values timeout
5885 at the same time. This is especially a problem shortly after the daemon starts,
5886 because all values were added to the internal cache at roughly the same time.
5890 =head2 Plugin C<sensors>
5892 The I<Sensors plugin> uses B<lm_sensors> to retrieve sensor-values. This means
5893 that all the needed modules have to be loaded and lm_sensors has to be
5894 configured (most likely by editing F</etc/sensors.conf>. Read
5895 L<sensors.conf(5)> for details.
5897 The B<lm_sensors> homepage can be found at
5898 L<http://secure.netroedge.com/~lm78/>.
5902 =item B<SensorConfigFile> I<File>
5904 Read the I<lm_sensors> configuration from I<File>. When unset (recommended),
5905 the library's default will be used.
5907 =item B<Sensor> I<chip-bus-address/type-feature>
5909 Selects the name of the sensor which you want to collect or ignore, depending
5910 on the B<IgnoreSelected> below. For example, the option "B<Sensor>
5911 I<it8712-isa-0290/voltage-in1>" will cause collectd to gather data for the
5912 voltage sensor I<in1> of the I<it8712> on the isa bus at the address 0290.
5914 =item B<IgnoreSelected> I<true>|I<false>
5916 If no configuration if given, the B<sensors>-plugin will collect data from all
5917 sensors. This may not be practical, especially for uninteresting sensors.
5918 Thus, you can use the B<Sensor>-option to pick the sensors you're interested
5919 in. Sometimes, however, it's easier/preferred to collect all sensors I<except> a
5920 few ones. This option enables you to do that: By setting B<IgnoreSelected> to
5921 I<true> the effect of B<Sensor> is inverted: All selected sensors are ignored
5922 and all other sensors are collected.
5926 =head2 Plugin C<sigrok>
5928 The I<sigrok plugin> uses I<libsigrok> to retrieve measurements from any device
5929 supported by the L<sigrok|http://sigrok.org/> project.
5935 <Device "AC Voltage">
5940 <Device "Sound Level">
5941 Driver "cem-dt-885x"
5948 =item B<LogLevel> B<0-5>
5950 The I<sigrok> logging level to pass on to the I<collectd> log, as a number
5951 between B<0> and B<5> (inclusive). These levels correspond to C<None>,
5952 C<Errors>, C<Warnings>, C<Informational>, C<Debug >and C<Spew>, respectively.
5953 The default is B<2> (C<Warnings>). The I<sigrok> log messages, regardless of
5954 their level, are always submitted to I<collectd> at its INFO log level.
5956 =item E<lt>B<Device> I<Name>E<gt>
5958 A sigrok-supported device, uniquely identified by this section's options. The
5959 I<Name> is passed to I<collectd> as the I<plugin instance>.
5961 =item B<Driver> I<DriverName>
5963 The sigrok driver to use for this device.
5965 =item B<Conn> I<ConnectionSpec>
5967 If the device cannot be auto-discovered, or more than one might be discovered
5968 by the driver, I<ConnectionSpec> specifies the connection string to the device.
5969 It can be of the form of a device path (e.g.E<nbsp>C</dev/ttyUSB2>), or, in
5970 case of a non-serial USB-connected device, the USB I<VendorID>B<.>I<ProductID>
5971 separated by a period (e.g.E<nbsp>C<0403.6001>). A USB device can also be
5972 specified as I<Bus>B<.>I<Address> (e.g.E<nbsp>C<1.41>).
5974 =item B<SerialComm> I<SerialSpec>
5976 For serial devices with non-standard port settings, this option can be used
5977 to specify them in a form understood by I<sigrok>, e.g.E<nbsp>C<9600/8n1>.
5978 This should not be necessary; drivers know how to communicate with devices they
5981 =item B<MinimumInterval> I<Seconds>
5983 Specifies the minimum time between measurement dispatches to I<collectd>, in
5984 seconds. Since some I<sigrok> supported devices can acquire measurements many
5985 times per second, it may be necessary to throttle these. For example, the
5986 I<RRD plugin> cannot process writes more than once per second.
5988 The default B<MinimumInterval> is B<0>, meaning measurements received from the
5989 device are always dispatched to I<collectd>. When throttled, unused
5990 measurements are discarded.
5994 =head2 Plugin C<smart>
5996 The C<smart> plugin collects SMART information from physical
5997 disks. Values collectd include temperature, power cycle count, poweron
5998 time and bad sectors. Also, all SMART attributes are collected along
5999 with the normalized current value, the worst value, the threshold and
6000 a human readable value.
6002 Using the following two options you can ignore some disks or configure the
6003 collection only of specific disks.
6007 =item B<Disk> I<Name>
6009 Select the disk I<Name>. Whether it is collected or ignored depends on the
6010 B<IgnoreSelected> setting, see below. As with other plugins that use the
6011 daemon's ignorelist functionality, a string that starts and ends with a slash
6012 is interpreted as a regular expression. Examples:
6017 =item B<IgnoreSelected> B<true>|B<false>
6019 Sets whether selected disks, i.E<nbsp>e. the ones matches by any of the B<Disk>
6020 statements, are ignored or if all other disks are ignored. The behavior
6021 (hopefully) is intuitive: If no B<Disk> option is configured, all disks are
6022 collected. If at least one B<Disk> option is given and no B<IgnoreSelected> or
6023 set to B<false>, B<only> matching disks will be collected. If B<IgnoreSelected>
6024 is set to B<true>, all disks are collected B<except> the ones matched.
6028 =head2 Plugin C<snmp>
6030 Since the configuration of the C<snmp plugin> is a little more complicated than
6031 other plugins, its documentation has been moved to an own manpage,
6032 L<collectd-snmp(5)>. Please see there for details.
6034 =head2 Plugin C<statsd>
6036 The I<statsd plugin> listens to a UDP socket, reads "events" in the statsd
6037 protocol and dispatches rates or other aggregates of these numbers
6040 The plugin implements the I<Counter>, I<Timer>, I<Gauge> and I<Set> types which
6041 are dispatched as the I<collectd> types C<derive>, C<latency>, C<gauge> and
6042 C<objects> respectively.
6044 The following configuration options are valid:
6048 =item B<Host> I<Host>
6050 Bind to the hostname / address I<Host>. By default, the plugin will bind to the
6051 "any" address, i.e. accept packets sent to any of the hosts addresses.
6053 =item B<Port> I<Port>
6055 UDP port to listen to. This can be either a service name or a port number.
6056 Defaults to C<8125>.
6058 =item B<DeleteCounters> B<false>|B<true>
6060 =item B<DeleteTimers> B<false>|B<true>
6062 =item B<DeleteGauges> B<false>|B<true>
6064 =item B<DeleteSets> B<false>|B<true>
6066 These options control what happens if metrics are not updated in an interval.
6067 If set to B<False>, the default, metrics are dispatched unchanged, i.e. the
6068 rate of counters and size of sets will be zero, timers report C<NaN> and gauges
6069 are unchanged. If set to B<True>, the such metrics are not dispatched and
6070 removed from the internal cache.
6072 =item B<TimerPercentile> I<Percent>
6074 Calculate and dispatch the configured percentile, i.e. compute the latency, so
6075 that I<Percent> of all reported timers are smaller than or equal to the
6076 computed latency. This is useful for cutting off the long tail latency, as it's
6077 often done in I<Service Level Agreements> (SLAs).
6079 Different percentiles can be calculated by setting this option several times.
6080 If none are specified, no percentiles are calculated / dispatched.
6082 =item B<TimerLower> B<false>|B<true>
6084 =item B<TimerUpper> B<false>|B<true>
6086 =item B<TimerSum> B<false>|B<true>
6088 =item B<TimerCount> B<false>|B<true>
6090 Calculate and dispatch various values out of I<Timer> metrics received during
6091 an interval. If set to B<False>, the default, these values aren't calculated /
6096 =head2 Plugin C<swap>
6098 The I<Swap plugin> collects information about used and available swap space. On
6099 I<Linux> and I<Solaris>, the following options are available:
6103 =item B<ReportByDevice> B<false>|B<true>
6105 Configures how to report physical swap devices. If set to B<false> (the
6106 default), the summary over all swap devices is reported only, i.e. the globally
6107 used and available space over all devices. If B<true> is configured, the used
6108 and available space of each device will be reported separately.
6110 This option is only available if the I<Swap plugin> can read C</proc/swaps>
6111 (under Linux) or use the L<swapctl(2)> mechanism (under I<Solaris>).
6113 =item B<ReportBytes> B<false>|B<true>
6115 When enabled, the I<swap I/O> is reported in bytes. When disabled, the default,
6116 I<swap I/O> is reported in pages. This option is available under Linux only.
6118 =item B<ValuesAbsolute> B<true>|B<false>
6120 Enables or disables reporting of absolute swap metrics, i.e. number of I<bytes>
6121 available and used. Defaults to B<true>.
6123 =item B<ValuesPercentage> B<false>|B<true>
6125 Enables or disables reporting of relative swap metrics, i.e. I<percent>
6126 available and free. Defaults to B<false>.
6128 This is useful for deploying I<collectd> in a heterogeneous environment, where
6129 swap sizes differ and you want to specify generic thresholds or similar.
6133 =head2 Plugin C<syslog>
6137 =item B<LogLevel> B<debug|info|notice|warning|err>
6139 Sets the log-level. If, for example, set to B<notice>, then all events with
6140 severity B<notice>, B<warning>, or B<err> will be submitted to the
6143 Please note that B<debug> is only available if collectd has been compiled with
6146 =item B<NotifyLevel> B<OKAY>|B<WARNING>|B<FAILURE>
6148 Controls which notifications should be sent to syslog. The default behaviour is
6149 not to send any. Less severe notifications always imply logging more severe
6150 notifications: Setting this to B<OKAY> means all notifications will be sent to
6151 syslog, setting this to B<WARNING> will send B<WARNING> and B<FAILURE>
6152 notifications but will dismiss B<OKAY> notifications. Setting this option to
6153 B<FAILURE> will only send failures to syslog.
6157 =head2 Plugin C<table>
6159 The C<table plugin> provides generic means to parse tabular data and dispatch
6160 user specified values. Values are selected based on column numbers. For
6161 example, this plugin may be used to get values from the Linux L<proc(5)>
6162 filesystem or CSV (comma separated values) files.
6165 <Table "/proc/slabinfo">
6170 InstancePrefix "active_objs"
6176 InstancePrefix "objperslab"
6183 The configuration consists of one or more B<Table> blocks, each of which
6184 configures one file to parse. Within each B<Table> block, there are one or
6185 more B<Result> blocks, which configure which data to select and how to
6188 The following options are available inside a B<Table> block:
6192 =item B<Instance> I<instance>
6194 If specified, I<instance> is used as the plugin instance. So, in the above
6195 example, the plugin name C<table-slabinfo> would be used. If omitted, the
6196 filename of the table is used instead, with all special characters replaced
6197 with an underscore (C<_>).
6199 =item B<Separator> I<string>
6201 Any character of I<string> is interpreted as a delimiter between the different
6202 columns of the table. A sequence of two or more contiguous delimiters in the
6203 table is considered to be a single delimiter, i.E<nbsp>e. there cannot be any
6204 empty columns. The plugin uses the L<strtok_r(3)> function to parse the lines
6205 of a table - see its documentation for more details. This option is mandatory.
6207 A horizontal tab, newline and carriage return may be specified by C<\\t>,
6208 C<\\n> and C<\\r> respectively. Please note that the double backslashes are
6209 required because of collectd's config parsing.
6213 The following options are available inside a B<Result> block:
6217 =item B<Type> I<type>
6219 Sets the type used to dispatch the values to the daemon. Detailed information
6220 about types and their configuration can be found in L<types.db(5)>. This
6221 option is mandatory.
6223 =item B<InstancePrefix> I<prefix>
6225 If specified, prepend I<prefix> to the type instance. If omitted, only the
6226 B<InstancesFrom> option is considered for the type instance.
6228 =item B<InstancesFrom> I<column0> [I<column1> ...]
6230 If specified, the content of the given columns (identified by the column
6231 number starting at zero) will be used to create the type instance for each
6232 row. Multiple values (and the instance prefix) will be joined together with
6233 dashes (I<->) as separation character. If omitted, only the B<InstancePrefix>
6234 option is considered for the type instance.
6236 The plugin itself does not check whether or not all built instances are
6237 different. It’s your responsibility to assure that each is unique. This is
6238 especially true, if you do not specify B<InstancesFrom>: B<You> have to make
6239 sure that the table only contains one row.
6241 If neither B<InstancePrefix> nor B<InstancesFrom> is given, the type instance
6244 =item B<ValuesFrom> I<column0> [I<column1> ...]
6246 Specifies the columns (identified by the column numbers starting at zero)
6247 whose content is used as the actual data for the data sets that are dispatched
6248 to the daemon. How many such columns you need is determined by the B<Type>
6249 setting above. If you specify too many or not enough columns, the plugin will
6250 complain about that and no data will be submitted to the daemon. The plugin
6251 uses L<strtoll(3)> and L<strtod(3)> to parse counter and gauge values
6252 respectively, so anything supported by those functions is supported by the
6253 plugin as well. This option is mandatory.
6257 =head2 Plugin C<tail>
6259 The C<tail plugin> follows logfiles, just like L<tail(1)> does, parses
6260 each line and dispatches found values. What is matched can be configured by the
6261 user using (extended) regular expressions, as described in L<regex(7)>.
6264 <File "/var/log/exim4/mainlog">
6268 Regex "S=([1-9][0-9]*)"
6274 Regex "\\<R=local_user\\>"
6275 ExcludeRegex "\\<R=local_user\\>.*mail_spool defer"
6278 Instance "local_user"
6283 The config consists of one or more B<File> blocks, each of which configures one
6284 logfile to parse. Within each B<File> block, there are one or more B<Match>
6285 blocks, which configure a regular expression to search for.
6287 The B<Instance> option in the B<File> block may be used to set the plugin
6288 instance. So in the above example the plugin name C<tail-foo> would be used.
6289 This plugin instance is for all B<Match> blocks that B<follow> it, until the
6290 next B<Instance> option. This way you can extract several plugin instances from
6291 one logfile, handy when parsing syslog and the like.
6293 The B<Interval> option allows you to define the length of time between reads. If
6294 this is not set, the default Interval will be used.
6296 Each B<Match> block has the following options to describe how the match should
6301 =item B<Regex> I<regex>
6303 Sets the regular expression to use for matching against a line. The first
6304 subexpression has to match something that can be turned into a number by
6305 L<strtoll(3)> or L<strtod(3)>, depending on the value of C<CounterAdd>, see
6306 below. Because B<extended> regular expressions are used, you do not need to use
6307 backslashes for subexpressions! If in doubt, please consult L<regex(7)>. Due to
6308 collectd's config parsing you need to escape backslashes, though. So if you
6309 want to match literal parentheses you need to do the following:
6311 Regex "SPAM \\(Score: (-?[0-9]+\\.[0-9]+)\\)"
6313 =item B<ExcludeRegex> I<regex>
6315 Sets an optional regular expression to use for excluding lines from the match.
6316 An example which excludes all connections from localhost from the match:
6318 ExcludeRegex "127\\.0\\.0\\.1"
6320 =item B<DSType> I<Type>
6322 Sets how the values are cumulated. I<Type> is one of:
6326 =item B<GaugeAverage>
6328 Calculate the average.
6332 Use the smallest number only.
6336 Use the greatest number only.
6340 Use the last number found.
6346 =item B<AbsoluteSet>
6348 The matched number is a counter. Simply I<sets> the internal counter to this
6349 value. Variants exist for C<COUNTER>, C<DERIVE>, and C<ABSOLUTE> data sources.
6357 Add the matched value to the internal counter. In case of B<DeriveAdd>, the
6358 matched number may be negative, which will effectively subtract from the
6367 Increase the internal counter by one. These B<DSType> are the only ones that do
6368 not use the matched subexpression, but simply count the number of matched
6369 lines. Thus, you may use a regular expression without submatch in this case.
6373 As you'd expect the B<Gauge*> types interpret the submatch as a floating point
6374 number, using L<strtod(3)>. The B<Counter*> and B<AbsoluteSet> types interpret
6375 the submatch as an unsigned integer using L<strtoull(3)>. The B<Derive*> types
6376 interpret the submatch as a signed integer using L<strtoll(3)>. B<CounterInc>
6377 and B<DeriveInc> do not use the submatch at all and it may be omitted in this
6380 =item B<Type> I<Type>
6382 Sets the type used to dispatch this value. Detailed information about types and
6383 their configuration can be found in L<types.db(5)>.
6385 =item B<Instance> I<TypeInstance>
6387 This optional setting sets the type instance to use.
6391 =head2 Plugin C<tail_csv>
6393 The I<tail_csv plugin> reads files in the CSV format, e.g. the statistics file
6394 written by I<Snort>.
6399 <Metric "snort-dropped">
6404 <File "/var/log/snort/snort.stats">
6405 Instance "snort-eth0"
6407 Collect "snort-dropped"
6411 The configuration consists of one or more B<Metric> blocks that define an index
6412 into the line of the CSV file and how this value is mapped to I<collectd's>
6413 internal representation. These are followed by one or more B<Instance> blocks
6414 which configure which file to read, in which interval and which metrics to
6419 =item E<lt>B<Metric> I<Name>E<gt>
6421 The B<Metric> block configures a new metric to be extracted from the statistics
6422 file and how it is mapped on I<collectd's> data model. The string I<Name> is
6423 only used inside the B<Instance> blocks to refer to this block, so you can use
6424 one B<Metric> block for multiple CSV files.
6428 =item B<Type> I<Type>
6430 Configures which I<Type> to use when dispatching this metric. Types are defined
6431 in the L<types.db(5)> file, see the appropriate manual page for more
6432 information on specifying types. Only types with a single I<data source> are
6433 supported by the I<tail_csv plugin>. The information whether the value is an
6434 absolute value (i.e. a C<GAUGE>) or a rate (i.e. a C<DERIVE>) is taken from the
6435 I<Type's> definition.
6437 =item B<Instance> I<TypeInstance>
6439 If set, I<TypeInstance> is used to populate the type instance field of the
6440 created value lists. Otherwise, no type instance is used.
6442 =item B<ValueFrom> I<Index>
6444 Configure to read the value from the field with the zero-based index I<Index>.
6445 If the value is parsed as signed integer, unsigned integer or double depends on
6446 the B<Type> setting, see above.
6450 =item E<lt>B<File> I<Path>E<gt>
6452 Each B<File> block represents one CSV file to read. There must be at least one
6453 I<File> block but there can be multiple if you have multiple CSV files.
6457 =item B<Instance> I<PluginInstance>
6459 Sets the I<plugin instance> used when dispatching the values.
6461 =item B<Collect> I<Metric>
6463 Specifies which I<Metric> to collect. This option must be specified at least
6464 once, and you can use this option multiple times to specify more than one
6465 metric to be extracted from this statistic file.
6467 =item B<Interval> I<Seconds>
6469 Configures the interval in which to read values from this instance / file.
6470 Defaults to the plugin's default interval.
6472 =item B<TimeFrom> I<Index>
6474 Rather than using the local time when dispatching a value, read the timestamp
6475 from the field with the zero-based index I<Index>. The value is interpreted as
6476 seconds since epoch. The value is parsed as a double and may be factional.
6482 =head2 Plugin C<teamspeak2>
6484 The C<teamspeak2 plugin> connects to the query port of a teamspeak2 server and
6485 polls interesting global and virtual server data. The plugin can query only one
6486 physical server but unlimited virtual servers. You can use the following
6487 options to configure it:
6491 =item B<Host> I<hostname/ip>
6493 The hostname or ip which identifies the physical server.
6496 =item B<Port> I<port>
6498 The query port of the physical server. This needs to be a string.
6501 =item B<Server> I<port>
6503 This option has to be added once for every virtual server the plugin should
6504 query. If you want to query the virtual server on port 8767 this is what the
6505 option would look like:
6509 This option, although numeric, needs to be a string, i.E<nbsp>e. you B<must>
6510 use quotes around it! If no such statement is given only global information
6515 =head2 Plugin C<ted>
6517 The I<TED> plugin connects to a device of "The Energy Detective", a device to
6518 measure power consumption. These devices are usually connected to a serial
6519 (RS232) or USB port. The plugin opens a configured device and tries to read the
6520 current energy readings. For more information on TED, visit
6521 L<http://www.theenergydetective.com/>.
6523 Available configuration options:
6527 =item B<Device> I<Path>
6529 Path to the device on which TED is connected. collectd will need read and write
6530 permissions on that file.
6532 Default: B</dev/ttyUSB0>
6534 =item B<Retries> I<Num>
6536 Apparently reading from TED is not that reliable. You can therefore configure a
6537 number of retries here. You only configure the I<retries> here, to if you
6538 specify zero, one reading will be performed (but no retries if that fails); if
6539 you specify three, a maximum of four readings are performed. Negative values
6546 =head2 Plugin C<tcpconns>
6548 The C<tcpconns plugin> counts the number of currently established TCP
6549 connections based on the local port and/or the remote port. Since there may be
6550 a lot of connections the default if to count all connections with a local port,
6551 for which a listening socket is opened. You can use the following options to
6552 fine-tune the ports you are interested in:
6556 =item B<ListeningPorts> I<true>|I<false>
6558 If this option is set to I<true>, statistics for all local ports for which a
6559 listening socket exists are collected. The default depends on B<LocalPort> and
6560 B<RemotePort> (see below): If no port at all is specifically selected, the
6561 default is to collect listening ports. If specific ports (no matter if local or
6562 remote ports) are selected, this option defaults to I<false>, i.E<nbsp>e. only
6563 the selected ports will be collected unless this option is set to I<true>
6566 =item B<LocalPort> I<Port>
6568 Count the connections to a specific local port. This can be used to see how
6569 many connections are handled by a specific daemon, e.E<nbsp>g. the mailserver.
6570 You have to specify the port in numeric form, so for the mailserver example
6571 you'd need to set B<25>.
6573 =item B<RemotePort> I<Port>
6575 Count the connections to a specific remote port. This is useful to see how
6576 much a remote service is used. This is most useful if you want to know how many
6577 connections a local service has opened to remote services, e.E<nbsp>g. how many
6578 connections a mail server or news server has to other mail or news servers, or
6579 how many connections a web proxy holds to web servers. You have to give the
6580 port in numeric form.
6582 =item B<AllPortsSummary> I<true>|I<false>
6584 If this option is set to I<true> a summary of statistics from all connections
6585 are collectd. This option defaults to I<false>.
6589 =head2 Plugin C<thermal>
6593 =item B<ForceUseProcfs> I<true>|I<false>
6595 By default, the I<Thermal plugin> tries to read the statistics from the Linux
6596 C<sysfs> interface. If that is not available, the plugin falls back to the
6597 C<procfs> interface. By setting this option to I<true>, you can force the
6598 plugin to use the latter. This option defaults to I<false>.
6600 =item B<Device> I<Device>
6602 Selects the name of the thermal device that you want to collect or ignore,
6603 depending on the value of the B<IgnoreSelected> option. This option may be
6604 used multiple times to specify a list of devices.
6606 =item B<IgnoreSelected> I<true>|I<false>
6608 Invert the selection: If set to true, all devices B<except> the ones that
6609 match the device names specified by the B<Device> option are collected. By
6610 default only selected devices are collected if a selection is made. If no
6611 selection is configured at all, B<all> devices are selected.
6615 =head2 Plugin C<threshold>
6617 The I<Threshold plugin> checks values collected or received by I<collectd>
6618 against a configurable I<threshold> and issues I<notifications> if values are
6621 Documentation for this plugin is available in the L<collectd-threshold(5)>
6624 =head2 Plugin C<tokyotyrant>
6626 The I<TokyoTyrant plugin> connects to a TokyoTyrant server and collects a
6627 couple metrics: number of records, and database size on disk.
6631 =item B<Host> I<Hostname/IP>
6633 The hostname or ip which identifies the server.
6634 Default: B<127.0.0.1>
6636 =item B<Port> I<Service/Port>
6638 The query port of the server. This needs to be a string, even if the port is
6639 given in its numeric form.
6644 =head2 Plugin C<turbostat>
6646 The I<Turbostat plugin> reads CPU frequency and C-state residency on modern
6647 Intel processors by using the new Model Specific Registers.
6651 =item B<CoreCstates> I<Bitmask(Integer)>
6653 Bitmask of the list of core C states supported by the processor.
6654 This option should only be used if the automated detection fails.
6655 Default value extracted from the cpu model and family.
6657 Currently supported C-states (by this plugin): 3, 6, 7
6659 Example: (1<<3)+(1<<6)+(1<<7) = 392 for all states
6661 =item B<PackageCstates> I<Bitmask(Integer)>
6663 Bitmask of the list of pacages C states supported by the processor.
6664 This option should only be used if the automated detection fails.
6665 Default value extracted from the cpu model and family.
6667 Currently supported C-states (by this plugin): 2, 3, 6, 7, 8, 9, 10
6669 Example: (1<<2)+(1<<3)+(1<<6)+(1<<7) = 396 for states 2, 3, 6 and 7
6671 =item B<SystemManagementInterrupt> I<true>|I<false>
6673 Boolean enabling the collection of the I/O System-Management Interrupt
6674 counter'. This option should only be used if the automated detection
6675 fails or if you want to disable this feature.
6677 =item B<DigitalTemperatureSensor> I<true>|I<false>
6679 Boolean enabling the collection of the temperature of each core.
6680 This option should only be used if the automated detectionfails or
6681 if you want to disable this feature.
6683 =item B<DigitalTemperatureSensor> I<true>|I<false>
6685 Boolean enabling the collection of the temperature of each package.
6686 This option should only be used if the automated detectionfails or
6687 if you want to disable this feature.
6689 =item B<TCCActivationTemp> I<Temperature>
6691 Thermal Control Circuit Activation Temperature of the installed
6692 CPU. This temperature is used when collecting the temperature of
6693 cores or packages. This option should only be used if the automated
6694 detection fails. Default value extracted from B<MSR_IA32_TEMPERATURE_TARGET>
6696 =item B<RunningAveragePowerLimit> I<Bitmask(Integer)>
6698 Bitmask of the list of elements to be thermally monitored. This option
6699 should only be used if the automated detection fails or if you want to
6700 disable some collections. The different bits of this bitmask accepted
6705 =item 0 ('1'): Package
6709 =item 2 ('4'): Cores
6711 =item 3 ('8'): Embedded graphic device
6717 =head2 Plugin C<unixsock>
6721 =item B<SocketFile> I<Path>
6723 Sets the socket-file which is to be created.
6725 =item B<SocketGroup> I<Group>
6727 If running as root change the group of the UNIX-socket after it has been
6728 created. Defaults to B<collectd>.
6730 =item B<SocketPerms> I<Permissions>
6732 Change the file permissions of the UNIX-socket after it has been created. The
6733 permissions must be given as a numeric, octal value as you would pass to
6734 L<chmod(1)>. Defaults to B<0770>.
6736 =item B<DeleteSocket> B<false>|B<true>
6738 If set to B<true>, delete the socket file before calling L<bind(2)>, if a file
6739 with the given name already exists. If I<collectd> crashes a socket file may be
6740 left over, preventing the daemon from opening a new socket when restarted.
6741 Since this is potentially dangerous, this defaults to B<false>.
6745 =head2 Plugin C<uuid>
6747 This plugin, if loaded, causes the Hostname to be taken from the machine's
6748 UUID. The UUID is a universally unique designation for the machine, usually
6749 taken from the machine's BIOS. This is most useful if the machine is running in
6750 a virtual environment such as Xen, in which case the UUID is preserved across
6751 shutdowns and migration.
6753 The following methods are used to find the machine's UUID, in order:
6759 Check I</etc/uuid> (or I<UUIDFile>).
6763 Check for UUID from HAL (L<http://www.freedesktop.org/wiki/Software/hal>) if
6768 Check for UUID from C<dmidecode> / SMBIOS.
6772 Check for UUID from Xen hypervisor.
6776 If no UUID can be found then the hostname is not modified.
6780 =item B<UUIDFile> I<Path>
6782 Take the UUID from the given file (default I</etc/uuid>).
6786 =head2 Plugin C<varnish>
6788 The I<varnish plugin> collects information about Varnish, an HTTP accelerator.
6789 It collects a subset of the values displayed by L<varnishstat(1)>, and
6790 organizes them in categories which can be enabled or disabled. Currently only
6791 metrics shown in L<varnishstat(1)>'s I<MAIN> section are collected. The exact
6792 meaning of each metric can be found in L<varnish-counters(7)>.
6797 <Instance "example">
6801 CollectConnections true
6802 CollectDirectorDNS false
6806 CollectObjects false
6808 CollectSession false
6818 CollectWorkers false
6822 The configuration consists of one or more E<lt>B<Instance>E<nbsp>I<Name>E<gt>
6823 blocks. I<Name> is the parameter passed to "varnishd -n". If left empty, it
6824 will collectd statistics from the default "varnishd" instance (this should work
6825 fine in most cases).
6827 Inside each E<lt>B<Instance>E<gt> blocks, the following options are recognized:
6831 =item B<CollectBackend> B<true>|B<false>
6833 Back-end connection statistics, such as successful, reused,
6834 and closed connections. True by default.
6836 =item B<CollectBan> B<true>|B<false>
6838 Statistics about ban operations, such as number of bans added, retired, and
6839 number of objects tested against ban operations. Only available with Varnish
6840 3.x and above. False by default.
6842 =item B<CollectCache> B<true>|B<false>
6844 Cache hits and misses. True by default.
6846 =item B<CollectConnections> B<true>|B<false>
6848 Number of client connections received, accepted and dropped. True by default.
6850 =item B<CollectDirectorDNS> B<true>|B<false>
6852 DNS director lookup cache statistics. Only available with Varnish 3.x. False by
6855 =item B<CollectESI> B<true>|B<false>
6857 Edge Side Includes (ESI) parse statistics. False by default.
6859 =item B<CollectFetch> B<true>|B<false>
6861 Statistics about fetches (HTTP requests sent to the backend). False by default.
6863 =item B<CollectHCB> B<true>|B<false>
6865 Inserts and look-ups in the crit bit tree based hash. Look-ups are
6866 divided into locked and unlocked look-ups. False by default.
6868 =item B<CollectObjects> B<true>|B<false>
6870 Statistics on cached objects: number of objects expired, nuked (prematurely
6871 expired), saved, moved, etc. False by default.
6873 =item B<CollectPurge> B<true>|B<false>
6875 Statistics about purge operations, such as number of purges added, retired, and
6876 number of objects tested against purge operations. Only available with Varnish
6877 2.x. False by default.
6879 =item B<CollectSession> B<true>|B<false>
6881 Client session statistics. Number of past and current sessions, session herd and
6882 linger counters, etc. False by default. Note that if using Varnish 4.x, some
6883 metrics found in the Connections and Threads sections with previous versions of
6884 Varnish have been moved here.
6886 =item B<CollectSHM> B<true>|B<false>
6888 Statistics about the shared memory log, a memory region to store
6889 log messages which is flushed to disk when full. True by default.
6891 =item B<CollectSMA> B<true>|B<false>
6893 malloc or umem (umem_alloc(3MALLOC) based) storage statistics. The umem storage
6894 component is Solaris specific. Only available with Varnish 2.x. False by
6897 =item B<CollectSMS> B<true>|B<false>
6899 synth (synthetic content) storage statistics. This storage
6900 component is used internally only. False by default.
6902 =item B<CollectSM> B<true>|B<false>
6904 file (memory mapped file) storage statistics. Only available with Varnish 2.x.
6907 =item B<CollectStruct> B<true>|B<false>
6909 Current varnish internal state statistics. Number of current sessions, objects
6910 in cache store, open connections to backends (with Varnish 2.x), etc. False by
6913 =item B<CollectTotals> B<true>|B<false>
6915 Collects overview counters, such as the number of sessions created,
6916 the number of requests and bytes transferred. False by default.
6918 =item B<CollectUptime> B<true>|B<false>
6920 Varnish uptime. Only available with Varnish 3.x and above. False by default.
6922 =item B<CollectVCL> B<true>|B<false>
6924 Number of total (available + discarded) VCL (config files). False by default.
6926 =item B<CollectVSM> B<true>|B<false>
6928 Collect statistics about Varnish's shared memory usage (used by the logging and
6929 statistics subsystems). Only available with Varnish 4.x. False by default.
6931 =item B<CollectWorkers> B<true>|B<false>
6933 Collect statistics about worker threads. False by default.
6937 =head2 Plugin C<virt>
6939 This plugin allows CPU, disk and network load to be collected for virtualized
6940 guests on the machine. This means that these metrics can be collected for guest
6941 systems without installing any software on them - I<collectd> only runs on the
6942 host system. The statistics are collected through libvirt
6943 (L<http://libvirt.org/>).
6945 Only I<Connection> is required.
6949 =item B<Connection> I<uri>
6951 Connect to the hypervisor given by I<uri>. For example if using Xen use:
6953 Connection "xen:///"
6955 Details which URIs allowed are given at L<http://libvirt.org/uri.html>.
6957 =item B<RefreshInterval> I<seconds>
6959 Refresh the list of domains and devices every I<seconds>. The default is 60
6960 seconds. Setting this to be the same or smaller than the I<Interval> will cause
6961 the list of domains and devices to be refreshed on every iteration.
6963 Refreshing the devices in particular is quite a costly operation, so if your
6964 virtualization setup is static you might consider increasing this. If this
6965 option is set to 0, refreshing is disabled completely.
6967 =item B<Domain> I<name>
6969 =item B<BlockDevice> I<name:dev>
6971 =item B<InterfaceDevice> I<name:dev>
6973 =item B<IgnoreSelected> B<true>|B<false>
6975 Select which domains and devices are collected.
6977 If I<IgnoreSelected> is not given or B<false> then only the listed domains and
6978 disk/network devices are collected.
6980 If I<IgnoreSelected> is B<true> then the test is reversed and the listed
6981 domains and disk/network devices are ignored, while the rest are collected.
6983 The domain name and device names may use a regular expression, if the name is
6984 surrounded by I</.../> and collectd was compiled with support for regexps.
6986 The default is to collect statistics for all domains and all their devices.
6990 BlockDevice "/:hdb/"
6991 IgnoreSelected "true"
6993 Ignore all I<hdb> devices on any domain, but other block devices (eg. I<hda>)
6996 =item B<HostnameFormat> B<name|uuid|hostname|...>
6998 When the virt plugin logs data, it sets the hostname of the collected data
6999 according to this setting. The default is to use the guest name as provided by
7000 the hypervisor, which is equal to setting B<name>.
7002 B<uuid> means use the guest's UUID. This is useful if you want to track the
7003 same guest across migrations.
7005 B<hostname> means to use the global B<Hostname> setting, which is probably not
7006 useful on its own because all guests will appear to have the same name.
7008 You can also specify combinations of these fields. For example B<name uuid>
7009 means to concatenate the guest name and UUID (with a literal colon character
7010 between, thus I<"foo:1234-1234-1234-1234">).
7012 =item B<InterfaceFormat> B<name>|B<address>
7014 When the virt plugin logs interface data, it sets the name of the collected
7015 data according to this setting. The default is to use the path as provided by
7016 the hypervisor (the "dev" property of the target node), which is equal to
7019 B<address> means use the interface's mac address. This is useful since the
7020 interface path might change between reboots of a guest or across migrations.
7022 =item B<PluginInstanceFormat> B<name|uuid>
7024 When the virt plugin logs data, it sets the plugin_instance of the collected
7025 data according to this setting. The default is to use the guest name as provided
7026 by the hypervisor, which is equal to setting B<name>.
7028 B<uuid> means use the guest's UUID.
7032 =head2 Plugin C<vmem>
7034 The C<vmem> plugin collects information about the usage of virtual memory.
7035 Since the statistics provided by the Linux kernel are very detailed, they are
7036 collected very detailed. However, to get all the details, you have to switch
7037 them on manually. Most people just want an overview over, such as the number of
7038 pages read from swap space.
7042 =item B<Verbose> B<true>|B<false>
7044 Enables verbose collection of information. This will start collecting page
7045 "actions", e.E<nbsp>g. page allocations, (de)activations, steals and so on.
7046 Part of these statistics are collected on a "per zone" basis.
7050 =head2 Plugin C<vserver>
7052 This plugin doesn't have any options. B<VServer> support is only available for
7053 Linux. It cannot yet be found in a vanilla kernel, though. To make use of this
7054 plugin you need a kernel that has B<VServer> support built in, i.E<nbsp>e. you
7055 need to apply the patches and compile your own kernel, which will then provide
7056 the F</proc/virtual> filesystem that is required by this plugin.
7058 The B<VServer> homepage can be found at L<http://linux-vserver.org/>.
7060 B<Note>: The traffic collected by this plugin accounts for the amount of
7061 traffic passing a socket which might be a lot less than the actual on-wire
7062 traffic (e.E<nbsp>g. due to headers and retransmission). If you want to
7063 collect on-wire traffic you could, for example, use the logging facilities of
7064 iptables to feed data for the guest IPs into the iptables plugin.
7066 =head2 Plugin C<write_graphite>
7068 The C<write_graphite> plugin writes data to I<Graphite>, an open-source metrics
7069 storage and graphing project. The plugin connects to I<Carbon>, the data layer
7070 of I<Graphite>, via I<TCP> or I<UDP> and sends data via the "line based"
7071 protocol (per default using portE<nbsp>2003). The data will be sent in blocks
7072 of at most 1428 bytes to minimize the number of network packets.
7076 <Plugin write_graphite>
7086 The configuration consists of one or more E<lt>B<Node>E<nbsp>I<Name>E<gt>
7087 blocks. Inside the B<Node> blocks, the following options are recognized:
7091 =item B<Host> I<Address>
7093 Hostname or address to connect to. Defaults to C<localhost>.
7095 =item B<Port> I<Service>
7097 Service name or port number to connect to. Defaults to C<2003>.
7099 =item B<Protocol> I<String>
7101 Protocol to use when connecting to I<Graphite>. Defaults to C<tcp>.
7103 =item B<LogSendErrors> B<false>|B<true>
7105 If set to B<true> (the default), logs errors when sending data to I<Graphite>.
7106 If set to B<false>, it will not log the errors. This is especially useful when
7107 using Protocol UDP since many times we want to use the "fire-and-forget"
7108 approach and logging errors fills syslog with unneeded messages.
7110 =item B<Prefix> I<String>
7112 When set, I<String> is added in front of the host name. Dots and whitespace are
7113 I<not> escaped in this string (see B<EscapeCharacter> below).
7115 =item B<Postfix> I<String>
7117 When set, I<String> is appended to the host name. Dots and whitespace are
7118 I<not> escaped in this string (see B<EscapeCharacter> below).
7120 =item B<EscapeCharacter> I<Char>
7122 I<Carbon> uses the dot (C<.>) as escape character and doesn't allow whitespace
7123 in the identifier. The B<EscapeCharacter> option determines which character
7124 dots, whitespace and control characters are replaced with. Defaults to
7127 =item B<StoreRates> B<false>|B<true>
7129 If set to B<true> (the default), convert counter values to rates. If set to
7130 B<false> counter values are stored as is, i.E<nbsp>e. as an increasing integer
7133 =item B<SeparateInstances> B<false>|B<true>
7135 If set to B<true>, the plugin instance and type instance will be in their own
7136 path component, for example C<host.cpu.0.cpu.idle>. If set to B<false> (the
7137 default), the plugin and plugin instance (and likewise the type and type
7138 instance) are put into one component, for example C<host.cpu-0.cpu-idle>.
7140 =item B<AlwaysAppendDS> B<false>|B<true>
7142 If set to B<true>, append the name of the I<Data Source> (DS) to the "metric"
7143 identifier. If set to B<false> (the default), this is only done when there is
7148 =head2 Plugin C<write_tsdb>
7150 The C<write_tsdb> plugin writes data to I<OpenTSDB>, a scalable open-source
7151 time series database. The plugin connects to a I<TSD>, a masterless, no shared
7152 state daemon that ingests metrics and stores them in HBase. The plugin uses
7153 I<TCP> over the "line based" protocol with a default port 4242. The data will
7154 be sent in blocks of at most 1428 bytes to minimize the number of network
7161 Host "tsd-1.my.domain"
7163 HostTags "status=production"
7167 The configuration consists of one or more E<lt>B<Node>E<nbsp>I<Name>E<gt>
7168 blocks. Inside the B<Node> blocks, the following options are recognized:
7172 =item B<Host> I<Address>
7174 Hostname or address to connect to. Defaults to C<localhost>.
7176 =item B<Port> I<Service>
7178 Service name or port number to connect to. Defaults to C<4242>.
7181 =item B<HostTags> I<String>
7183 When set, I<HostTags> is added to the end of the metric. It is intended to be
7184 used for name=value pairs that the TSD will tag the metric with. Dots and
7185 whitespace are I<not> escaped in this string.
7187 =item B<StoreRates> B<false>|B<true>
7189 If set to B<true>, convert counter values to rates. If set to B<false>
7190 (the default) counter values are stored as is, as an increasing
7193 =item B<AlwaysAppendDS> B<false>|B<true>
7195 If set the B<true>, append the name of the I<Data Source> (DS) to the "metric"
7196 identifier. If set to B<false> (the default), this is only done when there is
7201 =head2 Plugin C<write_mongodb>
7203 The I<write_mongodb plugin> will send values to I<MongoDB>, a schema-less
7208 <Plugin "write_mongodb">
7217 The plugin can send values to multiple instances of I<MongoDB> by specifying
7218 one B<Node> block for each instance. Within the B<Node> blocks, the following
7219 options are available:
7223 =item B<Host> I<Address>
7225 Hostname or address to connect to. Defaults to C<localhost>.
7227 =item B<Port> I<Service>
7229 Service name or port number to connect to. Defaults to C<27017>.
7231 =item B<Timeout> I<Milliseconds>
7233 Set the timeout for each operation on I<MongoDB> to I<Timeout> milliseconds.
7234 Setting this option to zero means no timeout, which is the default.
7236 =item B<StoreRates> B<false>|B<true>
7238 If set to B<true> (the default), convert counter values to rates. If set to
7239 B<false> counter values are stored as is, i.e. as an increasing integer
7242 =item B<Database> I<Database>
7244 =item B<User> I<User>
7246 =item B<Password> I<Password>
7248 Sets the information used when authenticating to a I<MongoDB> database. The
7249 fields are optional (in which case no authentication is attempted), but if you
7250 want to use authentication all three fields must be set.
7254 =head2 Plugin C<write_http>
7256 This output plugin submits values to an HTTP server using POST requests and
7257 encoding metrics with JSON or using the C<PUTVAL> command described in
7258 L<collectd-unixsock(5)>.
7262 <Plugin "write_http">
7264 URL "http://example.com/post-collectd"
7271 The plugin can send values to multiple HTTP servers by specifying one
7272 E<lt>B<Node>E<nbsp>I<Name>E<gt> block for each server. Within each B<Node>
7273 block, the following options are available:
7279 URL to which the values are submitted to. Mandatory.
7281 =item B<User> I<Username>
7283 Optional user name needed for authentication.
7285 =item B<Password> I<Password>
7287 Optional password needed for authentication.
7289 =item B<VerifyPeer> B<true>|B<false>
7291 Enable or disable peer SSL certificate verification. See
7292 L<http://curl.haxx.se/docs/sslcerts.html> for details. Enabled by default.
7294 =item B<VerifyHost> B<true|false>
7296 Enable or disable peer host name verification. If enabled, the plugin checks if
7297 the C<Common Name> or a C<Subject Alternate Name> field of the SSL certificate
7298 matches the host name provided by the B<URL> option. If this identity check
7299 fails, the connection is aborted. Obviously, only works when connecting to a
7300 SSL enabled server. Enabled by default.
7302 =item B<CACert> I<File>
7304 File that holds one or more SSL certificates. If you want to use HTTPS you will
7305 possibly need this option. What CA certificates come bundled with C<libcurl>
7306 and are checked by default depends on the distribution you use.
7308 =item B<CAPath> I<Directory>
7310 Directory holding one or more CA certificate files. You can use this if for
7311 some reason all the needed CA certificates aren't in the same file and can't be
7312 pointed to using the B<CACert> option. Requires C<libcurl> to be built against
7315 =item B<ClientKey> I<File>
7317 File that holds the private key in PEM format to be used for certificate-based
7320 =item B<ClientCert> I<File>
7322 File that holds the SSL certificate to be used for certificate-based
7325 =item B<ClientKeyPass> I<Password>
7327 Password required to load the private key in B<ClientKey>.
7329 =item B<SSLVersion> B<SSLv2>|B<SSLv3>|B<TLSv1>|B<TLSv1_0>|B<TLSv1_1>|B<TLSv1_2>
7331 Define which SSL protocol version must be used. By default C<libcurl> will
7332 attempt to figure out the remote SSL protocol version. See
7333 L<curl_easy_setopt(3)> for more details.
7335 =item B<Format> B<Command>|B<JSON>
7337 Format of the output to generate. If set to B<Command>, will create output that
7338 is understood by the I<Exec> and I<UnixSock> plugins. When set to B<JSON>, will
7339 create output in the I<JavaScript Object Notation> (JSON).
7341 Defaults to B<Command>.
7343 =item B<StoreRates> B<true|false>
7345 If set to B<true>, convert counter values to rates. If set to B<false> (the
7346 default) counter values are stored as is, i.e. as an increasing integer number.
7348 =item B<BufferSize> I<Bytes>
7350 Sets the send buffer size to I<Bytes>. By increasing this buffer, less HTTP
7351 requests will be generated, but more metrics will be batched / metrics are
7352 cached for longer before being sent, introducing additional delay until they
7353 are available on the server side. I<Bytes> must be at least 1024 and cannot
7354 exceed the size of an C<int>, i.e. 2E<nbsp>GByte.
7355 Defaults to C<4096>.
7357 =item B<LowSpeedLimit> I<Bytes per Second>
7359 Sets the minimal transfer rate in I<Bytes per Second> below which the
7360 connection with the HTTP server will be considered too slow and aborted. All
7361 the data submitted over this connection will probably be lost. Defaults to 0,
7362 which means no minimum transfer rate is enforced.
7364 =item B<Timeout> I<Timeout>
7366 Sets the maximum time in milliseconds given for HTTP POST operations to
7367 complete. When this limit is reached, the POST operation will be aborted, and
7368 all the data in the current send buffer will probably be lost. Defaults to 0,
7369 which means the connection never times out.
7371 The C<write_http> plugin regularly submits the collected values to the HTTP
7372 server. How frequently this happens depends on how much data you are collecting
7373 and the size of B<BufferSize>. The optimal value to set B<Timeout> to is
7374 slightly below this interval, which you can estimate by monitoring the network
7375 traffic between collectd and the HTTP server.
7379 =head2 Plugin C<write_kafka>
7381 The I<write_kafka plugin> will send values to a I<Kafka> topic, a distributed
7385 <Plugin "write_kafka">
7386 Property "metadata.broker.list" "broker1:9092,broker2:9092"
7392 The following options are understood by the I<write_kafka plugin>:
7396 =item E<lt>B<Topic> I<Name>E<gt>
7398 The plugin's configuration consists of one or more B<Topic> blocks. Each block
7399 is given a unique I<Name> and specifies one kafka producer.
7400 Inside the B<Topic> block, the following per-topic options are
7405 =item B<Property> I<String> I<String>
7407 Configure the named property for the current topic. Properties are
7408 forwarded to the kafka producer library B<librdkafka>.
7410 =item B<Key> I<String>
7412 Use the specified string as a partioning key for the topic. Kafka breaks
7413 topic into partitions and guarantees that for a given topology, the same
7414 consumer will be used for a specific key. The special (case insensitive)
7415 string B<Random> can be used to specify that an arbitrary partition should
7418 =item B<Format> B<Command>|B<JSON>|B<Graphite>
7420 Selects the format in which messages are sent to the broker. If set to
7421 B<Command> (the default), values are sent as C<PUTVAL> commands which are
7422 identical to the syntax used by the I<Exec> and I<UnixSock plugins>.
7424 If set to B<JSON>, the values are encoded in the I<JavaScript Object Notation>,
7425 an easy and straight forward exchange format.
7427 If set to B<Graphite>, values are encoded in the I<Graphite> format, which is
7428 C<E<lt>metricE<gt> E<lt>valueE<gt> E<lt>timestampE<gt>\n>.
7430 =item B<StoreRates> B<true>|B<false>
7432 Determines whether or not C<COUNTER>, C<DERIVE> and C<ABSOLUTE> data sources
7433 are converted to a I<rate> (i.e. a C<GAUGE> value). If set to B<false> (the
7434 default), no conversion is performed. Otherwise the conversion is performed
7435 using the internal value cache.
7437 Please note that currently this option is only used if the B<Format> option has
7438 been set to B<JSON>.
7440 =item B<GraphitePrefix> (B<Format>=I<Graphite> only)
7442 A prefix can be added in the metric name when outputting in the I<Graphite>
7443 format. It's added before the I<Host> name.
7445 C<E<lt>prefixE<gt>E<lt>hostE<gt>E<lt>postfixE<gt>E<lt>pluginE<gt>E<lt>typeE<gt>E<lt>nameE<gt>>
7447 =item B<GraphitePostfix> (B<Format>=I<Graphite> only)
7449 A postfix can be added in the metric name when outputting in the I<Graphite>
7450 format. It's added after the I<Host> name.
7452 C<E<lt>prefixE<gt>E<lt>hostE<gt>E<lt>postfixE<gt>E<lt>pluginE<gt>E<lt>typeE<gt>E<lt>nameE<gt>>
7454 =item B<GraphiteEscapeChar> (B<Format>=I<Graphite> only)
7456 Specify a character to replace dots (.) in the host part of the metric name.
7457 In I<Graphite> metric name, dots are used as separators between different
7458 metric parts (host, plugin, type).
7459 Default is C<_> (I<Underscore>).
7461 =item B<GraphiteSeparateInstances> B<false>|B<true>
7463 If set to B<true>, the plugin instance and type instance will be in their own
7464 path component, for example C<host.cpu.0.cpu.idle>. If set to B<false> (the
7465 default), the plugin and plugin instance (and likewise the type and type
7466 instance) are put into one component, for example C<host.cpu-0.cpu-idle>.
7468 =item B<StoreRates> B<true>|B<false>
7470 If set to B<true> (the default), convert counter values to rates. If set to
7471 B<false> counter values are stored as is, i.e. as an increasing integer number.
7473 This will be reflected in the C<ds_type> tag: If B<StoreRates> is enabled,
7474 converted values will have "rate" appended to the data source type, e.g.
7475 C<ds_type:derive:rate>.
7479 =item B<Property> I<String> I<String>
7481 Configure the kafka producer through properties, you almost always will
7482 want to set B<metadata.broker.list> to your Kafka broker list.
7486 =head2 Plugin C<write_redis>
7488 The I<write_redis plugin> submits values to I<Redis>, a data structure server.
7492 <Plugin "write_redis">
7501 Values are submitted to I<Sorted Sets>, using the metric name as the key, and
7502 the timestamp as the score. Retrieving a date range can then be done using the
7503 C<ZRANGEBYSCORE> I<Redis> command. Additionnally, all the identifiers of these
7504 I<Sorted Sets> are kept in a I<Set> called C<collectd/values> (or
7505 C<${prefix}/values> if the B<Prefix> option was specified) and can be retrieved
7506 using the C<SMEMBERS> I<Redis> command. See
7507 L<http://redis.io/commands#sorted_set> and L<http://redis.io/commands#set> for
7510 The information shown in the synopsis above is the I<default configuration>
7511 which is used by the plugin if no configuration is present.
7513 The plugin can send values to multiple instances of I<Redis> by specifying
7514 one B<Node> block for each instance. Within the B<Node> blocks, the following
7515 options are available:
7519 =item B<Node> I<Nodename>
7521 The B<Node> block identifies a new I<Redis> node, that is a new I<Redis>
7522 instance running in an specified host and port. The name for node is a
7523 canonical identifier which is used as I<plugin instance>. It is limited to
7524 64E<nbsp>characters in length.
7526 =item B<Host> I<Hostname>
7528 The B<Host> option is the hostname or IP-address where the I<Redis> instance is
7531 =item B<Port> I<Port>
7533 The B<Port> option is the TCP port on which the Redis instance accepts
7534 connections. Either a service name of a port number may be given. Please note
7535 that numerical port numbers must be given as a string, too.
7537 =item B<Timeout> I<Milliseconds>
7539 The B<Timeout> option sets the socket connection timeout, in milliseconds.
7541 =item B<Prefix> I<Prefix>
7543 Prefix used when constructing the name of the I<Sorted Sets> and the I<Set>
7544 containing all metrics. Defaults to C<collectd/>, so metrics will have names
7545 like C<collectd/cpu-0/cpu-user>. When setting this to something different, it
7546 is recommended but not required to include a trailing slash in I<Prefix>.
7550 =head2 Plugin C<write_riemann>
7552 The I<write_riemann plugin> will send values to I<Riemann>, a powerful stream
7553 aggregation and monitoring system. The plugin sends I<Protobuf> encoded data to
7554 I<Riemann> using UDP packets.
7558 <Plugin "write_riemann">
7564 AlwaysAppendDS false
7568 Attribute "foo" "bar"
7571 The following options are understood by the I<write_riemann plugin>:
7575 =item E<lt>B<Node> I<Name>E<gt>
7577 The plugin's configuration consists of one or more B<Node> blocks. Each block
7578 is given a unique I<Name> and specifies one connection to an instance of
7579 I<Riemann>. Indise the B<Node> block, the following per-connection options are
7584 =item B<Host> I<Address>
7586 Hostname or address to connect to. Defaults to C<localhost>.
7588 =item B<Port> I<Service>
7590 Service name or port number to connect to. Defaults to C<5555>.
7592 =item B<Protocol> B<UDP>|B<TCP>
7594 Specify the protocol to use when communicating with I<Riemann>. Defaults to
7597 =item B<Batch> B<true>|B<false>
7599 If set to B<true> and B<Protocol> is set to B<TCP>,
7600 events will be batched in memory and flushed at
7601 regular intervals or when B<BatchMaxSize> is exceeded.
7603 Notifications are not batched and sent as soon as possible.
7605 When enabled, it can occur that events get processed by the Riemann server
7606 close to or after their expiration time. Tune the B<TTLFactor> and
7607 B<BatchMaxSize> settings according to the amount of values collected, if this
7612 =item B<BatchMaxSize> I<size>
7614 Maximum payload size for a riemann packet. Defaults to 8192
7616 =item B<StoreRates> B<true>|B<false>
7618 If set to B<true> (the default), convert counter values to rates. If set to
7619 B<false> counter values are stored as is, i.e. as an increasing integer number.
7621 This will be reflected in the C<ds_type> tag: If B<StoreRates> is enabled,
7622 converted values will have "rate" appended to the data source type, e.g.
7623 C<ds_type:derive:rate>.
7625 =item B<AlwaysAppendDS> B<false>|B<true>
7627 If set the B<true>, append the name of the I<Data Source> (DS) to the
7628 "service", i.e. the field that, together with the "host" field, uniquely
7629 identifies a metric in I<Riemann>. If set to B<false> (the default), this is
7630 only done when there is more than one DS.
7632 =item B<TTLFactor> I<Factor>
7634 I<Riemann> events have a I<Time to Live> (TTL) which specifies how long each
7635 event is considered active. I<collectd> populates this field based on the
7636 metrics interval setting. This setting controls the factor with which the
7637 interval is multiplied to set the TTL. The default value is B<2.0>. Unless you
7638 know exactly what you're doing, you should only increase this setting from its
7641 =item B<Notifications> B<false>|B<true>
7643 If set to B<true>, create riemann events for notifications. This is B<true>
7644 by default. When processing thresholds from write_riemann, it might prove
7645 useful to avoid getting notification events.
7647 =item B<CheckThresholds> B<false>|B<true>
7649 If set to B<true>, attach state to events based on thresholds defined
7650 in the B<Threshold> plugin. Defaults to B<false>.
7652 =item B<EventServicePrefix> I<String>
7654 Add the given string as a prefix to the event service name.
7655 If B<EventServicePrefix> not set or set to an empty string (""),
7656 no prefix will be used.
7660 =item B<Tag> I<String>
7662 Add the given string as an additional tag to the metric being sent to
7665 =item B<Attribute> I<String> I<String>
7667 Consider the two given strings to be the key and value of an additional
7668 attribute for each metric being sent out to I<Riemann>.
7672 =head2 Plugin C<write_sensu>
7674 The I<write_sensu plugin> will send values to I<Sensu>, a powerful stream
7675 aggregation and monitoring system. The plugin sends I<JSON> encoded data to
7676 a local I<Sensu> client using a TCP socket.
7678 At the moment, the I<write_sensu plugin> does not send over a collectd_host
7679 parameter so it is not possible to use one collectd instance as a gateway for
7680 others. Each collectd host must pair with one I<Sensu> client.
7684 <Plugin "write_sensu">
7689 AlwaysAppendDS false
7690 MetricHandler "influx"
7691 MetricHandler "default"
7692 NotificationHandler "flapjack"
7693 NotificationHandler "howling_monkey"
7697 Attribute "foo" "bar"
7700 The following options are understood by the I<write_sensu plugin>:
7704 =item E<lt>B<Node> I<Name>E<gt>
7706 The plugin's configuration consists of one or more B<Node> blocks. Each block
7707 is given a unique I<Name> and specifies one connection to an instance of
7708 I<Sensu>. Inside the B<Node> block, the following per-connection options are
7713 =item B<Host> I<Address>
7715 Hostname or address to connect to. Defaults to C<localhost>.
7717 =item B<Port> I<Service>
7719 Service name or port number to connect to. Defaults to C<3030>.
7721 =item B<StoreRates> B<true>|B<false>
7723 If set to B<true> (the default), convert counter values to rates. If set to
7724 B<false> counter values are stored as is, i.e. as an increasing integer number.
7726 This will be reflected in the C<collectd_data_source_type> tag: If
7727 B<StoreRates> is enabled, converted values will have "rate" appended to the
7728 data source type, e.g. C<collectd_data_source_type:derive:rate>.
7730 =item B<AlwaysAppendDS> B<false>|B<true>
7732 If set the B<true>, append the name of the I<Data Source> (DS) to the
7733 "service", i.e. the field that, together with the "host" field, uniquely
7734 identifies a metric in I<Sensu>. If set to B<false> (the default), this is
7735 only done when there is more than one DS.
7737 =item B<Notifications> B<false>|B<true>
7739 If set to B<true>, create I<Sensu> events for notifications. This is B<false>
7740 by default. At least one of B<Notifications> or B<Metrics> should be enabled.
7742 =item B<Metrics> B<false>|B<true>
7744 If set to B<true>, create I<Sensu> events for metrics. This is B<false>
7745 by default. At least one of B<Notifications> or B<Metrics> should be enabled.
7748 =item B<Separator> I<String>
7750 Sets the separator for I<Sensu> metrics name or checks. Defaults to "/".
7752 =item B<MetricHandler> I<String>
7754 Add a handler that will be set when metrics are sent to I<Sensu>. You can add
7755 several of them, one per line. Defaults to no handler.
7757 =item B<NotificationHandler> I<String>
7759 Add a handler that will be set when notifications are sent to I<Sensu>. You can
7760 add several of them, one per line. Defaults to no handler.
7762 =item B<EventServicePrefix> I<String>
7764 Add the given string as a prefix to the event service name.
7765 If B<EventServicePrefix> not set or set to an empty string (""),
7766 no prefix will be used.
7770 =item B<Tag> I<String>
7772 Add the given string as an additional tag to the metric being sent to
7775 =item B<Attribute> I<String> I<String>
7777 Consider the two given strings to be the key and value of an additional
7778 attribute for each metric being sent out to I<Sensu>.
7782 =head2 Plugin C<zookeeper>
7784 The I<zookeeper plugin> will collect statistics from a I<Zookeeper> server
7785 using the mntr command. It requires Zookeeper 3.4.0+ and access to the
7790 <Plugin "zookeeper">
7797 =item B<Host> I<Address>
7799 Hostname or address to connect to. Defaults to C<localhost>.
7801 =item B<Port> I<Service>
7803 Service name or port number to connect to. Defaults to C<2181>.
7807 =head1 THRESHOLD CONFIGURATION
7809 Starting with version C<4.3.0> collectd has support for B<monitoring>. By that
7810 we mean that the values are not only stored or sent somewhere, but that they
7811 are judged and, if a problem is recognized, acted upon. The only action
7812 collectd takes itself is to generate and dispatch a "notification". Plugins can
7813 register to receive notifications and perform appropriate further actions.
7815 Since systems and what you expect them to do differ a lot, you can configure
7816 B<thresholds> for your values freely. This gives you a lot of flexibility but
7817 also a lot of responsibility.
7819 Every time a value is out of range a notification is dispatched. This means
7820 that the idle percentage of your CPU needs to be less then the configured
7821 threshold only once for a notification to be generated. There's no such thing
7822 as a moving average or similar - at least not now.
7824 Also, all values that match a threshold are considered to be relevant or
7825 "interesting". As a consequence collectd will issue a notification if they are
7826 not received for B<Timeout> iterations. The B<Timeout> configuration option is
7827 explained in section L<"GLOBAL OPTIONS">. If, for example, B<Timeout> is set to
7828 "2" (the default) and some hosts sends it's CPU statistics to the server every
7829 60 seconds, a notification will be dispatched after about 120 seconds. It may
7830 take a little longer because the timeout is checked only once each B<Interval>
7833 When a value comes within range again or is received after it was missing, an
7834 "OKAY-notification" is dispatched.
7836 Here is a configuration example to get you started. Read below for more
7849 <Plugin "interface">
7866 WarningMin 100000000
7872 There are basically two types of configuration statements: The C<Host>,
7873 C<Plugin>, and C<Type> blocks select the value for which a threshold should be
7874 configured. The C<Plugin> and C<Type> blocks may be specified further using the
7875 C<Instance> option. You can combine the block by nesting the blocks, though
7876 they must be nested in the above order, i.E<nbsp>e. C<Host> may contain either
7877 C<Plugin> and C<Type> blocks, C<Plugin> may only contain C<Type> blocks and
7878 C<Type> may not contain other blocks. If multiple blocks apply to the same
7879 value the most specific block is used.
7881 The other statements specify the threshold to configure. They B<must> be
7882 included in a C<Type> block. Currently the following statements are recognized:
7886 =item B<FailureMax> I<Value>
7888 =item B<WarningMax> I<Value>
7890 Sets the upper bound of acceptable values. If unset defaults to positive
7891 infinity. If a value is greater than B<FailureMax> a B<FAILURE> notification
7892 will be created. If the value is greater than B<WarningMax> but less than (or
7893 equal to) B<FailureMax> a B<WARNING> notification will be created.
7895 =item B<FailureMin> I<Value>
7897 =item B<WarningMin> I<Value>
7899 Sets the lower bound of acceptable values. If unset defaults to negative
7900 infinity. If a value is less than B<FailureMin> a B<FAILURE> notification will
7901 be created. If the value is less than B<WarningMin> but greater than (or equal
7902 to) B<FailureMin> a B<WARNING> notification will be created.
7904 =item B<DataSource> I<DSName>
7906 Some data sets have more than one "data source". Interesting examples are the
7907 C<if_octets> data set, which has received (C<rx>) and sent (C<tx>) bytes and
7908 the C<disk_ops> data set, which holds C<read> and C<write> operations. The
7909 system load data set, C<load>, even has three data sources: C<shortterm>,
7910 C<midterm>, and C<longterm>.
7912 Normally, all data sources are checked against a configured threshold. If this
7913 is undesirable, or if you want to specify different limits for each data
7914 source, you can use the B<DataSource> option to have a threshold apply only to
7917 =item B<Invert> B<true>|B<false>
7919 If set to B<true> the range of acceptable values is inverted, i.E<nbsp>e.
7920 values between B<FailureMin> and B<FailureMax> (B<WarningMin> and
7921 B<WarningMax>) are not okay. Defaults to B<false>.
7923 =item B<Persist> B<true>|B<false>
7925 Sets how often notifications are generated. If set to B<true> one notification
7926 will be generated for each value that is out of the acceptable range. If set to
7927 B<false> (the default) then a notification is only generated if a value is out
7928 of range but the previous value was okay.
7930 This applies to missing values, too: If set to B<true> a notification about a
7931 missing value is generated once every B<Interval> seconds. If set to B<false>
7932 only one such notification is generated until the value appears again.
7934 =item B<Percentage> B<true>|B<false>
7936 If set to B<true>, the minimum and maximum values given are interpreted as
7937 percentage value, relative to the other data sources. This is helpful for
7938 example for the "df" type, where you may want to issue a warning when less than
7939 5E<nbsp>% of the total space is available. Defaults to B<false>.
7941 =item B<Hits> I<Number>
7943 Delay creating the notification until the threshold has been passed I<Number>
7944 times. When a notification has been generated, or when a subsequent value is
7945 inside the threshold, the counter is reset. If, for example, a value is
7946 collected once every 10E<nbsp>seconds and B<Hits> is set to 3, a notification
7947 will be dispatched at most once every 30E<nbsp>seconds.
7949 This is useful when short bursts are not a problem. If, for example, 100% CPU
7950 usage for up to a minute is normal (and data is collected every
7951 10E<nbsp>seconds), you could set B<Hits> to B<6> to account for this.
7953 =item B<Hysteresis> I<Number>
7955 When set to non-zero, a hysteresis value is applied when checking minimum and
7956 maximum bounds. This is useful for values that increase slowly and fluctuate a
7957 bit while doing so. When these values come close to the threshold, they may
7958 "flap", i.e. switch between failure / warning case and okay case repeatedly.
7960 If, for example, the threshold is configures as
7965 then a I<Warning> notification is created when the value exceeds I<101> and the
7966 corresponding I<Okay> notification is only created once the value falls below
7967 I<99>, thus avoiding the "flapping".
7971 =head1 FILTER CONFIGURATION
7973 Starting with collectd 4.6 there is a powerful filtering infrastructure
7974 implemented in the daemon. The concept has mostly been copied from
7975 I<ip_tables>, the packet filter infrastructure for Linux. We'll use a similar
7976 terminology, so that users that are familiar with iptables feel right at home.
7980 The following are the terms used in the remainder of the filter configuration
7981 documentation. For an ASCII-art schema of the mechanism, see
7982 L<"General structure"> below.
7988 A I<match> is a criteria to select specific values. Examples are, of course, the
7989 name of the value or it's current value.
7991 Matches are implemented in plugins which you have to load prior to using the
7992 match. The name of such plugins starts with the "match_" prefix.
7996 A I<target> is some action that is to be performed with data. Such actions
7997 could, for example, be to change part of the value's identifier or to ignore
7998 the value completely.
8000 Some of these targets are built into the daemon, see L<"Built-in targets">
8001 below. Other targets are implemented in plugins which you have to load prior to
8002 using the target. The name of such plugins starts with the "target_" prefix.
8006 The combination of any number of matches and at least one target is called a
8007 I<rule>. The target actions will be performed for all values for which B<all>
8008 matches apply. If the rule does not have any matches associated with it, the
8009 target action will be performed for all values.
8013 A I<chain> is a list of rules and possibly default targets. The rules are tried
8014 in order and if one matches, the associated target will be called. If a value
8015 is handled by a rule, it depends on the target whether or not any subsequent
8016 rules are considered or if traversal of the chain is aborted, see
8017 L<"Flow control"> below. After all rules have been checked, the default targets
8022 =head2 General structure
8024 The following shows the resulting structure:
8031 +---------+ +---------+ +---------+ +---------+
8032 ! Rule !->! Match !->! Match !->! Target !
8033 +---------+ +---------+ +---------+ +---------+
8036 +---------+ +---------+ +---------+
8037 ! Rule !->! Target !->! Target !
8038 +---------+ +---------+ +---------+
8045 +---------+ +---------+ +---------+
8046 ! Rule !->! Match !->! Target !
8047 +---------+ +---------+ +---------+
8057 There are four ways to control which way a value takes through the filter
8064 The built-in B<jump> target can be used to "call" another chain, i.E<nbsp>e.
8065 process the value with another chain. When the called chain finishes, usually
8066 the next target or rule after the jump is executed.
8070 The stop condition, signaled for example by the built-in target B<stop>, causes
8071 all processing of the value to be stopped immediately.
8075 Causes processing in the current chain to be aborted, but processing of the
8076 value generally will continue. This means that if the chain was called via
8077 B<Jump>, the next target or rule after the jump will be executed. If the chain
8078 was not called by another chain, control will be returned to the daemon and it
8079 may pass the value to another chain.
8083 Most targets will signal the B<continue> condition, meaning that processing
8084 should continue normally. There is no special built-in target for this
8091 The configuration reflects this structure directly:
8093 PostCacheChain "PostCache"
8095 <Rule "ignore_mysql_show">
8098 Type "^mysql_command$"
8099 TypeInstance "^show_"
8109 The above configuration example will ignore all values where the plugin field
8110 is "mysql", the type is "mysql_command" and the type instance begins with
8111 "show_". All other values will be sent to the C<rrdtool> write plugin via the
8112 default target of the chain. Since this chain is run after the value has been
8113 added to the cache, the MySQL C<show_*> command statistics will be available
8114 via the C<unixsock> plugin.
8116 =head2 List of configuration options
8120 =item B<PreCacheChain> I<ChainName>
8122 =item B<PostCacheChain> I<ChainName>
8124 Configure the name of the "pre-cache chain" and the "post-cache chain". The
8125 argument is the name of a I<chain> that should be executed before and/or after
8126 the values have been added to the cache.
8128 To understand the implications, it's important you know what is going on inside
8129 I<collectd>. The following diagram shows how values are passed from the
8130 read-plugins to the write-plugins:
8136 + - - - - V - - - - +
8137 : +---------------+ :
8140 : +-------+-------+ :
8143 : +-------+-------+ : +---------------+
8144 : ! Cache !--->! Value Cache !
8145 : ! insert ! : +---+---+-------+
8146 : +-------+-------+ : ! !
8147 : ! ,------------' !
8149 : +-------+---+---+ : +-------+-------+
8150 : ! Post-Cache +--->! Write-Plugins !
8151 : ! Chain ! : +---------------+
8152 : +---------------+ :
8155 + - - - - - - - - - +
8157 After the values are passed from the "read" plugins to the dispatch functions,
8158 the pre-cache chain is run first. The values are added to the internal cache
8159 afterwards. The post-cache chain is run after the values have been added to the
8160 cache. So why is it such a huge deal if chains are run before or after the
8161 values have been added to this cache?
8163 Targets that change the identifier of a value list should be executed before
8164 the values are added to the cache, so that the name in the cache matches the
8165 name that is used in the "write" plugins. The C<unixsock> plugin, too, uses
8166 this cache to receive a list of all available values. If you change the
8167 identifier after the value list has been added to the cache, this may easily
8168 lead to confusion, but it's not forbidden of course.
8170 The cache is also used to convert counter values to rates. These rates are, for
8171 example, used by the C<value> match (see below). If you use the rate stored in
8172 the cache B<before> the new value is added, you will use the old, B<previous>
8173 rate. Write plugins may use this rate, too, see the C<csv> plugin, for example.
8174 The C<unixsock> plugin uses these rates too, to implement the C<GETVAL>
8177 Last but not last, the B<stop> target makes a difference: If the pre-cache
8178 chain returns the stop condition, the value will not be added to the cache and
8179 the post-cache chain will not be run.
8181 =item B<Chain> I<Name>
8183 Adds a new chain with a certain name. This name can be used to refer to a
8184 specific chain, for example to jump to it.
8186 Within the B<Chain> block, there can be B<Rule> blocks and B<Target> blocks.
8188 =item B<Rule> [I<Name>]
8190 Adds a new rule to the current chain. The name of the rule is optional and
8191 currently has no meaning for the daemon.
8193 Within the B<Rule> block, there may be any number of B<Match> blocks and there
8194 must be at least one B<Target> block.
8196 =item B<Match> I<Name>
8198 Adds a match to a B<Rule> block. The name specifies what kind of match should
8199 be performed. Available matches depend on the plugins that have been loaded.
8201 The arguments inside the B<Match> block are passed to the plugin implementing
8202 the match, so which arguments are valid here depends on the plugin being used.
8203 If you do not need any to pass any arguments to a match, you can use the
8208 Which is equivalent to:
8213 =item B<Target> I<Name>
8215 Add a target to a rule or a default target to a chain. The name specifies what
8216 kind of target is to be added. Which targets are available depends on the
8217 plugins being loaded.
8219 The arguments inside the B<Target> block are passed to the plugin implementing
8220 the target, so which arguments are valid here depends on the plugin being used.
8221 If you do not need any to pass any arguments to a target, you can use the
8226 This is the same as writing:
8233 =head2 Built-in targets
8235 The following targets are built into the core daemon and therefore need no
8236 plugins to be loaded:
8242 Signals the "return" condition, see the L<"Flow control"> section above. This
8243 causes the current chain to stop processing the value and returns control to
8244 the calling chain. The calling chain will continue processing targets and rules
8245 just after the B<jump> target (see below). This is very similar to the
8246 B<RETURN> target of iptables, see L<iptables(8)>.
8248 This target does not have any options.
8256 Signals the "stop" condition, see the L<"Flow control"> section above. This
8257 causes processing of the value to be aborted immediately. This is similar to
8258 the B<DROP> target of iptables, see L<iptables(8)>.
8260 This target does not have any options.
8268 Sends the value to "write" plugins.
8274 =item B<Plugin> I<Name>
8276 Name of the write plugin to which the data should be sent. This option may be
8277 given multiple times to send the data to more than one write plugin. If the
8278 plugin supports multiple instances, the plugin's instance(s) must also be
8283 If no plugin is explicitly specified, the values will be sent to all available
8286 Single-instance plugin example:
8292 Multi-instance plugin example:
8294 <Plugin "write_graphite">
8304 Plugin "write_graphite/foo"
8309 Starts processing the rules of another chain, see L<"Flow control"> above. If
8310 the end of that chain is reached, or a stop condition is encountered,
8311 processing will continue right after the B<jump> target, i.E<nbsp>e. with the
8312 next target or the next rule. This is similar to the B<-j> command line option
8313 of iptables, see L<iptables(8)>.
8319 =item B<Chain> I<Name>
8321 Jumps to the chain I<Name>. This argument is required and may appear only once.
8333 =head2 Available matches
8339 Matches a value using regular expressions.
8345 =item B<Host> I<Regex>
8347 =item B<Plugin> I<Regex>
8349 =item B<PluginInstance> I<Regex>
8351 =item B<Type> I<Regex>
8353 =item B<TypeInstance> I<Regex>
8355 Match values where the given regular expressions match the various fields of
8356 the identifier of a value. If multiple regular expressions are given, B<all>
8357 regexen must match for a value to match.
8359 =item B<Invert> B<false>|B<true>
8361 When set to B<true>, the result of the match is inverted, i.e. all value lists
8362 where all regular expressions apply are not matched, all other value lists are
8363 matched. Defaults to B<false>.
8370 Host "customer[0-9]+"
8376 Matches values that have a time which differs from the time on the server.
8378 This match is mainly intended for servers that receive values over the
8379 C<network> plugin and write them to disk using the C<rrdtool> plugin. RRDtool
8380 is very sensitive to the timestamp used when updating the RRD files. In
8381 particular, the time must be ever increasing. If a misbehaving client sends one
8382 packet with a timestamp far in the future, all further packets with a correct
8383 time will be ignored because of that one packet. What's worse, such corrupted
8384 RRD files are hard to fix.
8386 This match lets one match all values B<outside> a specified time range
8387 (relative to the server's time), so you can use the B<stop> target (see below)
8388 to ignore the value, for example.
8394 =item B<Future> I<Seconds>
8396 Matches all values that are I<ahead> of the server's time by I<Seconds> or more
8397 seconds. Set to zero for no limit. Either B<Future> or B<Past> must be
8400 =item B<Past> I<Seconds>
8402 Matches all values that are I<behind> of the server's time by I<Seconds> or
8403 more seconds. Set to zero for no limit. Either B<Future> or B<Past> must be
8415 This example matches all values that are five minutes or more ahead of the
8416 server or one hour (or more) lagging behind.
8420 Matches the actual value of data sources against given minimumE<nbsp>/ maximum
8421 values. If a data-set consists of more than one data-source, all data-sources
8422 must match the specified ranges for a positive match.
8428 =item B<Min> I<Value>
8430 Sets the smallest value which still results in a match. If unset, behaves like
8433 =item B<Max> I<Value>
8435 Sets the largest value which still results in a match. If unset, behaves like
8438 =item B<Invert> B<true>|B<false>
8440 Inverts the selection. If the B<Min> and B<Max> settings result in a match,
8441 no-match is returned and vice versa. Please note that the B<Invert> setting
8442 only effects how B<Min> and B<Max> are applied to a specific value. Especially
8443 the B<DataSource> and B<Satisfy> settings (see below) are not inverted.
8445 =item B<DataSource> I<DSName> [I<DSName> ...]
8447 Select one or more of the data sources. If no data source is configured, all
8448 data sources will be checked. If the type handled by the match does not have a
8449 data source of the specified name(s), this will always result in no match
8450 (independent of the B<Invert> setting).
8452 =item B<Satisfy> B<Any>|B<All>
8454 Specifies how checking with several data sources is performed. If set to
8455 B<Any>, the match succeeds if one of the data sources is in the configured
8456 range. If set to B<All> the match only succeeds if all data sources are within
8457 the configured range. Default is B<All>.
8459 Usually B<All> is used for positive matches, B<Any> is used for negative
8460 matches. This means that with B<All> you usually check that all values are in a
8461 "good" range, while with B<Any> you check if any value is within a "bad" range
8462 (or outside the "good" range).
8466 Either B<Min> or B<Max>, but not both, may be unset.
8470 # Match all values smaller than or equal to 100. Matches only if all data
8471 # sources are below 100.
8477 # Match if the value of any data source is outside the range of 0 - 100.
8485 =item B<empty_counter>
8487 Matches all values with one or more data sources of type B<COUNTER> and where
8488 all counter values are zero. These counters usually I<never> increased since
8489 they started existing (and are therefore uninteresting), or got reset recently
8490 or overflowed and you had really, I<really> bad luck.
8492 Please keep in mind that ignoring such counters can result in confusing
8493 behavior: Counters which hardly ever increase will be zero for long periods of
8494 time. If the counter is reset for some reason (machine or service restarted,
8495 usually), the graph will be empty (NAN) for a long time. People may not
8500 Calculates a hash value of the host name and matches values according to that
8501 hash value. This makes it possible to divide all hosts into groups and match
8502 only values that are in a specific group. The intended use is in load
8503 balancing, where you want to handle only part of all data and leave the rest
8506 The hashing function used tries to distribute the hosts evenly. First, it
8507 calculates a 32E<nbsp>bit hash value using the characters of the hostname:
8510 for (i = 0; host[i] != 0; i++)
8511 hash_value = (hash_value * 251) + host[i];
8513 The constant 251 is a prime number which is supposed to make this hash value
8514 more random. The code then checks the group for this host according to the
8515 I<Total> and I<Match> arguments:
8517 if ((hash_value % Total) == Match)
8522 Please note that when you set I<Total> to two (i.E<nbsp>e. you have only two
8523 groups), then the least significant bit of the hash value will be the XOR of
8524 all least significant bits in the host name. One consequence is that when you
8525 have two hosts, "server0.example.com" and "server1.example.com", where the host
8526 name differs in one digit only and the digits differ by one, those hosts will
8527 never end up in the same group.
8533 =item B<Match> I<Match> I<Total>
8535 Divide the data into I<Total> groups and match all hosts in group I<Match> as
8536 described above. The groups are numbered from zero, i.E<nbsp>e. I<Match> must
8537 be smaller than I<Total>. I<Total> must be at least one, although only values
8538 greater than one really do make any sense.
8540 You can repeat this option to match multiple groups, for example:
8545 The above config will divide the data into seven groups and match groups three
8546 and five. One use would be to keep every value on two hosts so that if one
8547 fails the missing data can later be reconstructed from the second host.
8553 # Operate on the pre-cache chain, so that ignored values are not even in the
8558 # Divide all received hosts in seven groups and accept all hosts in
8562 # If matched: Return and continue.
8565 # If not matched: Return and stop.
8571 =head2 Available targets
8575 =item B<notification>
8577 Creates and dispatches a notification.
8583 =item B<Message> I<String>
8585 This required option sets the message of the notification. The following
8586 placeholders will be replaced by an appropriate value:
8594 =item B<%{plugin_instance}>
8598 =item B<%{type_instance}>
8600 These placeholders are replaced by the identifier field of the same name.
8602 =item B<%{ds:>I<name>B<}>
8604 These placeholders are replaced by a (hopefully) human readable representation
8605 of the current rate of this data source. If you changed the instance name
8606 (using the B<set> or B<replace> targets, see below), it may not be possible to
8607 convert counter values to rates.
8611 Please note that these placeholders are B<case sensitive>!
8613 =item B<Severity> B<"FAILURE">|B<"WARNING">|B<"OKAY">
8615 Sets the severity of the message. If omitted, the severity B<"WARNING"> is
8622 <Target "notification">
8623 Message "Oops, the %{type_instance} temperature is currently %{ds:value}!"
8629 Replaces parts of the identifier using regular expressions.
8635 =item B<Host> I<Regex> I<Replacement>
8637 =item B<Plugin> I<Regex> I<Replacement>
8639 =item B<PluginInstance> I<Regex> I<Replacement>
8641 =item B<TypeInstance> I<Regex> I<Replacement>
8643 Match the appropriate field with the given regular expression I<Regex>. If the
8644 regular expression matches, that part that matches is replaced with
8645 I<Replacement>. If multiple places of the input buffer match a given regular
8646 expression, only the first occurrence will be replaced.
8648 You can specify each option multiple times to use multiple regular expressions
8656 # Replace "example.net" with "example.com"
8657 Host "\\<example.net\\>" "example.com"
8659 # Strip "www." from hostnames
8665 Sets part of the identifier of a value to a given string.
8671 =item B<Host> I<String>
8673 =item B<Plugin> I<String>
8675 =item B<PluginInstance> I<String>
8677 =item B<TypeInstance> I<String>
8679 =item B<MetaDataSet> I<String> I<String>
8681 Set the appropriate field to the given string. The strings for plugin instance
8682 and type instance may be empty, the strings for host and plugin may not be
8683 empty. It's currently not possible to set the type of a value this way.
8690 PluginInstance "coretemp"
8691 TypeInstance "core3"
8696 =head2 Backwards compatibility
8698 If you use collectd with an old configuration, i.E<nbsp>e. one without a
8699 B<Chain> block, it will behave as it used to. This is equivalent to the
8700 following configuration:
8706 If you specify a B<PostCacheChain>, the B<write> target will not be added
8707 anywhere and you will have to make sure that it is called where appropriate. We
8708 suggest to add the above snippet as default target to your "PostCache" chain.
8712 Ignore all values, where the hostname does not contain a dot, i.E<nbsp>e. can't
8728 L<collectd-exec(5)>,
8729 L<collectd-perl(5)>,
8730 L<collectd-unixsock(5)>,
8743 Florian Forster E<lt>octo@collectd.orgE<gt>