1 collectd - System information collection daemon
2 =================================================
8 collectd is a small daemon which collects system information periodically
9 and provides mechanisms to store and monitor the values in a variety of
16 * collectd is able to collect the following data:
19 Apache server utilization: Number of bytes transfered, number of
20 requests handled and detailed scoreboard statistics
23 APC UPS Daemon: UPS charge, load, input/output/battery voltage, etc.
26 Sensors in Macs running Mac OS X / Darwin: Temperature, fanspeed and
30 Statistics about Ascent, a free server for the game `World of Warcraft'.
33 Batterycharge, -current and voltage of ACPI and PMU based laptop
37 Name server and resolver statistics from the `statistics-channel'
38 interface of BIND 9.5, 9,6 and later.
41 Number of nf_conntrack entries.
44 Number of context switches done by the operating system.
47 CPU utilization: Time spent in the system, user, nice, idle, and related
51 CPU frequency (For laptops with speed step or a similar technology)
54 Parse statistics from websites using regular expressions.
57 Retrieves JSON data via cURL and parses it according to user
61 Retrieves XML data via cURL and parses it according to user
65 Executes SQL statements on various databases and interprets the returned
69 Mountpoint usage (Basically the values `df(1)' delivers)
72 Disk utilization: Sectors read/written, number of read/write actions,
73 average time an IO-operation took to complete.
76 DNS traffic: Query types, response codes, opcodes and traffic/octets
80 Email statistics: Count, traffic, spam scores and checks.
81 See collectd-email(5).
84 Amount of entropy available to the system.
87 Values gathered by a custom program or script.
91 Count the number of files in directories.
94 Linux file-system based caching framework statistics.
97 Receive multicast traffic from Ganglia instances.
100 Harddisk temperatures using hddtempd.
103 Interface traffic: Number of octets, packets and errors for each
107 Iptables' counters: Number of bytes that were matched by a certain
111 IPMI (Intelligent Platform Management Interface) sensors information.
114 IPVS connection statistics (number of connections, octets and packets
115 for each service and destination).
116 See http://www.linuxvirtualserver.org/software/index.html.
119 IRQ counters: Frequency in which certain interrupts occur.
122 Integrates a `Java Virtual Machine' (JVM) to execute plugins in Java
123 bytecode. See “Configuring with libjvm” below.
126 System load average over the last 1, 5 and 15 minutes.
129 CPU, disk and network I/O statistics from virtual machines.
132 Queries very detailed usage statistics from wireless LAN adapters and
133 interfaces that use the Atheros chipset and the MadWifi driver.
136 Motherboard sensors: temperature, fanspeed and voltage information,
140 Query and parse data from a memcache daemon (memcached).
143 Statistics of the memcached distributed caching system.
144 <http://www.danga.com/memcached/>
147 Memory utilization: Memory occupied by running processes, page cache,
148 buffer cache and free.
151 Reads values from Modbus/TCP enabled devices. Supports reading values
152 from multiple "slaves" so gateway devices can be used.
155 Information provided by serial multimeters, such as the `Metex
159 MySQL server statistics: Commands issued, handlers triggered, thread
160 usage, query cache utilization and traffic/octets sent and received.
163 Plugin to query performance values from a NetApp storage system using the
164 “Manage ONTAP” SDK provided by NetApp.
167 Very detailed Linux network interface and routing statistics. You can get
168 (detailed) information on interfaces, qdiscs, classes, and, if you can
169 make use of it, filters.
172 Receive values that were collected by other hosts. Large setups will
173 want to collect the data on one dedicated machine, and this is the
174 plugin of choice for that.
177 NFS Procedures: Which NFS command were called how often. Only NFSv2 and
181 Collects statistics from `nginx' (speak: engine X), a HTTP and mail
185 NTP daemon statistics: Local clock drift, offset to peers, etc.
188 Network UPS tools: UPS current, voltage, power, charge, utilisation,
189 temperature, etc. See upsd(8).
192 Queries routing information from the “Optimized Link State Routing”
195 - onewire (EXPERIMENTAL!)
196 Read onewire sensors using the owcapu library of the owfs project.
197 Please read in collectd.conf(5) why this plugin is experimental.
200 RX and TX of each client in openvpn-status.log (status-version 2).
201 <http://openvpn.net/index.php/documentation/howto.html>
204 Query data from an Oracle database.
207 The perl plugin implements a Perl-interpreter into collectd. You can
208 write your own plugins in Perl and return arbitrary values using this
209 API. See collectd-perl(5).
212 Receive and dispatch timing values from Pinba, a profiling extension for
216 Network latency: Time to reach the default gateway or another given
220 PostgreSQL database statistics: active server connections, transaction
221 numbers, block IO, table row manipulations.
224 PowerDNS name server statistics.
227 Process counts: Number of running, sleeping, zombie, ... processes.
230 Counts various aspects of network protocols such as IP, TCP, UDP, etc.
233 The python plugin implements a Python interpreter into collectd. This
234 makes it possible to write plugins in Python which are executed by
235 collectd without the need to start a heavy interpreter every interval.
236 See collectd-python(5) for details.
239 The redis plugin gathers information from a redis server, including:
240 uptime, used memory, total connections etc.
243 Query interface and wireless registration statistics from RouterOS.
246 RRDtool caching daemon (RRDcacheD) statistics.
249 System sensors, accessed using lm_sensors: Voltages, temperatures and
253 RX and TX of serial interfaces. Linux only; needs root privileges.
256 Read values from SNMP (Simple Network Management Protocol) enabled
257 network devices such as switches, routers, thermometers, rack monitoring
258 servers, etc. See collectd-snmp(5).
261 Pages swapped out onto harddisk or whatever is called `swap' by the OS..
264 Parse table-like structured files.
267 Follows (tails) logfiles, parses them by lines and submits matched
271 Bytes and operations read and written on tape devices. Solaris only.
274 Number of TCP connections to specific local and remote ports.
277 TeamSpeak2 server statistics.
280 Plugin to read values from `The Energy Detective' (TED).
283 Linux ACPI thermal zone information.
286 Reads the number of records and file size from a running Tokyo Tyrant
290 System uptime statistics.
293 Users currently logged in.
296 Various statistics from Varnish, an HTTP accelerator.
299 Virtual memory statistics, e. g. the number of page-ins/-outs or the
300 number of pagefaults.
303 System resources used by Linux VServers.
304 See <http://linux-vserver.org/>.
307 Link quality of wireless cards. Linux only.
310 Bitrate and frequency of music played with XMMS.
313 Statistics for ZFS' “Adaptive Replacement Cache” (ARC).
315 * Output can be written or sent to various destinations by the following
319 Sends JSON-encoded data to an Advanced Message Queuing Protocol (AMQP)
320 server, such as RabbitMQ.
323 Write to comma separated values (CSV) files. This needs lots of
324 diskspace but is extremely portable and can be analysed with almost
325 every program that can analyse anything. Even Microsoft's Excel..
328 Send the data to a remote host to save the data somehow. This is useful
329 for large setups where the data should be saved by a dedicated machine.
332 Of course the values are propagated to plugins written in Perl, too, so
333 you can easily do weird stuff with the plugins we didn't dare think of
334 ;) See collectd-perl(5).
337 It's possible to implement write plugins in Python using the python
338 plugin. See collectd-python(5) for details.
341 Output to round-robin-database (RRD) files using the RRDtool caching
342 daemon (RRDcacheD) - see rrdcached(1). That daemon provides a general
343 implementation of the caching done by the `rrdtool' plugin.
346 Output to round-robin-database (RRD) files using librrd. See rrdtool(1).
347 This is likely the most popular destination for such values. Since
348 updates to RRD-files are somewhat expensive this plugin can cache
349 updates to the files and write a bunch of updates at once, which lessens
353 One can query the values from the unixsock plugin whenever they're
354 needed. Please read collectd-unixsock(5) for a description on how that's
358 Sends the values collected by collectd to a web-server using HTTP POST
359 requests. The transmitted data is either in a form understood by the
360 Exec plugin or formatted in JSON.
362 * Logging is, as everything in collectd, provided by plugins. The following
363 plugins keep up informed about what's going on:
366 Writes logmessages to a file or STDOUT/STDERR.
369 Log messages are propagated to plugins written in Perl as well.
370 See collectd-perl(5).
373 It's possible to implement log plugins in Python using the python plugin.
374 See collectd-python(5) for details.
377 Logs to the standard UNIX logging mechanism, syslog.
379 * Notifications can be handled by the following plugins:
382 Send a desktop notification to a notification daemon, as defined in
383 the Desktop Notification Specification. To actually display the
384 notifications, notification-daemon is required.
385 See http://www.galago-project.org/specs/notification/.
388 Send an E-mail with the notification message to the configured
392 Execute a program or script to handle the notification.
393 See collectd-exec(5).
396 Writes the notification message to a file or STDOUT/STDERR.
399 Send the notification to a remote host to handle it somehow.
402 Notifications are propagated to plugins written in Perl as well.
403 See collectd-perl(5).
406 It's possible to implement notification plugins in Python using the
407 python plugin. See collectd-python(5) for details.
409 * Value processing can be controlled using the "filter chain" infrastructure
410 and "matches" and "targets". The following plugins are available:
412 - match_empty_counter
413 Match counter values which are currently zero.
416 Match values using a hash function of the hostname.
419 Match values by their identifier based on regular expressions.
422 Match values with an invalid timestamp.
425 Select values by their data sources' values.
427 - target_notification
428 Create and dispatch a notification.
431 Replace parts of an identifier using regular expressions.
434 Scale (multiply) values by an arbitrary value.
437 Set (overwrite) entire parts of an identifier.
439 * Miscellaneous plugins:
442 Sets the hostname to an unique identifier. This is meant for setups
443 where each client may migrate to another physical host, possibly going
444 through one or more name changes in the process.
446 * Performance: Since collectd is running as a daemon it doesn't spend much
447 time starting up again and again. With the exception of the exec plugin no
448 processes are forked. Caching in output plugins, such as the rrdtool and
449 network plugins, makes sure your resources are used efficiently. Also,
450 since collectd is programmed multithreaded it benefits from hyperthreading
451 and multicore processors and makes sure that the daemon isn't idle if only
452 one plugin waits for an IO-operation to complete.
454 * Once set up, hardly any maintenance is necessary. Setup is kept as easy
455 as possible and the default values should be okay for most users.
461 * collectd's configuration file can be found at `sysconfdir'/collectd.conf.
462 Run `collectd -h' for a list of builtin defaults. See `collectd.conf(5)'
463 for a list of options and a syntax description.
465 * When the `csv' or `rrdtool' plugins are loaded they'll write the values to
466 files. The usual place for these files is beneath `/var/lib/collectd'.
468 * When using some of the plugins, collectd needs to run as user root, since
469 only root can do certain things, such as craft ICMP packages needed to ping
470 other hosts. collectd should NOT be installed setuid root since it can be
471 used to overwrite valuable files!
473 * Sample scripts to generate graphs reside in `contrib/' in the source
474 package or somewhere near `/usr/share/doc/collectd' in most distributions.
475 Please be aware that those script are meant as a starting point for your
476 own experiments.. Some of them require the `RRDs' Perl module.
477 (`librrds-perl' on Debian) If you have written a more sophisticated
478 solution please share it with us.
480 * The RRAs of the automatically created RRD files depend on the `step'
481 and `heartbeat' settings given. If change these settings you may need to
482 re-create the files, losing all data. Please be aware of that when changing
483 the values and read the rrdtool(1) manpage thoroughly.
486 collectd and chkrootkit
487 -----------------------
489 If you are using the `dns' plugin chkrootkit(1) will report collectd as a
490 packet sniffer ("<iface>: PACKET SNIFFER(/usr/sbin/collectd[<pid>])"). The
491 plugin captures all UDP packets on port 53 to analyze the DNS traffic. In
492 this case, collectd is a legitimate sniffer and the report should be
493 considered to be a false positive. However, you might want to check that
494 this really is collectd and not some other, illegitimate sniffer.
500 To compile collectd from source you will need:
502 * Usual suspects: C compiler, linker, preprocessor, make, ...
504 * A POSIX-threads (pthread) implementation.
505 Since gathering some statistics is slow (network connections, slow devices,
506 etc) the collectd is parallelized. The POSIX threads interface is being
507 used and should be found in various implementations for hopefully all
510 * CoreFoundation.framework and IOKit.framework (optional)
511 For compiling on Darwin in general and the `apple_sensors' plugin in
513 <http://developer.apple.com/corefoundation/>
515 * libclntsh (optional)
516 Used by the `oracle' plugin.
518 * libcredis (optional)
519 Used by the redis plugin. Please note that you require a 0.2.2 version
520 or higher. <http://code.google.com/p/credis/>
523 If you want to use the `apache', `ascent', `curl', `nginx', or `write_http'
525 <http://curl.haxx.se/>
528 Used by the `dbi' plugin to connect to various databases.
529 <http://libdbi.sourceforge.net/>
531 * libesmtp (optional)
532 For the `notify_email' plugin.
533 <http://www.stafford.uklinux.net/libesmtp/>
535 * libganglia (optional)
536 Used by the `gmond' plugin to process data received from Ganglia.
537 <http://ganglia.info/>
539 * libgcrypt (optional)
540 Used by the `network' plugin for encryption and authentication.
541 <http://www.gnupg.org/>
544 If present, the uuid plugin will check for UUID from HAL.
545 <http://hal.freedesktop.org/>
548 For querying iptables counters.
549 <http://netfilter.org/>
551 If not found on the system, a version shipped with this distribution can
552 be used. It requires some Linux headers in /usr/include/linux. You can
553 force the build system to use the shipped version by specifying
554 --with-libiptc=shipped
555 when running the configure script.
558 Library that encapsulates the `Java Virtual Machine' (JVM). This library is
559 used by the Java plugin to execute Java bytecode. See “Configuring with
561 <http://openjdk.java.net/> (and others)
563 * libmemcached (optional)
564 Used by the `memcachec' plugin to connect to a memcache daemon.
565 <http://tangent.org/552/libmemcached.html>
567 * libmodbus (optional)
568 Used by the `modbus' plugin to communicate with Modbus/TCP devices.
569 <https://launchpad.net/libmodbus>
571 * libmysqlclient (optional)
572 Unsurprisingly used by the `mysql' plugin.
573 <http://dev.mysql.com/>
575 * libnetapp (optional)
576 Required for the “netapp” plugin.
577 This library is part of the “Manage ONTAP SDK” published by NetApp.
579 * libnetlink (optional)
580 Used, obviously, for the `netlink' plugin.
581 <http://www.linuxfoundation.org/en/Net:Iproute2>
583 * libnetsnmp (optional)
584 For the `snmp' plugin.
585 <http://www.net-snmp.org/>
587 * libnotify (optional)
588 For the `notify_desktop' plugin.
589 <http://www.galago-project.org/>
591 * liboping (optional)
592 Used by the `ping' plugin to send and receive ICMP packets.
593 <http://verplant.org/liboping/>
595 * libowcapi (optional)
596 Used by the `onewire' plugin to read values from onewire sensors (or the
598 <http://www.owfs.org/>
601 Used to capture packets by the `dns' plugin.
602 <http://www.tcpdump.org/>
605 Obviously used by the `perl' plugin. The library has to be compiled with
606 ithread support (introduced in Perl 5.6.0).
607 <http://www.perl.org/>
610 The PostgreSQL C client library used by the `postgresql' plugin.
611 <http://www.postgresql.org/>
613 * libprotobuf-c, protoc-c (optional)
614 Used by the `pinba' plugin to generate a parser for the network packets
615 sent by the Pinba PHP extension.
616 <http://code.google.com/p/protobuf-c/>
618 * libpython (optional)
619 Used by the `python' plugin. Currently, only 2.3 ≦ Python < 3 is supported.
620 <http://www.python.org/>
622 * librabbitmq (optional; also called “rabbitmq-c”)
623 Used by the AMQP plugin for AMQP connections, for example to RabbitMQ.
624 <http://hg.rabbitmq.com/rabbitmq-c/>
626 * librouteros (optional)
627 Used by the `routeros' plugin to connect to a device running `RouterOS'.
628 <http://verplant.org/librouteros/>
631 Used by the `rrdtool' and `rrdcached' plugins. The latter requires RRDtool
632 client support which was added after version 1.3 of RRDtool. Versions 1.0,
633 1.2 and 1.3 are known to work with the `rrdtool' plugin.
634 <http://oss.oetiker.ch/rrdtool/>
636 * librt, libsocket, libkstat, libdevinfo (optional)
637 Various standard Solaris libraries which provide system functions.
638 <http://developers.sun.com/solaris/>
640 * libsensors (optional)
641 To read from `lm_sensors', see the `sensors' plugin.
642 <http://www.lm-sensors.org/>
644 * libstatgrab (optional)
645 Used by various plugins to collect statistics on systems other than Linux
647 <http://www.i-scream.org/libstatgrab/>
649 * libtokyotyrant (optional)
650 Used by the tokyotyrant plugin.
651 <http://1978th.net/tokyotyrant/>
653 * libupsclient/nut (optional)
654 For the `nut' plugin which queries nut's `upsd'.
655 <http://networkupstools.org/>
658 Collect statistics from virtual machines.
659 <http://libvirt.org/>
662 Parse XML data. This is needed for the `ascent' and `libvirt' plugins.
663 <http://xmlsoft.org/>
666 <http://www.xmms.org/>
669 Parse JSON data. This is needed for the `curl_json' plugin.
670 <http://github.com/lloyd/yajl>
672 * libvarnish (optional)
673 Fetches statistics from a Varnish instance. This is needed for the Varnish plugin
674 <http://varnish-cache.org>
676 Configuring / Compiling / Installing
677 ------------------------------------
679 To configure, build and install collectd with the default settings, run
680 `./configure && make && make install'. For detailed, generic instructions
681 see INSTALL. For a complete list of configure options and their description,
682 run `./configure --help'.
684 By default, the configure script will check for all build dependencies and
685 disable all plugins whose requirements cannot be fulfilled (any other plugin
686 will be enabled). To enable a plugin, install missing dependencies (see
687 section `Prerequisites' above) and rerun `configure'. If you specify the
688 `--enable-<plugin>' configure option, the script will fail if the depen-
689 dencies for the specified plugin are not met. In that case you can force the
690 plugin to be built using the `--enable-<plugin>=force' configure option.
691 This will most likely fail though unless you're working in a very unusual
692 setup and you really know what you're doing. If you specify the
693 `--disable-<plugin>' configure option, the plugin will not be built. If you
694 specify the `--enable-all-plugins' or `--disable-all-plugins' configure
695 options, all plugins will be enabled or disabled respectively by default.
696 Explicitly enabling or disabling a plugin overwrites the default for the
697 specified plugin. These options are meant for package maintainers and should
698 not be used in everyday situations.
700 By default, collectd will be installed into `/opt/collectd'. You can adjust
701 this setting by specifying the `--prefix' configure option - see INSTALL for
702 details. If you pass DESTDIR=<path> to `make install', <path> will be
703 prefixed to all installation directories. This might be useful when creating
704 packages for collectd.
706 Configuring with libjvm
707 -----------------------
709 To determine the location of the required files of a Java installation is not
710 an easy task, because the locations vary with your kernel (Linux, SunOS, …)
711 and with your architecture (x86, SPARC, …) and there is no ‘java-config’
712 script we could use. Configuration of the JVM library is therefore a bit
715 The easiest way to use the `--with-java=$JAVA_HOME' option, where
716 `$JAVA_HOME' is usually something like:
717 /usr/lib/jvm/java-1.5.0-sun-1.5.0.14
719 The configure script will then use find(1) to look for the following files:
725 If found, appropriate CPP-flags and LD-flags are set and the following
726 library checks succeed.
728 If this doesn't work for you, you have the possibility to specify CPP-flags,
729 C-flags and LD-flags for the ‘Java’ plugin by hand, using the following three
730 (environment) variables:
736 For example (shortened for demonstration purposes):
738 ./configure JAVA_CPPFLAGS="-I$JAVA_HOME/include -I$JAVA_HOME/include/linux"
740 Adding "-ljvm" to the JAVA_LDFLAGS is done automatically, you don't have to
746 To compile correctly collectd needs to be able to initialize static
747 variables to NAN (Not A Number). Some C libraries, especially the GNU
748 libc, have a problem with that.
750 Luckily, with GCC it's possible to work around that problem: One can define
751 NAN as being (0.0 / 0.0) and `isnan' as `f != f'. However, to test this
752 ``implementation'' the configure script needs to compile and run a short
753 test program. Obviously running a test program when doing a cross-
754 compilation is, well, challenging.
756 If you run into this problem, you can use the `--with-nan-emulation'
757 configure option to force the use of this implementation. We can't promise
758 that the compiled binary actually behaves as it should, but since NANs
759 are likely never passed to the libm you have a good chance to be lucky.
761 Likewise, collectd needs to know the layout of doubles in memory, in order
762 to craft uniform network packets over different architectures. For this, it
763 needs to know how to convert doubles into the memory layout used by x86. The
764 configure script tries to figure this out by compiling and running a few
765 small test programs. This is of course not possible when cross-compiling.
766 You can use the `--with-fp-layout' option to tell the configure script which
767 conversion method to assume. Valid arguments are:
769 * `nothing' (12345678 -> 12345678)
770 * `endianflip' (12345678 -> 87654321)
771 * `intswap' (12345678 -> 56781234)
777 For questions, bug reports, development information and basically all other
778 concerns please send an email to collectd's mailing list at
779 <collectd at verplant.org>.
781 For live discussion and more personal contact visit us in IRC, we're in
782 channel #collectd on freenode.
788 Florian octo Forster <octo at verplant.org>,
789 Sebastian tokkee Harl <sh at tokkee.org>,
790 and many contributors (see `AUTHORS').
792 Please send bug reports and patches to the mailing list, see `Contact'