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 Executes SQL statements on various databases and interprets the returned
65 Mountpoint usage (Basically the values `df(1)' delivers)
68 Disk utilization: Sectors read/written, number of read/write actions,
69 average time an IO-operation took to complete.
72 DNS traffic: Query types, response codes, opcodes and traffic/octets
76 Email statistics: Count, traffic, spam scores and checks.
77 See collectd-email(5).
80 Amount of entropy available to the system.
83 Values gathered by a custom program or script.
87 Count the number of files in directories.
90 Linux file-system based caching framework statistics.
93 Receive multicast traffic from Ganglia instances.
96 Harddisk temperatures using hddtempd.
99 Interface traffic: Number of octets, packets and errors for each
103 Iptables' counters: Number of bytes that were matched by a certain
107 IPMI (Intelligent Platform Management Interface) sensors information.
110 IPVS connection statistics (number of connections, octets and packets
111 for each service and destination).
112 See http://www.linuxvirtualserver.org/software/index.html.
115 IRQ counters: Frequency in which certain interrupts occur.
118 Integrates a `Java Virtual Machine' (JVM) to execute plugins in Java
119 bytecode. See “Configuring with libjvm” below.
122 System load average over the last 1, 5 and 15 minutes.
125 CPU, disk and network I/O statistics from virtual machines.
128 Queries very detailed usage statistics from wireless LAN adapters and
129 interfaces that use the Atheros chipset and the MadWifi driver.
132 Motherboard sensors: temperature, fanspeed and voltage information,
136 Query and parse data from a memcache daemon (memcached).
139 Statistics of the memcached distributed caching system.
140 <http://www.danga.com/memcached/>
143 Memory utilization: Memory occupied by running processes, page cache,
144 buffer cache and free.
147 Information provided by serial multimeters, such as the `Metex
151 MySQL server statistics: Commands issued, handlers triggered, thread
152 usage, query cache utilization and traffic/octets sent and received.
155 Plugin to query performance values from a NetApp storage system using the
156 “Manage ONTAP” SDK provided by NetApp.
159 Very detailed Linux network interface and routing statistics. You can get
160 (detailed) information on interfaces, qdiscs, classes, and, if you can
161 make use of it, filters.
164 Receive values that were collected by other hosts. Large setups will
165 want to collect the data on one dedicated machine, and this is the
166 plugin of choice for that.
169 NFS Procedures: Which NFS command were called how often. Only NFSv2 and
173 Collects statistics from `nginx' (speak: engine X), a HTTP and mail
177 NTP daemon statistics: Local clock drift, offset to peers, etc.
180 Network UPS tools: UPS current, voltage, power, charge, utilisation,
181 temperature, etc. See upsd(8).
184 Queries routing information from the “Optimized Link State Routing”
187 - onewire (EXPERIMENTAL!)
188 Read onewire sensors using the owcapu library of the owfs project.
189 Please read in collectd.conf(5) why this plugin is experimental.
192 RX and TX of each client in openvpn-status.log (status-version 2).
193 <http://openvpn.net/index.php/documentation/howto.html>
196 Query data from an Oracle database.
199 The perl plugin implements a Perl-interpreter into collectd. You can
200 write your own plugins in Perl and return arbitrary values using this
201 API. See collectd-perl(5).
204 Network latency: Time to reach the default gateway or another given
208 PostgreSQL database statistics: active server connections, transaction
209 numbers, block IO, table row manipulations.
212 PowerDNS name server statistics.
215 Process counts: Number of running, sleeping, zombie, ... processes.
218 Counts various aspects of network protocols such as IP, TCP, UDP, etc.
221 The python plugin implements a Python interpreter into collectd. This
222 makes it possible to write plugins in Python which are executed by
223 collectd without the need to start a heavy interpreter every interval.
224 See collectd-python(5) for details.
227 RRDtool caching daemon (RRDcacheD) statistics.
230 System sensors, accessed using lm_sensors: Voltages, temperatures and
234 RX and TX of serial interfaces. Linux only; needs root privileges.
237 Read values from SNMP (Simple Network Management Protocol) enabled
238 network devices such as switches, routers, thermometers, rack monitoring
239 servers, etc. See collectd-snmp(5).
242 Pages swapped out onto harddisk or whatever is called `swap' by the OS..
245 Parse table-like structured files.
248 Follows (tails) logfiles, parses them by lines and submits matched
252 Bytes and operations read and written on tape devices. Solaris only.
255 Number of TCP connections to specific local and remote ports.
258 TeamSpeak2 server statistics.
261 Plugin to read values from `The Energy Detective' (TED).
264 Linux ACPI thermal zone information.
267 Reads the number of records and file size from a running Tokyo Tyrant
271 System uptime statistics.
274 Users currently logged in.
277 Virtual memory statistics, e. g. the number of page-ins/-outs or the
278 number of pagefaults.
281 System resources used by Linux VServers.
282 See <http://linux-vserver.org/>.
285 Link quality of wireless cards. Linux only.
288 Bitrate and frequency of music played with XMMS.
291 Statistics for ZFS' “Adaptive Replacement Cache” (ARC).
293 * Output can be written or send to various destinations by the following
297 Write to comma separated values (CSV) files. This needs lots of
298 diskspace but is extremely portable and can be analysed with almost
299 every program that can analyse anything. Even Microsoft's Excel..
302 Send the data to a remote host to save the data somehow. This is useful
303 for large setups where the data should be saved by a dedicated machine.
306 Of course the values are propagated to plugins written in Perl, too, so
307 you can easily do weird stuff with the plugins we didn't dare think of
308 ;) See collectd-perl(5).
311 It's possible to implement write plugins in Python using the python
312 plugin. See collectd-python(5) for details.
315 Output to round-robin-database (RRD) files using the RRDtool caching
316 daemon (RRDcacheD) - see rrdcached(1). That daemon provides a general
317 implementation of the caching done by the `rrdtool' plugin.
320 Output to round-robin-database (RRD) files using librrd. See rrdtool(1).
321 This is likely the most popular destination for such values. Since
322 updates to RRD-files are somewhat expensive this plugin can cache
323 updates to the files and write a bunch of updates at once, which lessens
327 One can query the values from the unixsock plugin whenever they're
328 needed. Please read collectd-unixsock(5) for a description on how that's
332 Sends the values collected by collectd to a web-server using HTTP POST
333 requests. The transmitted data is either in a form understood by the
334 Exec plugin or formatted in JSON.
336 * Logging is, as everything in collectd, provided by plugins. The following
337 plugins keep up informed about what's going on:
340 Writes logmessages to a file or STDOUT/STDERR.
343 Log messages are propagated to plugins written in Perl as well.
344 See collectd-perl(5).
347 It's possible to implement log plugins in Python using the python plugin.
348 See collectd-python(5) for details.
351 Logs to the standard UNIX logging mechanism, syslog.
353 * Notifications can be handled by the following plugins:
356 Send a desktop notification to a notification daemon, as defined in
357 the Desktop Notification Specification. To actually display the
358 notifications, notification-daemon is required.
359 See http://www.galago-project.org/specs/notification/.
362 Send an E-mail with the notification message to the configured
366 Execute a program or script to handle the notification.
367 See collectd-exec(5).
370 Writes the notification message to a file or STDOUT/STDERR.
373 Send the notification to a remote host to handle it somehow.
376 Notifications are propagated to plugins written in Perl as well.
377 See collectd-perl(5).
380 It's possible to implement notification plugins in Python using the
381 python plugin. See collectd-python(5) for details.
383 * Value processing can be controlled using the "filter chain" infrastructure
384 and "matches" and "targets". The following plugins are available:
386 - match_empty_counter
387 Match counter values which are currently zero.
390 Match values by their identifier based on regular expressions.
393 Match values with an invalid timestamp.
396 Select values by their data sources' values.
398 - target_notification
399 Create and dispatch a notification.
402 Replace parts of an identifier using regular expressions.
405 Set (overwrite) entire parts of an identifier.
407 * Miscellaneous plugins:
410 Sets the hostname to an unique identifier. This is meant for setups
411 where each client may migrate to another physical host, possibly going
412 through one or more name changes in the process.
414 * Performance: Since collectd is running as a daemon it doesn't spend much
415 time starting up again and again. With the exception of the exec plugin no
416 processes are forked. Caching in output plugins, such as the rrdtool and
417 network plugins, makes sure your resources are used efficiently. Also,
418 since collectd is programmed multithreaded it benefits from hyperthreading
419 and multicore processors and makes sure that the daemon isn't idle if only
420 one plugins waits for an IO-operation to complete.
422 * Once set up, hardly any maintenance is necessary. Setup is kept as easy
423 as possible and the default values should be okay for most users.
429 * collectd's configuration file can be found at `sysconfdir'/collectd.conf.
430 Run `collectd -h' for a list of builtin defaults. See `collectd.conf(5)'
431 for a list of options and a syntax description.
433 * When the `csv' or `rrdtool' plugins are loaded they'll write the values to
434 files. The usual place for these files is beneath `/var/lib/collectd'.
436 * When using some of the plugins, collectd needs to run as user root, since
437 only root can do certain things, such as craft ICMP packages needed to ping
438 other hosts. collectd should NOT be installed setuid root since it can be
439 used to overwrite valuable files!
441 * Sample scripts to generate graphs reside in `contrib/' in the source
442 package or somewhere near `/usr/share/doc/collectd' in most distributions.
443 Please be aware that those script are meant as a starting point for your
444 own experiments.. Some of them require the `RRDs' Perl module.
445 (`librrds-perl' on Debian) If you have written a more sophisticated
446 solution please share it with us.
448 * The RRAs of the automatically created RRD files depend on the `step'
449 and `heartbeat' settings given. If change these settings you may need to
450 re-create the files, losing all data. Please be aware of that when changing
451 the values and read the rrdtool(1) manpage thoroughly.
454 collectd and chkrootkit
455 -----------------------
457 If you are using the `dns' plugin chkrootkit(1) will report collectd as a
458 packet sniffer ("<iface>: PACKET SNIFFER(/usr/sbin/collectd[<pid>])"). The
459 plugin captures all UDP packets on port 53 to analyze the DNS traffic. In
460 this case, collectd is a legitimate sniffer and the report should be
461 considered to be a false positive. However, you might want to check that
462 this really is collectd and not some other, illegitimate sniffer.
468 To compile collectd from source you will need:
470 * Usual suspects: C compiler, linker, preprocessor, make, ...
472 * A POSIX-threads (pthread) implementation.
473 Since gathering some statistics is slow (network connections, slow devices,
474 etc) the collectd is parallelized. The POSIX threads interface is being
475 used and should be found in various implementations for hopefully all
478 * CoreFoundation.framework and IOKit.framework (optional)
479 For compiling on Darwin in general and the `apple_sensors' plugin in
481 <http://developer.apple.com/corefoundation/>
483 * libclntsh (optional)
484 Used by the `oracle' plugin.
487 If you want to use the `apache', `ascent', `curl', `nginx', or `write_http'
489 <http://curl.haxx.se/>
492 Used by the `dbi' plugin to connect to various databases.
493 <http://libdbi.sourceforge.net/>
495 * libesmtp (optional)
496 For the `notify_email' plugin.
497 <http://www.stafford.uklinux.net/libesmtp/>
499 * libganglia (optional)
500 Used by the `gmond' plugin to process data received from Ganglia.
501 <http://ganglia.info/>
503 * libgcrypt (optional)
504 Used by the `network' plugin for encryption and authentication.
505 <http://www.gnupg.org/>
508 If present, the uuid plugin will check for UUID from HAL.
509 <http://hal.freedesktop.org/>
512 For querying iptables counters.
513 <http://netfilter.org/>
515 If not found on the system, a version shipped with this distribution can
516 be used. It requires some Linux headers in /usr/include/linux. You can
517 force the build system to use the shipped version by specifying
518 --with-libiptc=shipped
519 when running the configure script.
522 Library that encapsulates the `Java Virtual Machine' (JVM). This library is
523 used by the Java plugin to execute Java bytecode. See “Configuring with
525 <http://openjdk.java.net/> (and others)
527 * libmemcached (optional)
528 Used by the `memcachec' plugin to connect to a memcache daemon.
529 <http://tangent.org/552/libmemcached.html>
531 * libmysqlclient (optional)
532 Unsurprisingly used by the `mysql' plugin.
533 <http://dev.mysql.com/>
535 * libnatapp (optional)
536 Required for the “netapp” plugin.
537 This library is part of the “Manage ONTAP SDK” published by NetApp.
539 * libnetlink (optional)
540 Used, obviously, for the `netlink' plugin.
541 <http://www.linuxfoundation.org/en/Net:Iproute2>
543 * libnetsnmp (optional)
544 For the `snmp' plugin.
545 <http://www.net-snmp.org/>
547 * libnotify (optional)
548 For the `notify_desktop' plugin.
549 <http://www.galago-project.org/>
551 * liboping (optional)
552 Used by the `ping' plugin to send and receive ICMP packets.
553 <http://verplant.org/liboping/>
555 * libowcapi (optional)
556 Used by the `onewire' plugin to read values from onewire sensors (or the
558 <http://www.owfs.org/>
561 Used to capture packets by the `dns' plugin.
562 <http://www.tcpdump.org/>
565 Obviously used by the `perl' plugin. The library has to be compiled with
566 ithread support (introduced in Perl 5.6.0).
567 <http://www.perl.org/>
570 The PostgreSQL C client library used by the `postgresql' plugin.
571 <http://www.postgresql.org/>
573 * libpython (optional)
574 Used by the `python' plugin. Currently, only 2.3 ≦ Python < 3 is supported.
575 <http://www.python.org/>
577 * librouteros (optional)
578 Used by the `routeros' plugin to connect to a device running `RouterOS'.
579 <http://verplant.org/librouteros/>
582 Used by the `rrdtool' and `rrdcached' plugins. The latter requires RRDtool
583 client support which was added after version 1.3 of RRDtool. Versions 1.0,
584 1.2 and 1.3 are known to work with the `rrdtool' plugin.
585 <http://oss.oetiker.ch/rrdtool/>
587 * librt, libsocket, libkstat, libdevinfo (optional)
588 Various standard Solaris libraries which provide system functions.
589 <http://developers.sun.com/solaris/>
591 * libsensors (optional)
592 To read from `lm_sensors', see the `sensors' plugin.
593 <http://www.lm-sensors.org/>
595 * libstatgrab (optional)
596 Used by various plugins to collect statistics on systems other than Linux
598 <http://www.i-scream.org/libstatgrab/>
600 * libtokyotyrant (optional)
601 Used by the tokyotyrant plugin.
602 <http://1978th.net/tokyotyrant/>
604 * libupsclient/nut (optional)
605 For the `nut' plugin which queries nut's `upsd'.
606 <http://networkupstools.org/>
609 Collect statistics from virtual machines.
610 <http://libvirt.org/>
613 Parse XML data. This is needed for the `ascent' and `libvirt' plugins.
614 <http://xmlsoft.org/>
617 <http://www.xmms.org/>
620 Parse JSON data. This is needed for the `curl_json' plugin.
621 <http://github.com/lloyd/yajl>
623 Configuring / Compiling / Installing
624 ------------------------------------
626 To configure, build and install collectd with the default settings, run
627 `./configure && make && make install'. For detailed, generic instructions
628 see INSTALL. For a complete list of configure options and their description,
629 run `./configure --help'.
631 By default, the configure script will check for all build dependencies and
632 disable all plugins whose requirements cannot be fulfilled (any other plugin
633 will be enabled). To enable a plugin, install missing dependencies (see
634 section `Prerequisites' above) and rerun `configure'. If you specify the
635 `--enable-<plugin>' configure option, the script will fail if the depen-
636 dencies for the specified plugin are not met. In that case you can force the
637 plugin to be built using the `--enable-<plugin>=force' configure option.
638 This will most likely fail though unless you're working in a very unusual
639 setup and you really know what you're doing. If you specify the
640 `--disable-<plugin>' configure option, the plugin will not be built. If you
641 specify the `--enable-all-plugins' or `--disable-all-plugins' configure
642 options, all plugins will be enabled or disabled respectively by default.
643 Explicitly enabling or disabling a plugin overwrites the default for the
644 specified plugin. These options are meant for package maintainers and should
645 not be used in everyday situations.
647 By default, collectd will be installed into `/opt/collectd'. You can adjust
648 this setting by specifying the `--prefix' configure option - see INSTALL for
649 details. If you pass DESTDIR=<path> to `make install', <path> will be
650 prefixed to all installation directories. This might be useful when creating
651 packages for collectd.
653 Configuring with libjvm
654 -----------------------
656 To determine the location of the required files of a Java installation is not
657 an easy task, because the locations vary with your kernel (Linux, SunOS, …)
658 and with your architecture (x86, SPARC, …) and there is no ‘java-config’
659 script we could use. Configuration of the JVM library is therefore a bit
662 The easiest way to use the `--with-java=$JAVA_HOME' option, where
663 `$JAVA_HOME' is usually something like:
664 /usr/lib/jvm/java-1.5.0-sun-1.5.0.14
666 The configure script will then use find(1) to look for the following files:
672 If found, appropriate CPP-flags and LD-flags are set and the following
673 library checks succeed.
675 If this doesn't work for you, you have the possibility to specify CPP-flags,
676 C-flags and LD-flags for the ‘Java’ plugin by hand, using the following three
677 (environment) variables:
683 For example (shortened for demonstration purposes):
685 ./configure JAVA_CPPFLAGS="-I$JAVA_HOME/include -I$JAVA_HOME/include/linux"
687 Adding "-ljvm" to the JAVA_LDFLAGS is done automatically, you don't have to
693 To compile correctly collectd needs to be able to initialize static
694 variables to NAN (Not A Number). Some C libraries, especially the GNU
695 libc, have a problem with that.
697 Luckily, with GCC it's possible to work around that problem: One can define
698 NAN as being (0.0 / 0.0) and `isnan' as `f != f'. However, to test this
699 ``implementation'' the configure script needs to compile and run a short
700 test program. Obviously running a test program when doing a cross-
701 compilation is, well, challenging.
703 If you run into this problem, you can use the `--with-nan-emulation'
704 configure option to force the use of this implementation. We can't promise
705 that the compiled binary actually behaves as it should, but since NANs
706 are likely never passed to the libm you have a good chance to be lucky.
708 Likewise, collectd needs to know the layout of doubles in memory, in order
709 to craft uniform network packets over different architectures. For this, it
710 needs to know how to convert doubles into the memory layout used by x86. The
711 configure script tries to figure this out by compiling and running a few
712 small test programs. This is of course not possible when cross-compiling.
713 You can use the `--with-fp-layout' option to tell the configure script which
714 conversion method to assume. Valid arguments are:
716 * `nothing' (12345678 -> 12345678)
717 * `endianflip' (12345678 -> 87654321)
718 * `intswap' (12345678 -> 56781234)
724 For questions, bug reports, development information and basically all other
725 concerns please send an email to collectd's mailing list at
726 <collectd at verplant.org>.
728 For live discussion and more personal contact visit us in IRC, we're in
729 channel #collectd on freenode.
735 Florian octo Forster <octo at verplant.org>,
736 Sebastian tokkee Harl <sh at tokkee.org>,
737 and many contributors (see `AUTHORS').
739 Please send bug reports and patches to the mailing list, see `Contact'