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 Parse statistics from websites using regular expressions.
40 Name server and resolver statistics from the `statistics-channel'
41 interface of BIND 9.5, 9,6 and later.
44 Number of nf_conntrack entries.
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 Executes SQL statements on various databases and interprets the returned
58 Mountpoint usage (Basically the values `df(1)' delivers)
61 Disk utilization: Sectors read/written, number of read/write actions,
62 average time an IO-operation took to complete.
65 DNS traffic: Query types, response codes, opcodes and traffic/octets
69 Email statistics: Count, traffic, spam scores and checks.
70 See collectd-email(5).
73 Amount of entropy available to the system.
76 Values gathered by a custom program or script.
80 Count the number of files in directories.
83 Receive multicast traffic from Ganglia instances.
86 Harddisk temperatures using hddtempd.
89 Interface traffic: Number of octets, packets and errors for each
93 Iptables' counters: Number of bytes that were matched by a certain
97 IPMI (Intelligent Platform Management Interface) sensors information.
100 IPVS connection statistics (number of connections, octets and packets
101 for each service and destination).
102 See http://www.linuxvirtualserver.org/software/index.html.
105 IRQ counters: Frequency in which certain interrupts occur.
108 System load average over the last 1, 5 and 15 minutes.
111 CPU, disk and network I/O statistics from virtual machines.
114 Motherboard sensors: temperature, fanspeed and voltage information,
118 Statistics of the memcached distributed caching system.
119 <http://www.danga.com/memcached/>
122 Memory utilization: Memory occupied by running processes, page cache,
123 buffer cache and free.
126 Information provided by serial multimeters, such as the `Metex
130 MySQL server statistics: Commands issued, handlers triggered, thread
131 usage, query cache utilization and traffic/octets sent and received.
134 Very detailed Linux network interface and routing statistics. You can get
135 (detailed) information on interfaces, qdiscs, classes, and, if you can
136 make use of it, filters.
139 Receive values that were collected by other hosts. Large setups will
140 want to collect the data on one dedicated machine, and this is the
141 plugin of choice for that.
144 NFS Procedures: Which NFS command were called how often. Only NFSv2 and
148 Collects statistics from `nginx' (speak: engine X), a HTTP and mail
152 NTP daemon statistics: Local clock drift, offset to peers, etc.
155 Network UPS tools: UPS current, voltage, power, charge, utilisation,
156 temperature, etc. See upsd(8).
158 - onewire (EXPERIMENTAL!)
159 Read onewire sensors using the owcapu library of the owfs project.
160 Please read in collectd.conf(5) why this plugin is experimental.
163 RX and TX of each client in openvpn-status.log (status-version 2).
164 <http://openvpn.net/index.php/documentation/howto.html>
167 Query data from an Oracle database.
170 The perl plugin implements a Perl-interpreter into collectd. You can
171 write your own plugins in Perl and return arbitrary values using this
172 API. See collectd-perl(5).
175 Network latency: Time to reach the default gateway or another given
179 PostgreSQL database statistics: active server connections, transaction
180 numbers, block IO, table row manipulations.
183 PowerDNS name server statistics.
186 Process counts: Number of running, sleeping, zombie, ... processes.
189 Counts various aspects of network protocols such as IP, TCP, UDP, etc.
192 RRDtool caching daemon (RRDcacheD) statistics.
195 System sensors, accessed using lm_sensors: Voltages, temperatures and
199 RX and TX of serial interfaces. Linux only; needs root privileges.
202 Read values from SNMP (Simple Network Management Protocol) enabled
203 network devices such as switches, routers, thermometers, rack monitoring
204 servers, etc. See collectd-snmp(5).
207 Pages swapped out onto harddisk or whatever is called `swap' by the OS..
210 Follows (tails) logfiles, parses them by lines and submits matched
214 Bytes and operations read and written on tape devices. Solaris only.
217 Number of TCP connections to specific local and remote ports.
220 TeamSpeak2 server statistics.
223 Plugin to read values from `The Energy Detective' (TED).
226 Linux ACPI thermal zone information.
229 Users currently logged in.
232 Virtual memory statistics, e. g. the number of page-ins/-outs or the
233 number of pagefaults.
236 System resources used by Linux VServers.
237 See <http://linux-vserver.org/>.
240 Link quality of wireless cards. Linux only.
243 Bitrate and frequency of music played with XMMS.
245 * Output can be written or send to various destinations by the following
249 Write to comma separated values (CSV) files. This needs lots of
250 diskspace but is extremely portable and can be analysed with almost
251 every program that can analyse anything. Even Microsoft's Excel..
254 Send the data to a remote host to save the data somehow. This is useful
255 for large setups where the data should be saved by a dedicated machine.
258 Of course the values are propagated to plugins written in Perl, too, so
259 you can easily do weird stuff with the plugins we didn't dare think of
260 ;) See collectd-perl(5).
263 Output to round-robin-database (RRD) files using the RRDtool caching
264 daemon (RRDcacheD) - see rrdcached(1). That daemon provides a general
265 implementation of the caching done by the `rrdtool' plugin.
268 Output to round-robin-database (RRD) files using librrd. See rrdtool(1).
269 This is likely the most popular destination for such values. Since
270 updates to RRD-files are somewhat expensive this plugin can cache
271 updates to the files and write a bunch of updates at once, which lessens
275 One can query the values from the unixsock plugin whenever they're
276 needed. Please read collectd-unixsock(5) for a description on how that's
279 * Logging is, as everything in collectd, provided by plugins. The following
280 plugins keep up informed about what's going on:
283 Writes logmessages to a file or STDOUT/STDERR.
286 Log messages are propagated to plugins written in Perl as well.
287 See collectd-perl(5).
290 Logs to the standard UNIX logging mechanism, syslog.
292 * Notifications can be handled by the following plugins:
295 Send a desktop notification to a notification daemon, as defined in
296 the Desktop Notification Specification. To actually display the
297 notifications, notification-daemon is required.
298 See http://www.galago-project.org/specs/notification/.
301 Send an E-mail with the notification message to the configured
305 Execute a program or script to handle the notification.
306 See collectd-exec(5).
309 Writes the notification message to a file or STDOUT/STDERR.
312 Send the notification to a remote host to handle it somehow.
315 Notifications are propagated to plugins written in Perl as well.
316 See collectd-perl(5).
318 * Value processing can be controlled using the "filter chain" infrastructure
319 and "matches" and "targets". The following plugins are available:
322 Match values by their identifier based on regular expressions.
325 Match values with an invalid timestamp.
328 Select values by their data sources' values.
330 - target_notification
331 Create and dispatch a notification.
334 Replace parts of an identifier using regular expressions.
337 Set (overwrite) entire parts of an identifier.
339 * Miscellaneous plugins:
342 Sets the hostname to an unique identifier. This is meant for setups
343 where each client may migrate to another physical host, possibly going
344 through one or more name changes in the process.
346 * Performance: Since collectd is running as a daemon it doesn't spend much
347 time starting up again and again. With the exception of the exec plugin no
348 processes are forked. Caching in output plugins, such as the rrdtool and
349 network plugins, makes sure your resources are used efficiently. Also,
350 since collectd is programmed multithreaded it benefits from hyperthreading
351 and multicore processors and makes sure that the daemon isn't idle if only
352 one plugins waits for an IO-operation to complete.
354 * Once set up, hardly any maintenance is necessary. Setup is kept as easy
355 as possible and the default values should be okay for most users.
361 * collectd's configuration file can be found at `sysconfdir'/collectd.conf.
362 Run `collectd -h' for a list of builtin defaults. See `collectd.conf(5)'
363 for a list of options and a syntax description.
365 * When the `csv' or `rrdtool' plugins are loaded they'll write the values to
366 files. The usual place for these files is beneath `/var/lib/collectd'.
368 * When using some of the plugins, collectd needs to run as user root, since
369 only root can do certain things, such as craft ICMP packages needed to ping
370 other hosts. collectd should NOT be installed setuid root since it can be
371 used to overwrite valuable files!
373 * Sample scripts to generate graphs reside in `contrib/' in the source
374 package or somewhere near `/usr/share/doc/collectd' in most distributions.
375 Please be aware that those script are meant as a starting point for your
376 own experiments.. Some of them require the `RRDs' Perl module.
377 (`librrds-perl' on Debian) If you have written a more sophisticated
378 solution please share it with us.
380 * The RRAs of the automatically created RRD files depend on the `step'
381 and `heartbeat' settings given. If change these settings you may need to
382 re-create the files, losing all data. Please be aware of that when changing
383 the values and read the rrdtool(1) manpage thoroughly.
386 collectd and chkrootkit
387 -----------------------
389 If you are using the `dns' plugin chkrootkit(1) will report collectd as a
390 packet sniffer ("<iface>: PACKET SNIFFER(/usr/sbin/collectd[<pid>])"). The
391 plugin captures all UDP packets on port 53 to analyze the DNS traffic. In
392 this case, collectd is a legitimate sniffer and the report should be
393 considered to be a false positive. However, you might want to check that
394 this really is collectd and not some other, illegitimate sniffer.
400 To compile collectd from source you will need:
402 * Usual suspects: C compiler, linker, preprocessor, make, ...
404 * A POSIX-threads (pthread) implementation.
405 Since gathering some statistics is slow (network connections, slow devices,
406 etc) the collectd is parallelized. The POSIX threads interface is being
407 used and should be found in various implementations for hopefully all
410 * CoreFoundation.framework and IOKit.framework (optional)
411 For compiling on Darwin in general and the `apple_sensors' plugin in
413 <http://developer.apple.com/corefoundation/>
415 * libclntsh (optional)
416 Used by the `oracle' plugin.
419 If you want to use the `apache', `ascent', `curl' or `nginx' plugin.
420 <http://curl.haxx.se/>
423 Used by the `dbi' plugin to connect to various databases.
424 <http://libdbi.sourceforge.net/>
426 * libesmtp (optional)
427 For the `notify_email' plugin.
428 <http://www.stafford.uklinux.net/libesmtp/>
431 If present, the uuid plugin will check for UUID from HAL.
432 <http://hal.freedesktop.org/>
434 * libiptc (optional, if not found a version shipped with this distribution
435 can be used if the Linux kernel headers are available)
436 For querying iptables counters.
437 <http://netfilter.org/>
439 * libmysqlclient (optional)
440 Unsurprisingly used by the `mysql' plugin.
441 <http://dev.mysql.com/>
443 * libnetlink (optional)
444 Used, obviously, for the `netlink' plugin.
445 <http://www.linuxfoundation.org/en/Net:Iproute2>
447 * libnetsnmp (optional)
448 For the `snmp' plugin.
449 <http://www.net-snmp.org/>
451 * libnotify (optional)
452 For the `notify_desktop' plugin.
453 <http://www.galago-project.org/>
455 * liboping (optional, if not found a version shipped with this distribution
457 Used by the `ping' plugin to send and receive ICMP packets.
458 <http://verplant.org/liboping/>
460 * libowcapi (optional)
461 Used by the `onewire' plugin to read values from onewire sensors (or the
463 <http://www.owfs.org/>
466 Used to capture packets by the `dns' plugin.
467 <http://www.tcpdump.org/>
470 Obviously used by the `perl' plugin. The library has to be compiled with
471 ithread support (introduced in Perl 5.6.0).
472 <http://www.perl.org/>
475 The PostgreSQL C client library used by the `postgresql' plugin.
476 <http://www.postgresql.org/>
479 Used by the `rrdtool' and `rrdcached' plugins. The latter requires RRDtool
480 client support which was added after version 1.3 of RRDtool. Versions 1.0,
481 1.2 and 1.3 are known to work with the `rrdtool' plugin.
482 <http://oss.oetiker.ch/rrdtool/>
484 * librt, libsocket, libkstat, libdevinfo (optional)
485 Various standard Solaris libraries which provide system functions.
486 <http://developers.sun.com/solaris/>
488 * libsensors (optional)
489 To read from `lm_sensors', see the `sensors' plugin.
490 <http://www.lm-sensors.org/>
492 * libstatgrab (optional)
493 Used by various plugins to collect statistics on systems other than Linux
495 <http://www.i-scream.org/libstatgrab/>
497 * libupsclient/nut (optional)
498 For the `nut' plugin which queries nut's `upsd'.
499 <http://networkupstools.org/>
502 Collect statistics from virtual machines.
503 <http://libvirt.org/>
506 Parse XML data. This is needed for the `ascent' and `libvirt' plugins.
507 <http://xmlsoft.org/>
510 <http://www.xmms.org/>
513 Configuring / Compiling / Installing
514 ------------------------------------
516 To configure, build and install collectd with the default settings, run
517 `./configure && make && make install'. For detailed, generic instructions
518 see INSTALL. For a complete list of configure options and their description,
519 run `./configure --help'.
521 By default, the configure script will check for all build dependencies and
522 disable all plugins whose requirements cannot be fulfilled (any other plugin
523 will be enabled). To enable a plugin, install missing dependencies (see
524 section `Prerequisites' above) and rerun `configure'. If you specify the
525 `--enable-<plugin>' configure option, the script will fail if the depen-
526 dencies for the specified plugin are not met. If you specify the
527 `--disable-<plugin>' configure option, the plugin will not be built. Both
528 options are meant for package maintainers and should not be used in everyday
531 By default, collectd will be installed into `/opt/collectd'. You can adjust
532 this setting by specifying the `--prefix' configure option - see INSTALL for
533 details. If you pass DESTDIR=<path> to `make install', <path> will be
534 prefixed to all installation directories. This might be useful when creating
535 packages for collectd.
541 To compile correctly collectd needs to be able to initialize static
542 variables to NAN (Not A Number). Some C libraries, especially the GNU
543 libc, have a problem with that.
545 Luckily, with GCC it's possible to work around that problem: One can define
546 NAN as being (0.0 / 0.0) and `isnan' as `f != f'. However, to test this
547 ``implementation'' the configure script needs to compile and run a short
548 test program. Obviously running a test program when doing a cross-
549 compilation is, well, challenging.
551 If you run into this problem, you can use the `--with-nan-emulation'
552 configure option to force the use of this implementation. We can't promise
553 that the compiled binary actually behaves as it should, but since NANs
554 are likely never passed to the libm you have a good chance to be lucky.
556 Likewise, collectd needs to know the layout of doubles in memory, in order
557 to craft uniform network packets over different architectures. For this, it
558 needs to know how to convert doubles into the memory layout used by x86. The
559 configure script tries to figure this out by compiling and running a few
560 small test programs. This is of course not possible when cross-compiling.
561 You can use the `--with-fp-layout' option to tell the configure script which
562 conversion method to assume. Valid arguments are:
564 * `nothing' (12345678 -> 12345678)
565 * `endianflip' (12345678 -> 87654321)
566 * `intswap' (12345678 -> 56781234)
572 For questions, bug reports, development information and basically all other
573 concerns please send an email to collectd's mailing list at
574 <collectd at verplant.org>.
576 For live discussion and more personal contact visit us in IRC, we're in
577 channel #collectd on freenode.
583 Florian octo Forster <octo at verplant.org>,
584 Sebastian tokkee Harl <sh at tokkee.org>,
585 and many contributors (see `AUTHORS').
587 Please send bug reports and patches to the mailing list, see `Contact'