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 CPU utilization: Time spent in the system, user, nice, idle, and related
45 CPU frequency (For laptops with speed step or a similar technology)
48 Executes SQL statements on various databases and interprets the returned
52 Mountpoint usage (Basically the values `df(1)' delivers)
55 Disk utilization: Sectors read/written, number of read/write actions,
56 average time an IO-operation took to complete.
59 DNS traffic: Query types, response codes, opcodes and traffic/octets
63 Email statistics: Count, traffic, spam scores and checks.
64 See collectd-email(5).
67 Amount of entropy available to the system.
70 Values gathered by a custom program or script.
74 Count the number of files in directories.
77 Harddisk temperatures using hddtempd.
80 Interface traffic: Number of octets, packets and errors for each
84 Iptables' counters: Number of bytes that were matched by a certain
88 IPMI (Intelligent Platform Management Interface) sensors information.
91 IPVS connection statistics (number of connections, octets and packets
92 for each service and destination).
93 See http://www.linuxvirtualserver.org/software/index.html.
96 IRQ counters: Frequency in which certain interrupts occur.
99 System load average over the last 1, 5 and 15 minutes.
102 CPU, disk and network I/O statistics from virtual machines.
105 Motherboard sensors: temperature, fanspeed and voltage information,
109 Statistics of the memcached distributed caching system.
110 <http://www.danga.com/memcached/>
113 Memory utilization: Memory occupied by running processes, page cache,
114 buffer cache and free.
117 Information provided by serial multimeters, such as the `Metex
121 MySQL server statistics: Commands issued, handlers triggered, thread
122 usage, query cache utilization and traffic/octets sent and received.
125 Very detailed Linux network interface and routing statistics. You can get
126 (detailed) information on interfaces, qdiscs, classes, and, if you can
127 make use of it, filters.
130 Receive values that were collected by other hosts. Large setups will
131 want to collect the data on one dedicated machine, and this is the
132 plugin of choice for that.
135 NFS Procedures: Which NFS command were called how often. Only NFSv2 and
139 Collects statistics from `nginx' (speak: engine X), a HTTP and mail
143 NTP daemon statistics: Local clock drift, offset to peers, etc.
146 Network UPS tools: UPS current, voltage, power, charge, utilisation,
147 temperature, etc. See upsd(8).
149 - onewire (EXPERIMENTAL!)
150 Read onewire sensors using the owcapu library of the owfs project.
151 Please read in collectd.conf(5) why this plugin is experimental.
154 RX and TX of each client in openvpn-status.log (status-version 2).
155 <http://openvpn.net/index.php/documentation/howto.html>
158 The perl plugin implements a Perl-interpreter into collectd. You can
159 write your own plugins in Perl and return arbitrary values using this
160 API. See collectd-perl(5).
163 Network latency: Time to reach the default gateway or another given
167 PostgreSQL database statistics: active server connections, transaction
168 numbers, block IO, table row manipulations.
171 PowerDNS name server statistics.
174 Process counts: Number of running, sleeping, zombie, ... processes.
177 RRDtool caching daemon (RRDcacheD) statistics.
180 System sensors, accessed using lm_sensors: Voltages, temperatures and
184 RX and TX of serial interfaces. Linux only; needs root privileges.
187 Read values from SNMP (Simple Network Management Protocol) enabled
188 network devices such as switches, routers, thermometers, rack monitoring
189 servers, etc. See collectd-snmp(5).
192 Pages swapped out onto harddisk or whatever is called `swap' by the OS..
195 Follows (tails) logfiles, parses them by lines and submits matched
199 Bytes and operations read and written on tape devices. Solaris only.
202 Number of TCP connections to specific local and remote ports.
205 TeamSpeak2 server statistics.
208 Linux ACPI thermal zone information.
211 Users currently logged in.
214 Virtual memory statistics, e. g. the number of page-ins/-outs or the
215 number of pagefaults.
218 System resources used by Linux VServers.
219 See <http://linux-vserver.org/>.
222 Link quality of wireless cards. Linux only.
225 Bitrate and frequency of music played with XMMS.
227 * Output can be written or send to various destinations by the following
231 Write to comma separated values (CSV) files. This needs lots of
232 diskspace but is extremely portable and can be analysed with almost
233 every program that can analyse anything. Even Microsoft's Excel..
236 Send the data to a remote host to save the data somehow. This is useful
237 for large setups where the data should be saved by a dedicated machine.
240 Of course the values are propagated to plugins written in Perl, too, so
241 you can easily do weird stuff with the plugins we didn't dare think of
242 ;) See collectd-perl(5).
245 Output to round-robin-database (RRD) files using the RRDtool caching
246 daemon (RRDcacheD) - see rrdcached(1). That daemon provides a general
247 implementation of the caching done by the `rrdtool' plugin.
250 Output to round-robin-database (RRD) files using librrd. See rrdtool(1).
251 This is likely the most popular destination for such values. Since
252 updates to RRD-files are somewhat expensive this plugin can cache
253 updates to the files and write a bunch of updates at once, which lessens
257 One can query the values from the unixsock plugin whenever they're
258 needed. Please read collectd-unixsock(5) for a description on how that's
261 * Filtering and rewriting values dispatched to collectd can be done by the
265 Filter and rewrite value lists based on Perl-compatible regular
268 * Logging is, as everything in collectd, provided by plugins. The following
269 plugins keep up informed about what's going on:
272 Writes logmessages to a file or STDOUT/STDERR.
275 Log messages are propagated to plugins written in Perl as well.
276 See collectd-perl(5).
279 Logs to the standard UNIX logging mechanism, syslog.
281 * Notifications can be handled by the following plugins:
284 Send a desktop notification to a notification daemon, as defined in
285 the Desktop Notification Specification. To actually display the
286 notifications, notification-daemon is required.
287 See http://www.galago-project.org/specs/notification/.
290 Send an E-mail with the notification message to the configured
294 Execute a program or script to handle the notification.
295 See collectd-exec(5).
298 Writes the notification message to a file or STDOUT/STDERR.
301 Send the notification to a remote host to handle it somehow.
304 Notifications are propagated to plugins written in Perl as well.
305 See collectd-perl(5).
307 * Miscellaneous plugins:
310 Sets the hostname to an unique identifier. This is meant for setups
311 where each client may migrate to another physical host, possibly going
312 through one or more name changes in the process.
314 * Performance: Since collectd is running as a daemon it doesn't spend much
315 time starting up again and again. With the exception of the exec plugin no
316 processes are forked. Caching in output plugins, such as the rrdtool and
317 network plugins, makes sure your resources are used efficiently. Also,
318 since collectd is programmed multithreaded it benefits from hyperthreading
319 and multicore processors and makes sure that the daemon isn't idle if only
320 one plugins waits for an IO-operation to complete.
322 * Once set up, hardly any maintenance is necessary. Setup is kept as easy
323 as possible and the default values should be okay for most users.
329 * collectd's configuration file can be found at `sysconfdir'/collectd.conf.
330 Run `collectd -h' for a list of builtin defaults. See `collectd.conf(5)'
331 for a list of options and a syntax description.
333 * When the `csv' or `rrdtool' plugins are loaded they'll write the values to
334 files. The usual place for these files is beneath `/var/lib/collectd'.
336 * When using some of the plugins, collectd needs to run as user root, since
337 only root can do certain things, such as craft ICMP packages needed to ping
338 other hosts. collectd should NOT be installed setuid root since it can be
339 used to overwrite valuable files!
341 * Sample scripts to generate graphs reside in `contrib/' in the source
342 package or somewhere near `/usr/share/doc/collectd' in most distributions.
343 Please be aware that those script are meant as a starting point for your
344 own experiments.. Some of them require the `RRDs' Perl module.
345 (`librrds-perl' on Debian) If you have written a more sophisticated
346 solution please share it with us.
348 * The RRAs of the automatically created RRD files depend on the `step'
349 and `heartbeat' settings given. If change these settings you may need to
350 re-create the files, losing all data. Please be aware of that when changing
351 the values and read the rrdtool(1) manpage thoroughly.
354 collectd and chkrootkit
355 -----------------------
357 If you are using the `dns' plugin chkrootkit(1) will report collectd as a
358 packet sniffer ("<iface>: PACKET SNIFFER(/usr/sbin/collectd[<pid>])"). The
359 plugin captures all UDP packets on port 53 to analyze the DNS traffic. In
360 this case, collectd is a legitimate sniffer and the report should be
361 considered to be a false positive. However, you might want to check that
362 this really is collectd and not some other, illegitimate sniffer.
368 To compile collectd from source you will need:
370 * Usual suspects: C compiler, linker, preprocessor, make, ...
372 * A POSIX-threads (pthread) implementation.
373 Since gathering some statistics is slow (network connections, slow devices,
374 etc) the collectd is parallelized. The POSIX threads interface is being
375 used and should be found in various implementations for hopefully all
378 * CoreFoundation.framework and IOKit.framework (optional)
379 For compiling on Darwin in general and the `apple_sensors' plugin in
381 <http://developer.apple.com/corefoundation/>
384 If you want to use the `apache', `ascent', or `nginx' plugin.
385 <http://curl.haxx.se/>
388 Used by the `dbi' plugin to connect to various databases.
389 <http://libdbi.sourceforge.net/>
391 * libesmtp (optional)
392 For the `notify_email' plugin.
393 <http://www.stafford.uklinux.net/libesmtp/>
396 If present, the uuid plugin will check for UUID from HAL.
397 <http://hal.freedesktop.org/>
399 * libiptc (optional, if not found a version shipped with this distribution
400 can be used if the Linux kernel headers are available)
401 For querying iptables counters.
402 <http://netfilter.org/>
404 * libmysqlclient (optional)
405 Unsurprisingly used by the `mysql' plugin.
406 <http://dev.mysql.com/>
408 * libnetlink (optional)
409 Used, obviously, for the `netlink' plugin.
410 <http://www.linuxfoundation.org/en/Net:Iproute2>
412 * libnetsnmp (optional)
413 For the `snmp' plugin.
414 <http://www.net-snmp.org/>
416 * libnotify (optional)
417 For the `notify_desktop' plugin.
418 <http://www.galago-project.org/>
420 * liboping (optional, if not found a version shipped with this distribution
422 Used by the `ping' plugin to send and receive ICMP packets.
423 <http://verplant.org/liboping/>
425 * libowcapi (optional)
426 Used by the `onewire' plugin to read values from onewire sensors (or the
428 <http://www.owfs.org/>
431 Used to capture packets by the `dns' plugin.
432 <http://www.tcpdump.org/>
435 Used by the `filter_pcre' plugin.
436 <http://www.pcre.org/>
439 Obviously used by the `perl' plugin. The library has to be compiled with
440 ithread support (introduced in Perl 5.6.0).
441 <http://www.perl.org/>
444 The PostgreSQL C client library used by the `postgresql' plugin.
445 <http://www.postgresql.org/>
448 Used by the `rrdtool' and `rrdcached' plugins. The latter requires RRDtool
449 client support which was added after version 1.3 of RRDtool. Versions 1.0,
450 1.2 and 1.3 are known to work with the `rrdtool' plugin.
451 <http://oss.oetiker.ch/rrdtool/>
453 * librt, libsocket, libkstat, libdevinfo (optional)
454 Various standard Solaris libraries which provide system functions.
455 <http://developers.sun.com/solaris/>
457 * libsensors (optional)
458 To read from `lm_sensors', see the `sensors' plugin.
459 <http://www.lm-sensors.org/>
461 * libstatgrab (optional)
462 Used by various plugins to collect statistics on systems other than Linux
464 <http://www.i-scream.org/libstatgrab/>
466 * libupsclient/nut (optional)
467 For the `nut' plugin which queries nut's `upsd'.
468 <http://networkupstools.org/>
471 Collect statistics from virtual machines.
472 <http://libvirt.org/>
475 Parse XML data. This is needed for the `ascent' and `libvirt' plugins.
476 <http://xmlsoft.org/>
479 <http://www.xmms.org/>
482 Configuring / Compiling / Installing
483 ------------------------------------
485 To configure, build and install collectd with the default settings, run
486 `./configure && make && make install'. For detailed, generic instructions
487 see INSTALL. For a complete list of configure options and their description,
488 run `./configure --help'.
490 By default, the configure script will check for all build dependencies and
491 disable all plugins whose requirements cannot be fulfilled (any other plugin
492 will be enabled). To enable a plugin, install missing dependencies (see
493 section `Prerequisites' above) and rerun `configure'. If you specify the
494 `--enable-<plugin>' configure option, the script will fail if the depen-
495 dencies for the specified plugin are not met. If you specify the
496 `--disable-<plugin>' configure option, the plugin will not be built. Both
497 options are meant for package maintainers and should not be used in everyday
500 By default, collectd will be installed into `/opt/collectd'. You can adjust
501 this setting by specifying the `--prefix' configure option - see INSTALL for
502 details. If you pass DESTDIR=<path> to `make install', <path> will be
503 prefixed to all installation directories. This might be useful when creating
504 packages for collectd.
510 To compile correctly collectd needs to be able to initialize static
511 variables to NAN (Not A Number). Some C libraries, especially the GNU
512 libc, have a problem with that.
514 Luckily, with GCC it's possible to work around that problem: One can define
515 NAN as being (0.0 / 0.0) and `isnan' as `f != f'. However, to test this
516 ``implementation'' the configure script needs to compile and run a short
517 test program. Obviously running a test program when doing a cross-
518 compilation is, well, challenging.
520 If you run into this problem, you can use the `--with-nan-emulation'
521 configure option to force the use of this implementation. We can't promise
522 that the compiled binary actually behaves as it should, but since NANs
523 are likely never passed to the libm you have a good chance to be lucky.
529 For questions, bugreports, development information and basically all other
530 concerns please send an email to collectd's mailinglist at
531 <collectd at verplant.org>.
533 For live discussion and more personal contact visit us in IRC, we're in
534 channel #collectd on freenode.
540 Florian octo Forster <octo at verplant.org>,
541 Sebastian tokkee Harl <sh at tokkee.org>,
542 and many contributors (see `AUTHORS').
544 Please send bugreports and patches to the mailinglist, see `Contact' above.