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 CPU utilization: Time spent in the system, user, nice, idle, and related
41 CPU frequency (For laptops with speed step or a similar technology)
44 Executes SQL statements on various databases and interprets the returned
48 Mountpoint usage (Basically the values `df(1)' delivers)
51 Disk utilization: Sectors read/written, number of read/write actions,
52 average time an IO-operation took to complete.
55 DNS traffic: Query types, response codes, opcodes and traffic/octets
59 Email statistics: Count, traffic, spam scores and checks.
60 See collectd-email(5).
63 Amount of entropy available to the system.
66 Values gathered by a custom program or script.
70 Count the number of files in directories.
73 Harddisk temperatures using hddtempd.
76 Interface traffic: Number of octets, packets and errors for each
80 Iptables' counters: Number of bytes that were matched by a certain
84 IPMI (Intelligent Platform Management Interface) sensors information.
87 IPVS connection statistics (number of connections, octets and packets
88 for each service and destination).
89 See http://www.linuxvirtualserver.org/software/index.html.
92 IRQ counters: Frequency in which certain interrupts occur.
95 System load average over the last 1, 5 and 15 minutes.
98 CPU, disk and network I/O statistics from virtual machines.
101 Motherboard sensors: temperature, fanspeed and voltage information,
105 Statistics of the memcached distributed caching system.
106 <http://www.danga.com/memcached/>
109 Memory utilization: Memory occupied by running processes, page cache,
110 buffer cache and free.
113 Information provided by serial multimeters, such as the `Metex
117 MySQL server statistics: Commands issued, handlers triggered, thread
118 usage, query cache utilization and traffic/octets sent and received.
121 Very detailed Linux network interface and routing statistics. You can get
122 (detailed) information on interfaces, qdiscs, classes, and, if you can
123 make use of it, filters.
126 Receive values that were collected by other hosts. Large setups will
127 want to collect the data on one dedicated machine, and this is the
128 plugin of choice for that.
131 NFS Procedures: Which NFS command were called how often. Only NFSv2 and
135 Collects statistics from `nginx' (speak: engine X), a HTTP and mail
139 NTP daemon statistics: Local clock drift, offset to peers, etc.
142 Network UPS tools: UPS current, voltage, power, charge, utilisation,
143 temperature, etc. See upsd(8).
145 - onewire (EXPERIMENTAL!)
146 Read onewire sensors using the owcapu library of the owfs project.
147 Please read in collectd.conf(5) why this plugin is experimental.
150 The perl plugin implements a Perl-interpreter into collectd. You can
151 write your own plugins in Perl and return arbitrary values using this
152 API. See collectd-perl(5).
155 Network latency: Time to reach the default gateway or another given
159 PostgreSQL database statistics: active server connections, transaction
160 numbers, block IO, table row manipulations.
163 PowerDNS name server statistics.
166 Process counts: Number of running, sleeping, zombie, ... processes.
169 RRDtool caching daemon (RRDcacheD) statistics.
172 System sensors, accessed using lm_sensors: Voltages, temperatures and
176 RX and TX of serial interfaces. Linux only; needs root privileges.
179 Read values from SNMP (Simple Network Management Protocol) enabled
180 network devices such as switches, routers, thermometers, rack monitoring
181 servers, etc. See collectd-snmp(5).
184 Pages swapped out onto harddisk or whatever is called `swap' by the OS..
187 Follows (tails) logfiles, parses them by lines and submits matched
191 Bytes and operations read and written on tape devices. Solaris only.
194 Number of TCP connections to specific local and remote ports.
197 TeamSpeak2 server statistics.
200 Linux ACPI thermal zone information.
203 Users currently logged in.
206 Virtual memory statistics, e. g. the number of page-ins/-outs or the
207 number of pagefaults.
210 System resources used by Linux VServers.
211 See <http://linux-vserver.org/>.
214 Link quality of wireless cards. Linux only.
217 Bitrate and frequency of music played with XMMS.
219 * Output can be written or send to various destinations by the following
223 Write to comma separated values (CSV) files. This needs lots of
224 diskspace but is extremely portable and can be analysed with almost
225 every program that can analyse anything. Even Microsoft's Excel..
228 Send the data to a remote host to save the data somehow. This is useful
229 for large setups where the data should be saved by a dedicated machine.
232 Of course the values are propagated to plugins written in Perl, too, so
233 you can easily do weird stuff with the plugins we didn't dare think of
234 ;) See collectd-perl(5).
237 Output to round-robin-database (RRD) files using the RRDtool caching
238 daemon (RRDcacheD) - see rrdcached(1). That daemon provides a general
239 implementation of the caching done by the `rrdtool' plugin.
242 Output to round-robin-database (RRD) files using librrd. See rrdtool(1).
243 This is likely the most popular destination for such values. Since
244 updates to RRD-files are somewhat expensive this plugin can cache
245 updates to the files and write a bunch of updates at once, which lessens
249 One can query the values from the unixsock plugin whenever they're
250 needed. Please read collectd-unixsock(5) for a description on how that's
253 * Filtering and rewriting values dispatched to collectd can be done by the
257 Filter and rewrite value lists based on Perl-compatible regular
260 * Logging is, as everything in collectd, provided by plugins. The following
261 plugins keep up informed about what's going on:
264 Writes logmessages to a file or STDOUT/STDERR.
267 Log messages are propagated to plugins written in Perl as well.
268 See collectd-perl(5).
271 Logs to the standard UNIX logging mechanism, syslog.
273 * Notifications can be handled by the following plugins:
276 Send a desktop notification to a notification daemon, as defined in
277 the Desktop Notification Specification. To actually display the
278 notifications, notification-daemon is required.
279 See http://www.galago-project.org/specs/notification/.
282 Send an E-mail with the notification message to the configured
286 Execute a program or script to handle the notification.
287 See collectd-exec(5).
290 Writes the notification message to a file or STDOUT/STDERR.
293 Send the notification to a remote host to handle it somehow.
296 Notifications are propagated to plugins written in Perl as well.
297 See collectd-perl(5).
299 * Miscellaneous plugins:
302 Sets the hostname to an unique identifier. This is meant for setups
303 where each client may migrate to another physical host, possibly going
304 through one or more name changes in the process.
306 * Performance: Since collectd is running as a daemon it doesn't spend much
307 time starting up again and again. With the exception of the exec plugin no
308 processes are forked. Caching in output plugins, such as the rrdtool and
309 network plugins, makes sure your resources are used efficiently. Also,
310 since collectd is programmed multithreaded it benefits from hyperthreading
311 and multicore processors and makes sure that the daemon isn't idle if only
312 one plugins waits for an IO-operation to complete.
314 * Once set up, hardly any maintenance is necessary. Setup is kept as easy
315 as possible and the default values should be okay for most users.
321 * collectd's configuration file can be found at `sysconfdir'/collectd.conf.
322 Run `collectd -h' for a list of builtin defaults. See `collectd.conf(5)'
323 for a list of options and a syntax description.
325 * When the `csv' or `rrdtool' plugins are loaded they'll write the values to
326 files. The usual place for these files is beneath `/var/lib/collectd'.
328 * When using some of the plugins, collectd needs to run as user root, since
329 only root can do certain things, such as craft ICMP packages needed to ping
330 other hosts. collectd should NOT be installed setuid root since it can be
331 used to overwrite valuable files!
333 * Sample scripts to generate graphs reside in `contrib/' in the source
334 package or somewhere near `/usr/share/doc/collectd' in most distributions.
335 Please be aware that those script are meant as a starting point for your
336 own experiments.. Some of them require the `RRDs' Perl module.
337 (`librrds-perl' on Debian) If you have written a more sophisticated
338 solution please share it with us.
340 * The RRAs of the automatically created RRD files depend on the `step'
341 and `heartbeat' settings given. If change these settings you may need to
342 re-create the files, losing all data. Please be aware of that when changing
343 the values and read the rrdtool(1) manpage thoroughly.
346 collectd and chkrootkit
347 -----------------------
349 If you are using the `dns' plugin chkrootkit(1) will report collectd as a
350 packet sniffer ("<iface>: PACKET SNIFFER(/usr/sbin/collectd[<pid>])"). The
351 plugin captures all UDP packets on port 53 to analyze the DNS traffic. In
352 this case, collectd is a legitimate sniffer and the report should be
353 considered to be a false positive. However, you might want to check that
354 this really is collectd and not some other, illegitimate sniffer.
360 To compile collectd from source you will need:
362 * Usual suspects: C compiler, linker, preprocessor, make, ...
364 * A POSIX-threads (pthread) implementation.
365 Since gathering some statistics is slow (network connections, slow devices,
366 etc) the collectd is parallelized. The POSIX threads interface is being
367 used and should be found in various implementations for hopefully all
370 * CoreFoundation.framework and IOKit.framework (optional)
371 For compiling on Darwin in general and the `apple_sensors' plugin in
373 <http://developer.apple.com/corefoundation/>
376 If you want to use the `apache', `ascent', or `nginx' plugin.
377 <http://curl.haxx.se/>
380 Used by the `dbi' plugin to connect to various databases.
381 <http://libdbi.sourceforge.net/>
383 * libesmtp (optional)
384 For the `notify_email' plugin.
385 <http://www.stafford.uklinux.net/libesmtp/>
388 If present, the uuid plugin will check for UUID from HAL.
389 <http://hal.freedesktop.org/>
391 * libiptc (optional, if not found a version shipped with this distribution
392 can be used if the Linux kernel headers are available)
393 For querying iptables counters.
394 <http://netfilter.org/>
396 * libmysqlclient (optional)
397 Unsurprisingly used by the `mysql' plugin.
398 <http://dev.mysql.com/>
400 * libnetlink (optional)
401 Used, obviously, for the `netlink' plugin.
402 <http://www.linuxfoundation.org/en/Net:Iproute2>
404 * libnetsnmp (optional)
405 For the `snmp' plugin.
406 <http://www.net-snmp.org/>
408 * libnotify (optional)
409 For the `notify_desktop' plugin.
410 <http://www.galago-project.org/>
412 * liboping (optional, if not found a version shipped with this distribution
414 Used by the `ping' plugin to send and receive ICMP packets.
415 <http://verplant.org/liboping/>
417 * libowcapi (optional)
418 Used by the `onewire' plugin to read values from onewire sensors (or the
420 <http://www.owfs.org/>
423 Used to capture packets by the `dns' plugin.
424 <http://www.tcpdump.org/>
427 Used by the `filter_pcre' plugin.
428 <http://www.pcre.org/>
431 Obviously used by the `perl' plugin. The library has to be compiled with
432 ithread support (introduced in Perl 5.6.0).
433 <http://www.perl.org/>
436 The PostgreSQL C client library used by the `postgresql' plugin.
437 <http://www.postgresql.org/>
440 Used by the `rrdtool' and `rrdcached' plugins. The latter requires RRDtool
441 client support which was added after version 1.3 of RRDtool. Versions 1.0,
442 1.2 and 1.3 are known to work with the `rrdtool' plugin.
443 <http://oss.oetiker.ch/rrdtool/>
445 * librt, libsocket, libkstat, libdevinfo (optional)
446 Various standard Solaris libraries which provide system functions.
447 <http://developers.sun.com/solaris/>
449 * libsensors (optional)
450 To read from `lm_sensors', see the `sensors' plugin.
451 <http://www.lm-sensors.org/>
453 * libstatgrab (optional)
454 Used by various plugins to collect statistics on systems other than Linux
456 <http://www.i-scream.org/libstatgrab/>
458 * libupsclient/nut (optional)
459 For the `nut' plugin which queries nut's `upsd'.
460 <http://networkupstools.org/>
463 Collect statistics from virtual machines.
464 <http://libvirt.org/>
467 Parse XML data. This is needed for the `ascent' and `libvirt' plugins.
468 <http://xmlsoft.org/>
471 <http://www.xmms.org/>
474 Configuring / Compiling / Installing
475 ------------------------------------
477 To configure, build and install collectd with the default settings, run
478 `./configure && make && make install'. For detailed, generic instructions
479 see INSTALL. For a complete list of configure options and their description,
480 run `./configure --help'.
482 By default, the configure script will check for all build dependencies and
483 disable all plugins whose requirements cannot be fulfilled (any other plugin
484 will be enabled). To enable a plugin, install missing dependencies (see
485 section `Prerequisites' above) and rerun `configure'. If you specify the
486 `--enable-<plugin>' configure option, the script will fail if the depen-
487 dencies for the specified plugin are not met. If you specify the
488 `--disable-<plugin>' configure option, the plugin will not be built. Both
489 options are meant for package maintainers and should not be used in everyday
492 By default, collectd will be installed into `/opt/collectd'. You can adjust
493 this setting by specifying the `--prefix' configure option - see INSTALL for
494 details. If you pass DESTDIR=<path> to `make install', <path> will be
495 prefixed to all installation directories. This might be useful when creating
496 packages for collectd.
502 To compile correctly collectd needs to be able to initialize static
503 variables to NAN (Not A Number). Some C libraries, especially the GNU
504 libc, have a problem with that.
506 Luckily, with GCC it's possible to work around that problem: One can define
507 NAN as being (0.0 / 0.0) and `isnan' as `f != f'. However, to test this
508 ``implementation'' the configure script needs to compile and run a short
509 test program. Obviously running a test program when doing a cross-
510 compilation is, well, challenging.
512 If you run into this problem, you can use the `--with-nan-emulation'
513 configure option to force the use of this implementation. We can't promise
514 that the compiled binary actually behaves as it should, but since NANs
515 are likely never passed to the libm you have a good chance to be lucky.
521 For questions, bugreports, development information and basically all other
522 concerns please send an email to collectd's mailinglist at
523 <collectd at verplant.org>.
525 For live discussion and more personal contact visit us in IRC, we're in
526 channel #collectd on freenode.
532 Florian octo Forster <octo at verplant.org>,
533 Sebastian tokkee Harl <sh at tokkee.org>,
534 and many contributors (see `AUTHORS').
536 Please send bugreports and patches to the mailinglist, see `Contact' above.