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 Parse statistics from CouchDB JSON documents.
50 CPU utilization: Time spent in the system, user, nice, idle, and related
54 CPU frequency (For laptops with speed step or a similar technology)
57 Executes SQL statements on various databases and interprets the returned
61 Mountpoint usage (Basically the values `df(1)' delivers)
64 Disk utilization: Sectors read/written, number of read/write actions,
65 average time an IO-operation took to complete.
68 DNS traffic: Query types, response codes, opcodes and traffic/octets
72 Email statistics: Count, traffic, spam scores and checks.
73 See collectd-email(5).
76 Amount of entropy available to the system.
79 Values gathered by a custom program or script.
83 Count the number of files in directories.
86 Linux file-system based caching framework statistics.
89 Receive multicast traffic from Ganglia instances.
92 Harddisk temperatures using hddtempd.
95 Interface traffic: Number of octets, packets and errors for each
99 Iptables' counters: Number of bytes that were matched by a certain
103 IPMI (Intelligent Platform Management Interface) sensors information.
106 IPVS connection statistics (number of connections, octets and packets
107 for each service and destination).
108 See http://www.linuxvirtualserver.org/software/index.html.
111 IRQ counters: Frequency in which certain interrupts occur.
114 Integrates a `Java Virtual Machine' (JVM) to execute plugins in Java
115 bytecode. See “Configuring with libjvm” below.
118 System load average over the last 1, 5 and 15 minutes.
121 CPU, disk and network I/O statistics from virtual machines.
124 Motherboard sensors: temperature, fanspeed and voltage information,
128 Query and parse data from a memcache daemon (memcached).
131 Statistics of the memcached distributed caching system.
132 <http://www.danga.com/memcached/>
135 Memory utilization: Memory occupied by running processes, page cache,
136 buffer cache and free.
139 Information provided by serial multimeters, such as the `Metex
143 MySQL server statistics: Commands issued, handlers triggered, thread
144 usage, query cache utilization and traffic/octets sent and received.
147 Very detailed Linux network interface and routing statistics. You can get
148 (detailed) information on interfaces, qdiscs, classes, and, if you can
149 make use of it, filters.
152 Receive values that were collected by other hosts. Large setups will
153 want to collect the data on one dedicated machine, and this is the
154 plugin of choice for that.
157 NFS Procedures: Which NFS command were called how often. Only NFSv2 and
161 Collects statistics from `nginx' (speak: engine X), a HTTP and mail
165 NTP daemon statistics: Local clock drift, offset to peers, etc.
168 Network UPS tools: UPS current, voltage, power, charge, utilisation,
169 temperature, etc. See upsd(8).
171 - onewire (EXPERIMENTAL!)
172 Read onewire sensors using the owcapu library of the owfs project.
173 Please read in collectd.conf(5) why this plugin is experimental.
176 RX and TX of each client in openvpn-status.log (status-version 2).
177 <http://openvpn.net/index.php/documentation/howto.html>
180 Query data from an Oracle database.
183 The perl plugin implements a Perl-interpreter into collectd. You can
184 write your own plugins in Perl and return arbitrary values using this
185 API. See collectd-perl(5).
188 Network latency: Time to reach the default gateway or another given
192 PostgreSQL database statistics: active server connections, transaction
193 numbers, block IO, table row manipulations.
196 PowerDNS name server statistics.
199 Process counts: Number of running, sleeping, zombie, ... processes.
202 Counts various aspects of network protocols such as IP, TCP, UDP, etc.
205 RRDtool caching daemon (RRDcacheD) statistics.
208 System sensors, accessed using lm_sensors: Voltages, temperatures and
212 RX and TX of serial interfaces. Linux only; needs root privileges.
215 Read values from SNMP (Simple Network Management Protocol) enabled
216 network devices such as switches, routers, thermometers, rack monitoring
217 servers, etc. See collectd-snmp(5).
220 Pages swapped out onto harddisk or whatever is called `swap' by the OS..
223 Parse table-like structured files.
226 Follows (tails) logfiles, parses them by lines and submits matched
230 Bytes and operations read and written on tape devices. Solaris only.
233 Number of TCP connections to specific local and remote ports.
236 TeamSpeak2 server statistics.
239 Plugin to read values from `The Energy Detective' (TED).
242 Linux ACPI thermal zone information.
245 Record in and file size of a Tokyo Cabinet database file.
248 System uptime statistics.
251 Users currently logged in.
254 Virtual memory statistics, e. g. the number of page-ins/-outs or the
255 number of pagefaults.
258 System resources used by Linux VServers.
259 See <http://linux-vserver.org/>.
262 Link quality of wireless cards. Linux only.
265 Bitrate and frequency of music played with XMMS.
267 * Output can be written or send to various destinations by the following
271 Write to comma separated values (CSV) files. This needs lots of
272 diskspace but is extremely portable and can be analysed with almost
273 every program that can analyse anything. Even Microsoft's Excel..
276 Send the data to a remote host to save the data somehow. This is useful
277 for large setups where the data should be saved by a dedicated machine.
280 Of course the values are propagated to plugins written in Perl, too, so
281 you can easily do weird stuff with the plugins we didn't dare think of
282 ;) See collectd-perl(5).
285 Output to round-robin-database (RRD) files using the RRDtool caching
286 daemon (RRDcacheD) - see rrdcached(1). That daemon provides a general
287 implementation of the caching done by the `rrdtool' plugin.
290 Output to round-robin-database (RRD) files using librrd. See rrdtool(1).
291 This is likely the most popular destination for such values. Since
292 updates to RRD-files are somewhat expensive this plugin can cache
293 updates to the files and write a bunch of updates at once, which lessens
297 One can query the values from the unixsock plugin whenever they're
298 needed. Please read collectd-unixsock(5) for a description on how that's
301 * Logging is, as everything in collectd, provided by plugins. The following
302 plugins keep up informed about what's going on:
305 Writes logmessages to a file or STDOUT/STDERR.
308 Log messages are propagated to plugins written in Perl as well.
309 See collectd-perl(5).
312 Logs to the standard UNIX logging mechanism, syslog.
314 * Notifications can be handled by the following plugins:
317 Send a desktop notification to a notification daemon, as defined in
318 the Desktop Notification Specification. To actually display the
319 notifications, notification-daemon is required.
320 See http://www.galago-project.org/specs/notification/.
323 Send an E-mail with the notification message to the configured
327 Execute a program or script to handle the notification.
328 See collectd-exec(5).
331 Writes the notification message to a file or STDOUT/STDERR.
334 Send the notification to a remote host to handle it somehow.
337 Notifications are propagated to plugins written in Perl as well.
338 See collectd-perl(5).
340 * Value processing can be controlled using the "filter chain" infrastructure
341 and "matches" and "targets". The following plugins are available:
343 - match_empty_counter
344 Match counter values which are currently zero.
347 Match values by their identifier based on regular expressions.
350 Match values with an invalid timestamp.
353 Select values by their data sources' values.
355 - target_notification
356 Create and dispatch a notification.
359 Replace parts of an identifier using regular expressions.
362 Set (overwrite) entire parts of an identifier.
364 * Miscellaneous plugins:
367 Sets the hostname to an unique identifier. This is meant for setups
368 where each client may migrate to another physical host, possibly going
369 through one or more name changes in the process.
371 * Performance: Since collectd is running as a daemon it doesn't spend much
372 time starting up again and again. With the exception of the exec plugin no
373 processes are forked. Caching in output plugins, such as the rrdtool and
374 network plugins, makes sure your resources are used efficiently. Also,
375 since collectd is programmed multithreaded it benefits from hyperthreading
376 and multicore processors and makes sure that the daemon isn't idle if only
377 one plugins waits for an IO-operation to complete.
379 * Once set up, hardly any maintenance is necessary. Setup is kept as easy
380 as possible and the default values should be okay for most users.
386 * collectd's configuration file can be found at `sysconfdir'/collectd.conf.
387 Run `collectd -h' for a list of builtin defaults. See `collectd.conf(5)'
388 for a list of options and a syntax description.
390 * When the `csv' or `rrdtool' plugins are loaded they'll write the values to
391 files. The usual place for these files is beneath `/var/lib/collectd'.
393 * When using some of the plugins, collectd needs to run as user root, since
394 only root can do certain things, such as craft ICMP packages needed to ping
395 other hosts. collectd should NOT be installed setuid root since it can be
396 used to overwrite valuable files!
398 * Sample scripts to generate graphs reside in `contrib/' in the source
399 package or somewhere near `/usr/share/doc/collectd' in most distributions.
400 Please be aware that those script are meant as a starting point for your
401 own experiments.. Some of them require the `RRDs' Perl module.
402 (`librrds-perl' on Debian) If you have written a more sophisticated
403 solution please share it with us.
405 * The RRAs of the automatically created RRD files depend on the `step'
406 and `heartbeat' settings given. If change these settings you may need to
407 re-create the files, losing all data. Please be aware of that when changing
408 the values and read the rrdtool(1) manpage thoroughly.
411 collectd and chkrootkit
412 -----------------------
414 If you are using the `dns' plugin chkrootkit(1) will report collectd as a
415 packet sniffer ("<iface>: PACKET SNIFFER(/usr/sbin/collectd[<pid>])"). The
416 plugin captures all UDP packets on port 53 to analyze the DNS traffic. In
417 this case, collectd is a legitimate sniffer and the report should be
418 considered to be a false positive. However, you might want to check that
419 this really is collectd and not some other, illegitimate sniffer.
425 To compile collectd from source you will need:
427 * Usual suspects: C compiler, linker, preprocessor, make, ...
429 * A POSIX-threads (pthread) implementation.
430 Since gathering some statistics is slow (network connections, slow devices,
431 etc) the collectd is parallelized. The POSIX threads interface is being
432 used and should be found in various implementations for hopefully all
435 * CoreFoundation.framework and IOKit.framework (optional)
436 For compiling on Darwin in general and the `apple_sensors' plugin in
438 <http://developer.apple.com/corefoundation/>
440 * libclntsh (optional)
441 Used by the `oracle' plugin.
444 If you want to use the `apache', `ascent', `curl' or `nginx' plugin.
445 <http://curl.haxx.se/>
448 Used by the `dbi' plugin to connect to various databases.
449 <http://libdbi.sourceforge.net/>
451 * libesmtp (optional)
452 For the `notify_email' plugin.
453 <http://www.stafford.uklinux.net/libesmtp/>
455 * libganglia (optional)
456 Used by the `gmond' plugin to process data received from Ganglia.
458 * libgcrypt (optional)
459 Used by the `network' plugin for encryption and authentication.
462 If present, the uuid plugin will check for UUID from HAL.
463 <http://hal.freedesktop.org/>
465 * libiptc (optional, if not found a version shipped with this distribution
466 can be used if the Linux kernel headers are available)
467 For querying iptables counters.
468 <http://netfilter.org/>
471 Library that encapsulates the `Java Virtual Machine' (JVM). This library is
472 used by the Java plugin to execute Java bytecode. See “Configuring with
475 * libmemcached (optional)
476 Used by the `memcachec' plugin to connect to a memcache daemon.
478 * libmysqlclient (optional)
479 Unsurprisingly used by the `mysql' plugin.
480 <http://dev.mysql.com/>
482 * libnetlink (optional)
483 Used, obviously, for the `netlink' plugin.
484 <http://www.linuxfoundation.org/en/Net:Iproute2>
486 * libnetsnmp (optional)
487 For the `snmp' plugin.
488 <http://www.net-snmp.org/>
490 * libnotify (optional)
491 For the `notify_desktop' plugin.
492 <http://www.galago-project.org/>
494 * liboping (optional, if not found a version shipped with this distribution
496 Used by the `ping' plugin to send and receive ICMP packets.
497 <http://verplant.org/liboping/>
499 * libowcapi (optional)
500 Used by the `onewire' plugin to read values from onewire sensors (or the
502 <http://www.owfs.org/>
505 Used to capture packets by the `dns' plugin.
506 <http://www.tcpdump.org/>
509 Obviously used by the `perl' plugin. The library has to be compiled with
510 ithread support (introduced in Perl 5.6.0).
511 <http://www.perl.org/>
514 The PostgreSQL C client library used by the `postgresql' plugin.
515 <http://www.postgresql.org/>
518 Used by the `rrdtool' and `rrdcached' plugins. The latter requires RRDtool
519 client support which was added after version 1.3 of RRDtool. Versions 1.0,
520 1.2 and 1.3 are known to work with the `rrdtool' plugin.
521 <http://oss.oetiker.ch/rrdtool/>
523 * librt, libsocket, libkstat, libdevinfo (optional)
524 Various standard Solaris libraries which provide system functions.
525 <http://developers.sun.com/solaris/>
527 * libsensors (optional)
528 To read from `lm_sensors', see the `sensors' plugin.
529 <http://www.lm-sensors.org/>
531 * libstatgrab (optional)
532 Used by various plugins to collect statistics on systems other than Linux
534 <http://www.i-scream.org/libstatgrab/>
536 * libupsclient/nut (optional)
537 For the `nut' plugin which queries nut's `upsd'.
538 <http://networkupstools.org/>
541 Collect statistics from virtual machines.
542 <http://libvirt.org/>
545 Parse XML data. This is needed for the `ascent' and `libvirt' plugins.
546 <http://xmlsoft.org/>
549 <http://www.xmms.org/>
552 Parse JSON data. This is needed for the `couchdb' plugin.
553 <http://www.lloydforge.org/projects/yajl/>
555 Configuring / Compiling / Installing
556 ------------------------------------
558 To configure, build and install collectd with the default settings, run
559 `./configure && make && make install'. For detailed, generic instructions
560 see INSTALL. For a complete list of configure options and their description,
561 run `./configure --help'.
563 By default, the configure script will check for all build dependencies and
564 disable all plugins whose requirements cannot be fulfilled (any other plugin
565 will be enabled). To enable a plugin, install missing dependencies (see
566 section `Prerequisites' above) and rerun `configure'. If you specify the
567 `--enable-<plugin>' configure option, the script will fail if the depen-
568 dencies for the specified plugin are not met. In that case you can force the
569 plugin to be built using the `--enable-<plugin>=force' configure option.
570 This will most likely fail though unless you're working in a very unusual
571 setup and you really know what you're doing. If you specify the
572 `--disable-<plugin>' configure option, the plugin will not be built. If you
573 specify the `--enable-all-plugins' or `--disable-all-plugins' configure
574 options, all plugins will be enabled or disabled respectively by default.
575 Explicitly enabling or disabling a plugin overwrites the default for the
576 specified plugin. These options are meant for package maintainers and should
577 not be used in everyday situations.
579 By default, collectd will be installed into `/opt/collectd'. You can adjust
580 this setting by specifying the `--prefix' configure option - see INSTALL for
581 details. If you pass DESTDIR=<path> to `make install', <path> will be
582 prefixed to all installation directories. This might be useful when creating
583 packages for collectd.
585 Configuring with libjvm
586 -----------------------
588 To determine the location of the required files of a Java installation is not
589 an easy task, because the locations vary with your kernel (Linux, SunOS, …)
590 and with your architecture (x86, SPARC, …) and there is no ‘java-config’
591 script we could use. Configuration of the JVM library is therefore a bit
594 The easiest way to use the `--with-java=$JAVA_HOME' option, where
595 `$JAVA_HOME' is usually something like:
596 /usr/lib/jvm/java-1.5.0-sun-1.5.0.14
598 The configure script will then use find(1) to look for the following files:
604 If found, appropriate CPP-flags and LD-flags are set and the following
605 library checks succeed.
607 If this doesn't work for you, you have the possibility to specify CPP-flags,
608 C-flags and LD-flags for the ‘Java’ plugin by hand, using the following three
609 (environment) variables:
615 For example (shortened for demonstration purposes):
617 ./configure JAVA_CPPFLAGS="-I$JAVA_HOME/include -I$JAVA_HOME/include/linux"
619 Adding "-ljvm" to the JAVA_LDFLAGS is done automatically, you don't have to
625 To compile correctly collectd needs to be able to initialize static
626 variables to NAN (Not A Number). Some C libraries, especially the GNU
627 libc, have a problem with that.
629 Luckily, with GCC it's possible to work around that problem: One can define
630 NAN as being (0.0 / 0.0) and `isnan' as `f != f'. However, to test this
631 ``implementation'' the configure script needs to compile and run a short
632 test program. Obviously running a test program when doing a cross-
633 compilation is, well, challenging.
635 If you run into this problem, you can use the `--with-nan-emulation'
636 configure option to force the use of this implementation. We can't promise
637 that the compiled binary actually behaves as it should, but since NANs
638 are likely never passed to the libm you have a good chance to be lucky.
640 Likewise, collectd needs to know the layout of doubles in memory, in order
641 to craft uniform network packets over different architectures. For this, it
642 needs to know how to convert doubles into the memory layout used by x86. The
643 configure script tries to figure this out by compiling and running a few
644 small test programs. This is of course not possible when cross-compiling.
645 You can use the `--with-fp-layout' option to tell the configure script which
646 conversion method to assume. Valid arguments are:
648 * `nothing' (12345678 -> 12345678)
649 * `endianflip' (12345678 -> 87654321)
650 * `intswap' (12345678 -> 56781234)
656 For questions, bug reports, development information and basically all other
657 concerns please send an email to collectd's mailing list at
658 <collectd at verplant.org>.
660 For live discussion and more personal contact visit us in IRC, we're in
661 channel #collectd on freenode.
667 Florian octo Forster <octo at verplant.org>,
668 Sebastian tokkee Harl <sh at tokkee.org>,
669 and many contributors (see `AUTHORS').
671 Please send bug reports and patches to the mailing list, see `Contact'