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 CPU utilization: Time spent in the system, user, nice, idle, and related
48 CPU frequency (For laptops with speed step or a similar technology)
51 Executes SQL statements on various databases and interprets the returned
55 Mountpoint usage (Basically the values `df(1)' delivers)
58 Disk utilization: Sectors read/written, number of read/write actions,
59 average time an IO-operation took to complete.
62 DNS traffic: Query types, response codes, opcodes and traffic/octets
66 Email statistics: Count, traffic, spam scores and checks.
67 See collectd-email(5).
70 Amount of entropy available to the system.
73 Values gathered by a custom program or script.
77 Count the number of files in directories.
80 Receive multicast traffic from Ganglia instances.
83 Harddisk temperatures using hddtempd.
86 Interface traffic: Number of octets, packets and errors for each
90 Iptables' counters: Number of bytes that were matched by a certain
94 IPMI (Intelligent Platform Management Interface) sensors information.
97 IPVS connection statistics (number of connections, octets and packets
98 for each service and destination).
99 See http://www.linuxvirtualserver.org/software/index.html.
102 IRQ counters: Frequency in which certain interrupts occur.
105 System load average over the last 1, 5 and 15 minutes.
108 CPU, disk and network I/O statistics from virtual machines.
111 Motherboard sensors: temperature, fanspeed and voltage information,
115 Statistics of the memcached distributed caching system.
116 <http://www.danga.com/memcached/>
119 Memory utilization: Memory occupied by running processes, page cache,
120 buffer cache and free.
123 Information provided by serial multimeters, such as the `Metex
127 MySQL server statistics: Commands issued, handlers triggered, thread
128 usage, query cache utilization and traffic/octets sent and received.
131 Very detailed Linux network interface and routing statistics. You can get
132 (detailed) information on interfaces, qdiscs, classes, and, if you can
133 make use of it, filters.
136 Receive values that were collected by other hosts. Large setups will
137 want to collect the data on one dedicated machine, and this is the
138 plugin of choice for that.
141 NFS Procedures: Which NFS command were called how often. Only NFSv2 and
145 Collects statistics from `nginx' (speak: engine X), a HTTP and mail
149 NTP daemon statistics: Local clock drift, offset to peers, etc.
152 Network UPS tools: UPS current, voltage, power, charge, utilisation,
153 temperature, etc. See upsd(8).
155 - onewire (EXPERIMENTAL!)
156 Read onewire sensors using the owcapu library of the owfs project.
157 Please read in collectd.conf(5) why this plugin is experimental.
160 RX and TX of each client in openvpn-status.log (status-version 2).
161 <http://openvpn.net/index.php/documentation/howto.html>
164 Query data from an Oracle database.
167 The perl plugin implements a Perl-interpreter into collectd. You can
168 write your own plugins in Perl and return arbitrary values using this
169 API. See collectd-perl(5).
172 Network latency: Time to reach the default gateway or another given
176 PostgreSQL database statistics: active server connections, transaction
177 numbers, block IO, table row manipulations.
180 PowerDNS name server statistics.
183 Process counts: Number of running, sleeping, zombie, ... processes.
186 Counts various aspects of network protocols such as IP, TCP, UDP, etc.
189 RRDtool caching daemon (RRDcacheD) statistics.
192 System sensors, accessed using lm_sensors: Voltages, temperatures and
196 RX and TX of serial interfaces. Linux only; needs root privileges.
199 Read values from SNMP (Simple Network Management Protocol) enabled
200 network devices such as switches, routers, thermometers, rack monitoring
201 servers, etc. See collectd-snmp(5).
204 Pages swapped out onto harddisk or whatever is called `swap' by the OS..
207 Follows (tails) logfiles, parses them by lines and submits matched
211 Bytes and operations read and written on tape devices. Solaris only.
214 Number of TCP connections to specific local and remote ports.
217 TeamSpeak2 server statistics.
220 Linux ACPI thermal zone information.
223 Users currently logged in.
226 Virtual memory statistics, e. g. the number of page-ins/-outs or the
227 number of pagefaults.
230 System resources used by Linux VServers.
231 See <http://linux-vserver.org/>.
234 Link quality of wireless cards. Linux only.
237 Bitrate and frequency of music played with XMMS.
239 * Output can be written or send to various destinations by the following
243 Write to comma separated values (CSV) files. This needs lots of
244 diskspace but is extremely portable and can be analysed with almost
245 every program that can analyse anything. Even Microsoft's Excel..
248 Send the data to a remote host to save the data somehow. This is useful
249 for large setups where the data should be saved by a dedicated machine.
252 Of course the values are propagated to plugins written in Perl, too, so
253 you can easily do weird stuff with the plugins we didn't dare think of
254 ;) See collectd-perl(5).
257 Output to round-robin-database (RRD) files using the RRDtool caching
258 daemon (RRDcacheD) - see rrdcached(1). That daemon provides a general
259 implementation of the caching done by the `rrdtool' plugin.
262 Output to round-robin-database (RRD) files using librrd. See rrdtool(1).
263 This is likely the most popular destination for such values. Since
264 updates to RRD-files are somewhat expensive this plugin can cache
265 updates to the files and write a bunch of updates at once, which lessens
269 One can query the values from the unixsock plugin whenever they're
270 needed. Please read collectd-unixsock(5) for a description on how that's
273 * Logging is, as everything in collectd, provided by plugins. The following
274 plugins keep up informed about what's going on:
277 Writes logmessages to a file or STDOUT/STDERR.
280 Log messages are propagated to plugins written in Perl as well.
281 See collectd-perl(5).
284 Logs to the standard UNIX logging mechanism, syslog.
286 * Notifications can be handled by the following plugins:
289 Send a desktop notification to a notification daemon, as defined in
290 the Desktop Notification Specification. To actually display the
291 notifications, notification-daemon is required.
292 See http://www.galago-project.org/specs/notification/.
295 Send an E-mail with the notification message to the configured
299 Execute a program or script to handle the notification.
300 See collectd-exec(5).
303 Writes the notification message to a file or STDOUT/STDERR.
306 Send the notification to a remote host to handle it somehow.
309 Notifications are propagated to plugins written in Perl as well.
310 See collectd-perl(5).
312 * Value processing can be controlled using the "filter chain" infrastructure
313 and "matches" and "targets". The following plugins are available:
316 Match values by their identifier based on regular expressions.
319 Match values with an invalid timestamp.
322 Select values by their data sources' values.
324 - target_notification
325 Create and dispatch a notification.
328 Replace parts of an identifier using regular expressions.
331 Set (overwrite) entire parts of an identifier.
333 * Miscellaneous plugins:
336 Sets the hostname to an unique identifier. This is meant for setups
337 where each client may migrate to another physical host, possibly going
338 through one or more name changes in the process.
340 * Performance: Since collectd is running as a daemon it doesn't spend much
341 time starting up again and again. With the exception of the exec plugin no
342 processes are forked. Caching in output plugins, such as the rrdtool and
343 network plugins, makes sure your resources are used efficiently. Also,
344 since collectd is programmed multithreaded it benefits from hyperthreading
345 and multicore processors and makes sure that the daemon isn't idle if only
346 one plugins waits for an IO-operation to complete.
348 * Once set up, hardly any maintenance is necessary. Setup is kept as easy
349 as possible and the default values should be okay for most users.
355 * collectd's configuration file can be found at `sysconfdir'/collectd.conf.
356 Run `collectd -h' for a list of builtin defaults. See `collectd.conf(5)'
357 for a list of options and a syntax description.
359 * When the `csv' or `rrdtool' plugins are loaded they'll write the values to
360 files. The usual place for these files is beneath `/var/lib/collectd'.
362 * When using some of the plugins, collectd needs to run as user root, since
363 only root can do certain things, such as craft ICMP packages needed to ping
364 other hosts. collectd should NOT be installed setuid root since it can be
365 used to overwrite valuable files!
367 * Sample scripts to generate graphs reside in `contrib/' in the source
368 package or somewhere near `/usr/share/doc/collectd' in most distributions.
369 Please be aware that those script are meant as a starting point for your
370 own experiments.. Some of them require the `RRDs' Perl module.
371 (`librrds-perl' on Debian) If you have written a more sophisticated
372 solution please share it with us.
374 * The RRAs of the automatically created RRD files depend on the `step'
375 and `heartbeat' settings given. If change these settings you may need to
376 re-create the files, losing all data. Please be aware of that when changing
377 the values and read the rrdtool(1) manpage thoroughly.
380 collectd and chkrootkit
381 -----------------------
383 If you are using the `dns' plugin chkrootkit(1) will report collectd as a
384 packet sniffer ("<iface>: PACKET SNIFFER(/usr/sbin/collectd[<pid>])"). The
385 plugin captures all UDP packets on port 53 to analyze the DNS traffic. In
386 this case, collectd is a legitimate sniffer and the report should be
387 considered to be a false positive. However, you might want to check that
388 this really is collectd and not some other, illegitimate sniffer.
394 To compile collectd from source you will need:
396 * Usual suspects: C compiler, linker, preprocessor, make, ...
398 * A POSIX-threads (pthread) implementation.
399 Since gathering some statistics is slow (network connections, slow devices,
400 etc) the collectd is parallelized. The POSIX threads interface is being
401 used and should be found in various implementations for hopefully all
404 * CoreFoundation.framework and IOKit.framework (optional)
405 For compiling on Darwin in general and the `apple_sensors' plugin in
407 <http://developer.apple.com/corefoundation/>
409 * libclntsh (optional)
410 Used by the `oracle' plugin.
413 If you want to use the `apache', `ascent', `curl' or `nginx' plugin.
414 <http://curl.haxx.se/>
417 Used by the `dbi' plugin to connect to various databases.
418 <http://libdbi.sourceforge.net/>
420 * libesmtp (optional)
421 For the `notify_email' plugin.
422 <http://www.stafford.uklinux.net/libesmtp/>
425 If present, the uuid plugin will check for UUID from HAL.
426 <http://hal.freedesktop.org/>
428 * libiptc (optional, if not found a version shipped with this distribution
429 can be used if the Linux kernel headers are available)
430 For querying iptables counters.
431 <http://netfilter.org/>
433 * libmysqlclient (optional)
434 Unsurprisingly used by the `mysql' plugin.
435 <http://dev.mysql.com/>
437 * libnetlink (optional)
438 Used, obviously, for the `netlink' plugin.
439 <http://www.linuxfoundation.org/en/Net:Iproute2>
441 * libnetsnmp (optional)
442 For the `snmp' plugin.
443 <http://www.net-snmp.org/>
445 * libnotify (optional)
446 For the `notify_desktop' plugin.
447 <http://www.galago-project.org/>
449 * liboping (optional, if not found a version shipped with this distribution
451 Used by the `ping' plugin to send and receive ICMP packets.
452 <http://verplant.org/liboping/>
454 * libowcapi (optional)
455 Used by the `onewire' plugin to read values from onewire sensors (or the
457 <http://www.owfs.org/>
460 Used to capture packets by the `dns' plugin.
461 <http://www.tcpdump.org/>
464 Obviously used by the `perl' plugin. The library has to be compiled with
465 ithread support (introduced in Perl 5.6.0).
466 <http://www.perl.org/>
469 The PostgreSQL C client library used by the `postgresql' plugin.
470 <http://www.postgresql.org/>
473 Used by the `rrdtool' and `rrdcached' plugins. The latter requires RRDtool
474 client support which was added after version 1.3 of RRDtool. Versions 1.0,
475 1.2 and 1.3 are known to work with the `rrdtool' plugin.
476 <http://oss.oetiker.ch/rrdtool/>
478 * librt, libsocket, libkstat, libdevinfo (optional)
479 Various standard Solaris libraries which provide system functions.
480 <http://developers.sun.com/solaris/>
482 * libsensors (optional)
483 To read from `lm_sensors', see the `sensors' plugin.
484 <http://www.lm-sensors.org/>
486 * libstatgrab (optional)
487 Used by various plugins to collect statistics on systems other than Linux
489 <http://www.i-scream.org/libstatgrab/>
491 * libupsclient/nut (optional)
492 For the `nut' plugin which queries nut's `upsd'.
493 <http://networkupstools.org/>
496 Collect statistics from virtual machines.
497 <http://libvirt.org/>
500 Parse XML data. This is needed for the `ascent' and `libvirt' plugins.
501 <http://xmlsoft.org/>
504 <http://www.xmms.org/>
507 Configuring / Compiling / Installing
508 ------------------------------------
510 To configure, build and install collectd with the default settings, run
511 `./configure && make && make install'. For detailed, generic instructions
512 see INSTALL. For a complete list of configure options and their description,
513 run `./configure --help'.
515 By default, the configure script will check for all build dependencies and
516 disable all plugins whose requirements cannot be fulfilled (any other plugin
517 will be enabled). To enable a plugin, install missing dependencies (see
518 section `Prerequisites' above) and rerun `configure'. If you specify the
519 `--enable-<plugin>' configure option, the script will fail if the depen-
520 dencies for the specified plugin are not met. If you specify the
521 `--disable-<plugin>' configure option, the plugin will not be built. Both
522 options are meant for package maintainers and should not be used in everyday
525 By default, collectd will be installed into `/opt/collectd'. You can adjust
526 this setting by specifying the `--prefix' configure option - see INSTALL for
527 details. If you pass DESTDIR=<path> to `make install', <path> will be
528 prefixed to all installation directories. This might be useful when creating
529 packages for collectd.
535 To compile correctly collectd needs to be able to initialize static
536 variables to NAN (Not A Number). Some C libraries, especially the GNU
537 libc, have a problem with that.
539 Luckily, with GCC it's possible to work around that problem: One can define
540 NAN as being (0.0 / 0.0) and `isnan' as `f != f'. However, to test this
541 ``implementation'' the configure script needs to compile and run a short
542 test program. Obviously running a test program when doing a cross-
543 compilation is, well, challenging.
545 If you run into this problem, you can use the `--with-nan-emulation'
546 configure option to force the use of this implementation. We can't promise
547 that the compiled binary actually behaves as it should, but since NANs
548 are likely never passed to the libm you have a good chance to be lucky.
550 Likewise, collectd needs to know the layout of doubles in memory, in order
551 to craft uniform network packets over different architectures. For this, it
552 needs to know how to convert doubles into the memory layout used by x86. The
553 configure script tries to figure this out by compiling and running a few
554 small test programs. This is of course not possible when cross-compiling.
555 You can use the `--with-fp-layout' option to tell the configure script which
556 conversion method to assume. Valid arguments are:
558 * `nothing' (12345678 -> 12345678)
559 * `endianflip' (12345678 -> 87654321)
560 * `intswap' (12345678 -> 56781234)
566 For questions, bug reports, development information and basically all other
567 concerns please send an email to collectd's mailing list at
568 <collectd at verplant.org>.
570 For live discussion and more personal contact visit us in IRC, we're in
571 channel #collectd on freenode.
577 Florian octo Forster <octo at verplant.org>,
578 Sebastian tokkee Harl <sh at tokkee.org>,
579 and many contributors (see `AUTHORS').
581 Please send bug reports and patches to the mailing list, see `Contact'