1SYSTEMCTL(1) systemctl SYSTEMCTL(1)
2
6 systemctl - Control the systemd system and service manager
7
9 systemctl [OPTIONS...] COMMAND [UNIT...]
10
12 systemctl may be used to introspect and control the state of the
13 "systemd" system and service manager. Please refer to systemd(1) for an
14 introduction into the basic concepts and functionality this tool
15 manages.
16
18 The following commands are understood:
19 Unit Commands (Introspection and Modification)
21 list-units [PATTERN...]
22 List units that systemd currently has in memory. This includes
23 units that are either referenced directly or through a dependency,
24 units that are pinned by applications programmatically, or units
25 that were active in the past and have failed. By default only units
26 which are active, have pending jobs, or have failed are shown; this
27 can be changed with option --all. If one or more PATTERNs are
28 specified, only units matching one of them are shown. The units
29 that are shown are additionally filtered by --type= and --state= if
30 those options are specified.
31 Produces output similar to
33 UNIT LOAD ACTIVE SUB DESCRIPTION
35 sys-module-fuse.device loaded active plugged /sys/module/fuse
36 -.mount loaded active mounted Root Mount
37 boot-efi.mount loaded active mounted /boot/efi
38 systemd-journald.service loaded active running Journal Service
39 systemd-logind.service loaded active running Login Service
40 ● user@1000.service loaded failed failed User Manager for UID 1000
41 ...
42 systemd-tmpfiles-clean.timer loaded active waiting Daily Cleanup of Temporary Directories
43 LOAD = Reflects whether the unit definition was properly loaded.
45 ACTIVE = The high-level unit activation state, i.e. generalization of SUB.
46 SUB = The low-level unit activation state, values depend on unit type.
47 123 loaded units listed. Pass --all to see loaded but inactive units, too.
49 To show all installed unit files use 'systemctl list-unit-files'.
50 The header and the last unit of a given type are underlined if the
53 terminal supports that. A colored dot is shown next to services
54 which were masked, not found, or otherwise failed.
55 The LOAD column shows the load state, one of loaded, not-found,
57 bad-setting, error, masked. The ACTIVE columns shows the general
58 unit state, one of active, reloading, inactive, failed, activating,
59 deactivating. The SUB column shows the unit-type-specific detailed
60 state of the unit, possible values vary by unit type. The list of
61 possible LOAD, ACTIVE, and SUB states is not constant and new
62 systemd releases may both add and remove values.
63 systemctl --state=help
65 command maybe be used to display the current set of possible
67 values.
68 This is the default command.
70 list-sockets [PATTERN...]
72 List socket units currently in memory, ordered by listening
73 address. If one or more PATTERNs are specified, only socket units
74 matching one of them are shown. Produces output similar to
75 LISTEN UNIT ACTIVATES
77 /dev/initctl systemd-initctl.socket systemd-initctl.service
78 ...
79 [::]:22 sshd.socket sshd.service
80 kobject-uevent 1 systemd-udevd-kernel.socket systemd-udevd.service
81 5 sockets listed.
83 Note: because the addresses might contains spaces, this output is
85 not suitable for programmatic consumption.
86 Also see --show-types, --all, and --state=.
88 list-timers [PATTERN...]
90 List timer units currently in memory, ordered by the time they
91 elapse next. If one or more PATTERNs are specified, only units
92 matching one of them are shown. Produces output similar to
93 NEXT LEFT LAST PASSED UNIT ACTIVATES
95 n/a n/a Thu 2017-02-23 13:40:29 EST 3 days ago ureadahead-stop.timer ureadahead-stop.service
96 Sun 2017-02-26 18:55:42 EST 1min 14s left Thu 2017-02-23 13:54:44 EST 3 days ago systemd-tmpfiles-clean.timer systemd-tmpfiles-clean.service
97 Sun 2017-02-26 20:37:16 EST 1h 42min left Sun 2017-02-26 11:56:36 EST 6h ago apt-daily.timer apt-daily.service
98 Sun 2017-02-26 20:57:49 EST 2h 3min left Sun 2017-02-26 11:56:36 EST 6h ago snapd.refresh.timer snapd.refresh.service
99 NEXT shows the next time the timer will run.
102 LEFT shows how long till the next time the timer runs.
104 LAST shows the last time the timer ran.
106 PASSED shows how long has passed since the timer last ran.
108 UNIT shows the name of the timer
110 ACTIVATES shows the name the service the timer activates when it
112 runs.
113 Also see --all and --state=.
115 is-active PATTERN...
117 Check whether any of the specified units are active (i.e. running).
118 Returns an exit code 0 if at least one is active, or non-zero
119 otherwise. Unless --quiet is specified, this will also print the
120 current unit state to standard output.
121 is-failed PATTERN...
123 Check whether any of the specified units are in a "failed" state.
124 Returns an exit code 0 if at least one has failed, non-zero
125 otherwise. Unless --quiet is specified, this will also print the
126 current unit state to standard output.
127 status [PATTERN...|PID...]]
129 Show terse runtime status information about one or more units,
130 followed by most recent log data from the journal. If no units are
131 specified, show system status. If combined with --all, also show
132 the status of all units (subject to limitations specified with -t).
133 If a PID is passed, show information about the unit the process
134 belongs to.
135 This function is intended to generate human-readable output. If you
137 are looking for computer-parsable output, use show instead. By
138 default, this function only shows 10 lines of output and ellipsizes
139 lines to fit in the terminal window. This can be changed with
140 --lines and --full, see above. In addition, journalctl --unit=NAME
141 or journalctl --user-unit=NAME use a similar filter for messages
142 and might be more convenient.
143 systemd implicitly loads units as necessary, so just running the
145 status will attempt to load a file. The command is thus not useful
146 for determining if something was already loaded or not. The units
147 may possibly also be quickly unloaded after the operation is
148 completed if there's no reason to keep it in memory thereafter.
149 Example 1. Example output from systemctl status
151 $ systemctl status bluetooth
153 ● bluetooth.service - Bluetooth service
154 Loaded: loaded (/usr/lib/systemd/system/bluetooth.service; enabled; vendor preset: enabled)
155 Active: active (running) since Wed 2017-01-04 13:54:04 EST; 1 weeks 0 days ago
156 Docs: man:bluetoothd(8)
157 Main PID: 930 (bluetoothd)
158 Status: "Running"
159 Tasks: 1
160 Memory: 648.0K
161 CPU: 435ms
162 CGroup: /system.slice/bluetooth.service
163 └─930 /usr/lib/bluetooth/bluetoothd
164 Jan 12 10:46:45 example.com bluetoothd[8900]: Not enough free handles to register service
166 Jan 12 10:46:45 example.com bluetoothd[8900]: Current Time Service could not be registered
167 Jan 12 10:46:45 example.com bluetoothd[8900]: gatt-time-server: Input/output error (5)
168 The dot ("●") uses color on supported terminals to summarize the
170 unit state at a glance. Along with its color, its shape varies
171 according to its state: "inactive" or "maintenance" is a white
172 circle ("○"), "active" is a green dot ("●"), "deactivating" is a
173 white dot, "failed" or "error" is a red cross ("×"), and
174 "reloading" is a green clockwise circle arrow ("↻").
175 The "Loaded:" line in the output will show "loaded" if the unit has
177 been loaded into memory. Other possible values for "Loaded:"
178 include: "error" if there was a problem loading it, "not-found" if
179 no unit file was found for this unit, "bad-setting" if an essential
180 unit file setting could not be parsed and "masked" if the unit file
181 has been masked. Along with showing the path to the unit file, this
182 line will also show the enablement state. Enabled commands start at
183 boot. See the full table of possible enablement states — including
184 the definition of "masked" — in the documentation for the
185 is-enabled command.
186 The "Active:" line shows active state. The value is usually
188 "active" or "inactive". Active could mean started, bound, plugged
189 in, etc depending on the unit type. The unit could also be in
190 process of changing states, reporting a state of "activating" or
191 "deactivating". A special "failed" state is entered when the
192 service failed in some way, such as a crash, exiting with an error
193 code or timing out. If the failed state is entered the cause will
194 be logged for later reference.
195 show [PATTERN...|JOB...]
197 Show properties of one or more units, jobs, or the manager itself.
198 If no argument is specified, properties of the manager will be
199 shown. If a unit name is specified, properties of the unit are
200 shown, and if a job ID is specified, properties of the job are
201 shown. By default, empty properties are suppressed. Use --all to
202 show those too. To select specific properties to show, use
203 --property=. This command is intended to be used whenever
204 computer-parsable output is required. Use status if you are looking
205 for formatted human-readable output.
206 Many properties shown by systemctl show map directly to
208 configuration settings of the system and service manager and its
209 unit files. Note that the properties shown by the command are
210 generally more low-level, normalized versions of the original
211 configuration settings and expose runtime state in addition to
212 configuration. For example, properties shown for service units
213 include the service's current main process identifier as "MainPID"
214 (which is runtime state), and time settings are always exposed as
215 properties ending in the "...USec" suffix even if a matching
216 configuration options end in "...Sec", because microseconds is the
217 normalized time unit used internally by the system and service
218 manager.
219 For details about many of these properties, see the documentation
221 of the D-Bus interface backing these properties, see
222 org.freedesktop.systemd1(5).
223 cat PATTERN...
225 Show backing files of one or more units. Prints the "fragment" and
226 "drop-ins" (source files) of units. Each file is preceded by a
227 comment which includes the file name. Note that this shows the
228 contents of the backing files on disk, which may not match the
229 system manager's understanding of these units if any unit files
230 were updated on disk and the daemon-reload command wasn't issued
231 since.
232 help PATTERN...|PID...
234 Show manual pages for one or more units, if available. If a PID is
235 given, the manual pages for the unit the process belongs to are
236 shown.
237 list-dependencies [UNIT...]
239 Shows units required and wanted by the specified units. This
240 recursively lists units following the Requires=, Requisite=,
241 ConsistsOf=, Wants=, BindsTo= dependencies. If no units are
242 specified, default.target is implied.
243 By default, only target units are recursively expanded. When --all
245 is passed, all other units are recursively expanded as well.
246 Options --reverse, --after, --before may be used to change what
248 types of dependencies are shown.
249 Note that this command only lists units currently loaded into
251 memory by the service manager. In particular, this command is not
252 suitable to get a comprehensive list at all reverse dependencies on
253 a specific unit, as it won't list the dependencies declared by
254 units currently not loaded.
255 start PATTERN...
257 Start (activate) one or more units specified on the command line.
258 Note that unit glob patterns expand to names of units currently in
260 memory. Units which are not active and are not in a failed state
261 usually are not in memory, and will not be matched by any pattern.
262 In addition, in case of instantiated units, systemd is often
263 unaware of the instance name until the instance has been started.
264 Therefore, using glob patterns with start has limited usefulness.
265 Also, secondary alias names of units are not considered.
266 Option --all may be used to also operate on inactive units which
268 are referenced by other loaded units. Note that this is not the
269 same as operating on "all" possible units, because as the previous
270 paragraph describes, such a list is ill-defined. Nevertheless,
271 systemctl start --all GLOB may be useful if all the units that
272 should match the pattern are pulled in by some target which is
273 known to be loaded.
274 stop PATTERN...
276 Stop (deactivate) one or more units specified on the command line.
277 This command will fail if the unit does not exist or if stopping of
279 the unit is prohibited (see RefuseManualStop= in systemd.unit(5)).
280 It will not fail if any of the commands configured to stop the unit
281 (ExecStop=, etc.) fail, because the manager will still forcibly
282 terminate the unit.
283 reload PATTERN...
285 Asks all units listed on the command line to reload their
286 configuration. Note that this will reload the service-specific
287 configuration, not the unit configuration file of systemd. If you
288 want systemd to reload the configuration file of a unit, use the
289 daemon-reload command. In other words: for the example case of
290 Apache, this will reload Apache's httpd.conf in the web server, not
291 the apache.service systemd unit file.
292 This command should not be confused with the daemon-reload command.
294 restart PATTERN...
296 Stop and then start one or more units specified on the command
297 line. If the units are not running yet, they will be started.
298 Note that restarting a unit with this command does not necessarily
300 flush out all of the unit's resources before it is started again.
301 For example, the per-service file descriptor storage facility (see
302 FileDescriptorStoreMax= in systemd.service(5)) will remain intact
303 as long as the unit has a job pending, and is only cleared when the
304 unit is fully stopped and no jobs are pending anymore. If it is
305 intended that the file descriptor store is flushed out, too, during
306 a restart operation an explicit systemctl stop command followed by
307 systemctl start should be issued.
308 try-restart PATTERN...
310 Stop and then start one or more units specified on the command line
311 if the units are running. This does nothing if units are not
312 running.
313 reload-or-restart PATTERN...
315 Reload one or more units if they support it. If not, stop and then
316 start them instead. If the units are not running yet, they will be
317 started.
318 try-reload-or-restart PATTERN...
320 Reload one or more units if they support it. If not, stop and then
321 start them instead. This does nothing if the units are not running.
322 isolate UNIT
324 Start the unit specified on the command line and its dependencies
325 and stop all others, unless they have IgnoreOnIsolate=yes (see
326 systemd.unit(5)). If a unit name with no extension is given, an
327 extension of ".target" will be assumed.
328 This command is dangerous, since it will immediately stop processes
330 that are not enabled in the new target, possibly including the
331 graphical environment or terminal you are currently using.
332 Note that this is allowed only on units where AllowIsolate= is
334 enabled. See systemd.unit(5) for details.
335 kill PATTERN...
337 Send a signal to one or more processes of the unit. Use --kill-who=
338 to select which process to kill. Use --signal= to select the signal
339 to send.
340 clean PATTERN...
342 Remove the configuration, state, cache, logs or runtime data of the
343 specified units. Use --what= to select which kind of resource to
344 remove. For service units this may be used to remove the
345 directories configured with ConfigurationDirectory=,
346 StateDirectory=, CacheDirectory=, LogsDirectory= and
347 RuntimeDirectory=, see systemd.exec(5) for details. For timer units
348 this may be used to clear out the persistent timestamp data if
349 Persistent= is used and --what=state is selected, see
350 systemd.timer(5). This command only applies to units that use
351 either of these settings. If --what= is not specified, both the
352 cache and runtime data are removed (as these two types of data are
353 generally redundant and reproducible on the next invocation of the
354 unit).
355 freeze PATTERN...
357 Freeze one or more units specified on the command line using cgroup
358 freezer
359 Freezing the unit will cause all processes contained within the
361 cgroup corresponding to the unit to be suspended. Being suspended
362 means that unit's processes won't be scheduled to run on CPU until
363 thawed. Note that this command is supported only on systems that
364 use unified cgroup hierarchy. Unit is automatically thawed just
365 before we execute a job against the unit, e.g. before the unit is
366 stopped.
367 thaw PATTERN...
369 Thaw (unfreeze) one or more units specified on the command line.
370 This is the inverse operation to the freeze command and resumes the
372 execution of processes in the unit's cgroup.
373 set-property UNIT PROPERTY=VALUE...
375 Set the specified unit properties at runtime where this is
376 supported. This allows changing configuration parameter properties
377 such as resource control settings at runtime. Not all properties
378 may be changed at runtime, but many resource control settings
379 (primarily those in systemd.resource-control(5)) may. The changes
380 are applied immediately, and stored on disk for future boots,
381 unless --runtime is passed, in which case the settings only apply
382 until the next reboot. The syntax of the property assignment
383 follows closely the syntax of assignments in unit files.
384 Example: systemctl set-property foobar.service CPUWeight=200
386 If the specified unit appears to be inactive, the changes will be
388 only stored on disk as described previously hence they will be
389 effective when the unit will be started.
390 Note that this command allows changing multiple properties at the
392 same time, which is preferable over setting them individually.
393 Example: systemctl set-property foobar.service CPUWeight=200
395 MemoryMax=2G IPAccounting=yes
396 Like with unit file configuration settings, assigning an empty
398 setting usually resets a property to its defaults.
399 Example: systemctl set-property avahi-daemon.service IPAddressDeny=
401 bind UNIT PATH [PATH]
403 Bind mounts a file or directory from the host into the specified
404 unit's view. The first path argument is the source file or
405 directory on the host, the second path argument is the destination
406 file or directory in the unit's view. When the latter is omitted,
407 the destination path in the unit's view is the same as the source
408 path on the host. When combined with the --read-only switch, a
409 ready-only bind mount is created. When combined with the --mkdir
410 switch, the destination path is first created before the mount is
411 applied. Note that this option is currently only supported for
412 units that run within a mount namespace (e.g.: with RootImage=,
413 PrivateMounts=, etc.). This command supports bind mounting
414 directories, regular files, device nodes, AF_UNIX socket nodes, as
415 well as FIFOs. The bind mount is ephemeral, and it is undone as
416 soon as the current unit process exists. Note that the namespace
417 mentioned here, where the bind mount will be added to, is the one
418 where the main service process runs, as other processes run in
419 distinct namespaces (e.g.: ExecReload=, ExecStartPre=, etc.)
420 mount-image UNIT IMAGE [PATH [PARTITION_NAME:MOUNT_OPTIONS]]
422 Mounts an image from the host into the specified unit's view. The
423 first path argument is the source image on the host, the second
424 path argument is the destination directory in the unit's view (ie:
425 inside RootImage=/RootDirectory=). Any following argument is
426 interpreted as a colon-separated tuple of partition name and
427 comma-separated list of mount options for that partition. The
428 format is the same as the service MountImages= setting. When
429 combined with the --read-only switch, a ready-only mount is
430 created. When combined with the --mkdir switch, the destination
431 path is first created before the mount is applied. Note that this
432 option is currently only supported for units that run within a
433 mount namespace (e.g.: with RootImage=, PrivateMounts=, etc.). Note
434 that the namespace mentioned here, where the image mount will be
435 added to, is the one where the main service process runs, as other
436 processes run in distinct namespaces (e.g.: ExecReload=,
437 ExecStartPre=, etc.). Example:
438 systemctl mount-image foo.service /tmp/img.raw /var/lib/image root:ro,nosuid
440 systemctl mount-image --mkdir bar.service /tmp/img.raw /var/lib/baz/img
444 service-log-level SERVICE [LEVEL]
446 If the LEVEL argument is not given, print the current log level as
447 reported by service SERVICE.
448 If the optional argument LEVEL is provided, then change the current
450 log level of the service to LEVEL. The log level should be a
451 typical syslog log level, i.e. a value in the range 0...7 or one of
452 the strings emerg, alert, crit, err, warning, notice, info, debug;
453 see syslog(3) for details.
454 The service must have the appropriate BusName=destination property
456 and also implement the generic org.freedesktop.LogControl1(5)
457 interface. (systemctl will use the generic D-Bus protocol to access
458 the org.freedesktop.LogControl1.LogLevel interface for the D-Bus
459 name destination.)
460 service-log-target SERVICE [TARGET]
462 If the TARGET argument is not given, print the current log target
463 as reported by service SERVICE.
464 If the optional argument TARGET is provided, then change the
466 current log target of the service to TARGET. The log target should
467 be one of the strings console (for log output to the service's
468 standard error stream), kmsg (for log output to the kernel log
469 buffer), journal (for log output to systemd-journald.service(8)
470 using the native journal protocol), syslog (for log output to the
471 classic syslog socket /dev/log), null (for no log output
472 whatsoever) or auto (for an automatically determined choice,
473 typically equivalent to console if the service is invoked
474 interactively, and journal or syslog otherwise).
475 For most services, only a small subset of log targets make sense.
477 In particular, most "normal" services should only implement
478 console, journal, and null. Anything else is only appropriate for
479 low-level services that are active in very early boot before proper
480 logging is established.
481 The service must have the appropriate BusName=destination property
483 and also implement the generic org.freedesktop.LogControl1(5)
484 interface. (systemctl will use the generic D-Bus protocol to access
485 the org.freedesktop.LogControl1.LogLevel interface for the D-Bus
486 name destination.)
487 reset-failed [PATTERN...]
489 Reset the "failed" state of the specified units, or if no unit name
490 is passed, reset the state of all units. When a unit fails in some
491 way (i.e. process exiting with non-zero error code, terminating
492 abnormally or timing out), it will automatically enter the "failed"
493 state and its exit code and status is recorded for introspection by
494 the administrator until the service is stopped/re-started or reset
495 with this command.
496 In addition to resetting the "failed" state of a unit it also
498 resets various other per-unit properties: the start rate limit
499 counter of all unit types is reset to zero, as is the restart
500 counter of service units. Thus, if a unit's start limit (as
501 configured with StartLimitIntervalSec=/StartLimitBurst=) is hit and
502 the unit refuses to be started again, use this command to make it
503 startable again.
504 Unit File Commands
506 list-unit-files [PATTERN...]
507 List unit files installed on the system, in combination with their
508 enablement state (as reported by is-enabled). If one or more
509 PATTERNs are specified, only unit files whose name matches one of
510 them are shown (patterns matching unit file system paths are not
511 supported).
512 enable UNIT..., enable PATH...
514 Enable one or more units or unit instances. This will create a set
515 of symlinks, as encoded in the [Install] sections of the indicated
516 unit files. After the symlinks have been created, the system
517 manager configuration is reloaded (in a way equivalent to
518 daemon-reload), in order to ensure the changes are taken into
519 account immediately. Note that this does not have the effect of
520 also starting any of the units being enabled. If this is desired,
521 combine this command with the --now switch, or invoke start with
522 appropriate arguments later. Note that in case of unit instance
523 enablement (i.e. enablement of units of the form foo@bar.service),
524 symlinks named the same as instances are created in the unit
525 configuration directory, however they point to the single template
526 unit file they are instantiated from.
527 This command expects either valid unit names (in which case various
529 unit file directories are automatically searched for unit files
530 with appropriate names), or absolute paths to unit files (in which
531 case these files are read directly). If a specified unit file is
532 located outside of the usual unit file directories, an additional
533 symlink is created, linking it into the unit configuration path,
534 thus ensuring it is found when requested by commands such as start.
535 The file system where the linked unit files are located must be
536 accessible when systemd is started (e.g. anything underneath /home/
537 or /var/ is not allowed, unless those directories are located on
538 the root file system).
539 This command will print the file system operations executed. This
541 output may be suppressed by passing --quiet.
542 Note that this operation creates only the symlinks suggested in the
544 [Install] section of the unit files. While this command is the
545 recommended way to manipulate the unit configuration directory, the
546 administrator is free to make additional changes manually by
547 placing or removing symlinks below this directory. This is
548 particularly useful to create configurations that deviate from the
549 suggested default installation. In this case, the administrator
550 must make sure to invoke daemon-reload manually as necessary, in
551 order to ensure the changes are taken into account.
552 Enabling units should not be confused with starting (activating)
554 units, as done by the start command. Enabling and starting units is
555 orthogonal: units may be enabled without being started and started
556 without being enabled. Enabling simply hooks the unit into various
557 suggested places (for example, so that the unit is automatically
558 started on boot or when a particular kind of hardware is plugged
559 in). Starting actually spawns the daemon process (in case of
560 service units), or binds the socket (in case of socket units), and
561 so on.
562 Depending on whether --system, --user, --runtime, or --global is
564 specified, this enables the unit for the system, for the calling
565 user only, for only this boot of the system, or for all future
566 logins of all users. Note that in the last case, no systemd daemon
567 configuration is reloaded.
568 Using enable on masked units is not supported and results in an
570 error.
571 disable UNIT...
573 Disables one or more units. This removes all symlinks to the unit
574 files backing the specified units from the unit configuration
575 directory, and hence undoes any changes made by enable or link.
576 Note that this removes all symlinks to matching unit files,
577 including manually created symlinks, and not just those actually
578 created by enable or link. Note that while disable undoes the
579 effect of enable, the two commands are otherwise not symmetric, as
580 disable may remove more symlinks than a prior enable invocation of
581 the same unit created.
582 This command expects valid unit names only, it does not accept
584 paths to unit files.
585 In addition to the units specified as arguments, all units are
587 disabled that are listed in the Also= setting contained in the
588 [Install] section of any of the unit files being operated on.
589 This command implicitly reloads the system manager configuration
591 after completing the operation. Note that this command does not
592 implicitly stop the units that are being disabled. If this is
593 desired, either combine this command with the --now switch, or
594 invoke the stop command with appropriate arguments later.
595 This command will print information about the file system
597 operations (symlink removals) executed. This output may be
598 suppressed by passing --quiet.
599 This command honors --system, --user, --runtime and --global in a
601 similar way as enable.
602 reenable UNIT...
604 Reenable one or more units, as specified on the command line. This
605 is a combination of disable and enable and is useful to reset the
606 symlinks a unit file is enabled with to the defaults configured in
607 its [Install] section. This command expects a unit name only, it
608 does not accept paths to unit files.
609 preset UNIT...
611 Reset the enable/disable status one or more unit files, as
612 specified on the command line, to the defaults configured in the
613 preset policy files. This has the same effect as disable or enable,
614 depending how the unit is listed in the preset files.
615 Use --preset-mode= to control whether units shall be enabled and
617 disabled, or only enabled, or only disabled.
618 If the unit carries no install information, it will be silently
620 ignored by this command. UNIT must be the real unit name, any
621 alias names are ignored silently.
622 For more information on the preset policy format, see
624 systemd.preset(5).
625 preset-all
627 Resets all installed unit files to the defaults configured in the
628 preset policy file (see above).
629 Use --preset-mode= to control whether units shall be enabled and
631 disabled, or only enabled, or only disabled.
632 is-enabled UNIT...
634 Checks whether any of the specified unit files are enabled (as with
635 enable). Returns an exit code of 0 if at least one is enabled,
636 non-zero otherwise. Prints the current enable status (see table).
637 To suppress this output, use --quiet. To show installation targets,
638 use --full.
639 Table 1. is-enabled output
641 ┌──────────────────┬─────────────────────────┬───────────┐
642 │Name │ Description │ Exit Code │
643 ├──────────────────┼─────────────────────────┼───────────┤
644 │"enabled" │ Enabled via │ │
645 ├──────────────────┤ .wants/, .requires/ │ │
646 │"enabled-runtime" │ or Alias= symlinks │ │
647 │ │ (permanently in │ 0 │
648 │ │ /etc/systemd/system/, │ │
649 │ │ or transiently in │ │
650 │ │ /run/systemd/system/). │ │
651 ├──────────────────┼─────────────────────────┼───────────┤
652 │"linked" │ Made available through │ │
653 ├──────────────────┤ one or more symlinks │ │
654 │"linked-runtime" │ to the unit file │ │
655 │ │ (permanently in │ │
656 │ │ /etc/systemd/system/ │ │
657 │ │ or transiently in │ > 0 │
658 │ │ /run/systemd/system/), │ │
659 │ │ even though the unit │ │
660 │ │ file might reside │ │
661 │ │ outside of the unit │ │
662 │ │ file search path. │ │
663 ├──────────────────┼─────────────────────────┼───────────┤
664 │"alias" │ The name is an alias │ 0 │
665 │ │ (symlink to another │ │
666 │ │ unit file). │ │
667 ├──────────────────┼─────────────────────────┼───────────┤
668 │"masked" │ Completely disabled, │ │
669 ├──────────────────┤ so that any start │ │
670 │"masked-runtime" │ operation on it fails │ │
671 │ │ (permanently in │ > 0 │
672 │ │ /etc/systemd/system/ │ │
673 │ │ or transiently in │ │
674 │ │ /run/systemd/systemd/). │ │
675 ├──────────────────┼─────────────────────────┼───────────┤
676 │"static" │ The unit file is not │ 0 │
677 │ │ enabled, and has no │ │
678 │ │ provisions for enabling │ │
679 │ │ in the [Install] unit │ │
680 │ │ file section. │ │
681 ├──────────────────┼─────────────────────────┼───────────┤
682 │"indirect" │ The unit file itself is │ 0 │
683 │ │ not enabled, but it has │ │
684 │ │ a non-empty Also= │ │
685 │ │ setting in the │ │
686 │ │ [Install] unit file │ │
687 │ │ section, listing other │ │
688 │ │ unit files that might │ │
689 │ │ be enabled, or it has │ │
690 │ │ an alias under a │ │
691 │ │ different name through │ │
692 │ │ a symlink that is not │ │
693 │ │ specified in Also=. For │ │
694 │ │ template unit files, an │ │
695 │ │ instance different than │ │
696 │ │ the one specified in │ │
697 │ │ DefaultInstance= is │ │
698 │ │ enabled. │ │
699 ├──────────────────┼─────────────────────────┼───────────┤
700 │"disabled" │ The unit file is not │ > 0 │
701 │ │ enabled, but contains │ │
702 │ │ an [Install] section │ │
703 │ │ with installation │ │
704 │ │ instructions. │ │
705 ├──────────────────┼─────────────────────────┼───────────┤
706 │"generated" │ The unit file was │ 0 │
707 │ │ generated dynamically │ │
708 │ │ via a generator tool. │ │
709 │ │ See │ │
710 │ │ systemd.generator(7). │ │
711 │ │ Generated unit files │ │
712 │ │ may not be enabled, │ │
713 │ │ they are enabled │ │
714 │ │ implicitly by their │ │
715 │ │ generator. │ │
716 ├──────────────────┼─────────────────────────┼───────────┤
717 │"transient" │ The unit file has been │ 0 │
718 │ │ created dynamically │ │
719 │ │ with the runtime API. │ │
720 │ │ Transient units may not │ │
721 │ │ be enabled. │ │
722 ├──────────────────┼─────────────────────────┼───────────┤
723 │"bad" │ The unit file is │ > 0 │
724 │ │ invalid or another │ │
725 │ │ error occurred. Note │ │
726 │ │ that is-enabled will │ │
727 │ │ not actually return │ │
728 │ │ this state, but print │ │
729 │ │ an error message │ │
730 │ │ instead. However the │ │
731 │ │ unit file listing │ │
732 │ │ printed by │ │
733 │ │ list-unit-files might │ │
734 │ │ show it. │ │
735 └──────────────────┴─────────────────────────┴───────────┘
736 mask UNIT...
738 Mask one or more units, as specified on the command line. This will
739 link these unit files to /dev/null, making it impossible to start
740 them. This is a stronger version of disable, since it prohibits all
741 kinds of activation of the unit, including enablement and manual
742 activation. Use this option with care. This honors the --runtime
743 option to only mask temporarily until the next reboot of the
744 system. The --now option may be used to ensure that the units are
745 also stopped. This command expects valid unit names only, it does
746 not accept unit file paths.
747 unmask UNIT...
749 Unmask one or more unit files, as specified on the command line.
750 This will undo the effect of mask. This command expects valid unit
751 names only, it does not accept unit file paths.
752 link PATH...
754 Link a unit file that is not in the unit file search paths into the
755 unit file search path. This command expects an absolute path to a
756 unit file. The effect of this may be undone with disable. The
757 effect of this command is that a unit file is made available for
758 commands such as start, even though it is not installed directly in
759 the unit search path. The file system where the linked unit files
760 are located must be accessible when systemd is started (e.g.
761 anything underneath /home/ or /var/ is not allowed, unless those
762 directories are located on the root file system).
763 revert UNIT...
765 Revert one or more unit files to their vendor versions. This
766 command removes drop-in configuration files that modify the
767 specified units, as well as any user-configured unit file that
768 overrides a matching vendor supplied unit file. Specifically, for a
769 unit "foo.service" the matching directories "foo.service.d/" with
770 all their contained files are removed, both below the persistent
771 and runtime configuration directories (i.e. below
772 /etc/systemd/system and /run/systemd/system); if the unit file has
773 a vendor-supplied version (i.e. a unit file located below /usr/)
774 any matching persistent or runtime unit file that overrides it is
775 removed, too. Note that if a unit file has no vendor-supplied
776 version (i.e. is only defined below /etc/systemd/system or
777 /run/systemd/system, but not in a unit file stored below /usr/),
778 then it is not removed. Also, if a unit is masked, it is unmasked.
779 Effectively, this command may be used to undo all changes made with
781 systemctl edit, systemctl set-property and systemctl mask and puts
782 the original unit file with its settings back in effect.
783 add-wants TARGET UNIT..., add-requires TARGET UNIT...
785 Adds "Wants=" or "Requires=" dependencies, respectively, to the
786 specified TARGET for one or more units.
787 This command honors --system, --user, --runtime and --global in a
789 way similar to enable.
790 edit UNIT...
792 Edit a drop-in snippet or a whole replacement file if --full is
793 specified, to extend or override the specified unit.
794 Depending on whether --system (the default), --user, or --global is
796 specified, this command creates a drop-in file for each unit either
797 for the system, for the calling user, or for all futures logins of
798 all users. Then, the editor (see the "Environment" section below)
799 is invoked on temporary files which will be written to the real
800 location if the editor exits successfully.
801 If --full is specified, this will copy the original units instead
803 of creating drop-in files.
804 If --force is specified and any units do not already exist, new
806 unit files will be opened for editing.
807 If --runtime is specified, the changes will be made temporarily in
809 /run/ and they will be lost on the next reboot.
810 If the temporary file is empty upon exit, the modification of the
812 related unit is canceled.
813 After the units have been edited, systemd configuration is reloaded
815 (in a way that is equivalent to daemon-reload).
816 Note that this command cannot be used to remotely edit units and
818 that you cannot temporarily edit units which are in /etc/, since
819 they take precedence over /run/.
820 get-default
822 Return the default target to boot into. This returns the target
823 unit name default.target is aliased (symlinked) to.
824 set-default TARGET
826 Set the default target to boot into. This sets (symlinks) the
827 default.target alias to the given target unit.
828 Machine Commands
830 list-machines [PATTERN...]
831 List the host and all running local containers with their state. If
832 one or more PATTERNs are specified, only containers matching one of
833 them are shown.
834 Job Commands
836 list-jobs [PATTERN...]
837 List jobs that are in progress. If one or more PATTERNs are
838 specified, only jobs for units matching one of them are shown.
839 When combined with --after or --before the list is augmented with
841 information on which other job each job is waiting for, and which
842 other jobs are waiting for it, see above.
843 cancel JOB...
845 Cancel one or more jobs specified on the command line by their
846 numeric job IDs. If no job ID is specified, cancel all pending
847 jobs.
848 Environment Commands
850 systemd supports an environment block that is passed to processes the
851 manager spawns. The names of the variables can contain ASCII letters,
852 digits, and the underscore character. Variable names cannot be empty or
853 start with a digit. In variable values, most characters are allowed,
854 but the whole sequence must be valid UTF-8. (Note that control
855 characters like newline (NL), tab (TAB), or the escape character (ESC),
856 are valid ASCII and thus valid UTF-8). The total length of the
857 environment block is limited to _SC_ARG_MAX value defined by
858 sysconf(3).
859 show-environment
861 Dump the systemd manager environment block. This is the environment
862 block that is passed to all processes the manager spawns. The
863 environment block will be dumped in straight-forward form suitable
864 for sourcing into most shells. If no special characters or
865 whitespace is present in the variable values, no escaping is
866 performed, and the assignments have the form "VARIABLE=value". If
867 whitespace or characters which have special meaning to the shell
868 are present, dollar-single-quote escaping is used, and assignments
869 have the form "VARIABLE=$'value'". This syntax is known to be
870 supported by bash(1), zsh(1), ksh(1), and busybox(1)'s ash(1), but
871 not dash(1) or fish(1).
872 set-environment VARIABLE=VALUE...
874 Set one or more systemd manager environment variables, as specified
875 on the command line. This command will fail if variable names and
876 values do not conform to the rules listed above.
877 unset-environment VARIABLE...
879 Unset one or more systemd manager environment variables. If only a
880 variable name is specified, it will be removed regardless of its
881 value. If a variable and a value are specified, the variable is
882 only removed if it has the specified value.
883 import-environment VARIABLE...
885 Import all, one or more environment variables set on the client
886 into the systemd manager environment block. If a list of
887 environment variable names is passed, client-side values are then
888 imported into the manager's environment block. If any names are not
889 valid environment variable names or have invalid values according
890 to the rules described above, an error is raised. If no arguments
891 are passed, the entire environment block inherited by the systemctl
892 process is imported. In this mode, any inherited invalid
893 environment variables are quietly ignored.
894 Importing of the full inherited environment block (calling this
896 command without any arguments) is deprecated. A shell will set
897 dozens of variables which only make sense locally and are only
898 meant for processes which are descendants of the shell. Such
899 variables in the global environment block are confusing to other
900 processes.
901 Manager State Commands
903 daemon-reload
904 Reload the systemd manager configuration. This will rerun all
905 generators (see systemd.generator(7)), reload all unit files, and
906 recreate the entire dependency tree. While the daemon is being
907 reloaded, all sockets systemd listens on behalf of user
908 configuration will stay accessible.
909 This command should not be confused with the reload command.
911 daemon-reexec
913 Reexecute the systemd manager. This will serialize the manager
914 state, reexecute the process and deserialize the state again. This
915 command is of little use except for debugging and package upgrades.
916 Sometimes, it might be helpful as a heavy-weight daemon-reload.
917 While the daemon is being reexecuted, all sockets systemd listening
918 on behalf of user configuration will stay accessible.
919 log-level [LEVEL]
921 If no argument is given, print the current log level of the
922 manager. If an optional argument LEVEL is provided, then the
923 command changes the current log level of the manager to LEVEL
924 (accepts the same values as --log-level= described in systemd(1)).
925 log-target [TARGET]
927 If no argument is given, print the current log target of the
928 manager. If an optional argument TARGET is provided, then the
929 command changes the current log target of the manager to TARGET
930 (accepts the same values as --log-target=, described in
931 systemd(1)).
932 service-watchdogs [yes|no]
934 If no argument is given, print the current state of service runtime
935 watchdogs of the manager. If an optional boolean argument is
936 provided, then globally enables or disables the service runtime
937 watchdogs (WatchdogSec=) and emergency actions (e.g. OnFailure= or
938 StartLimitAction=); see systemd.service(5). The hardware watchdog
939 is not affected by this setting.
940 System Commands
942 is-system-running
943 Checks whether the system is operational. This returns success
944 (exit code 0) when the system is fully up and running, specifically
945 not in startup, shutdown or maintenance mode, and with no failed
946 services. Failure is returned otherwise (exit code non-zero). In
947 addition, the current state is printed in a short string to
948 standard output, see the table below. Use --quiet to suppress this
949 output.
950 Use --wait to wait until the boot process is completed before
952 printing the current state and returning the appropriate error
953 status. If --wait is in use, states initializing or starting will
954 not be reported, instead the command will block until a later state
955 (such as running or degraded) is reached.
956 Table 2. is-system-running output
958 ┌─────────────┬─────────────────────┬───────────┐
959 │Name │ Description │ Exit Code │
960 ├─────────────┼─────────────────────┼───────────┤
961 │initializing │ Early bootup, │ > 0 │
962 │ │ before basic.target │ │
963 │ │ is reached or the │ │
964 │ │ maintenance state │ │
965 │ │ entered. │ │
966 ├─────────────┼─────────────────────┼───────────┤
967 │starting │ Late bootup, before │ > 0 │
968 │ │ the job queue │ │
969 │ │ becomes idle for │ │
970 │ │ the first time, or │ │
971 │ │ one of the rescue │ │
972 │ │ targets are │ │
973 │ │ reached. │ │
974 ├─────────────┼─────────────────────┼───────────┤
975 │running │ The system is fully │ 0 │
976 │ │ operational. │ │
977 ├─────────────┼─────────────────────┼───────────┤
978 │degraded │ The system is │ > 0 │
979 │ │ operational but one │ │
980 │ │ or more units │ │
981 │ │ failed. │ │
982 ├─────────────┼─────────────────────┼───────────┤
983 │maintenance │ The rescue or │ > 0 │
984 │ │ emergency target is │ │
985 │ │ active. │ │
986 ├─────────────┼─────────────────────┼───────────┤
987 │stopping │ The manager is │ > 0 │
988 │ │ shutting down. │ │
989 ├─────────────┼─────────────────────┼───────────┤
990 │offline │ The manager is not │ > 0 │
991 │ │ running. │ │
992 │ │ Specifically, this │ │
993 │ │ is the operational │ │
994 │ │ state if an │ │
995 │ │ incompatible │ │
996 │ │ program is running │ │
997 │ │ as system manager │ │
998 │ │ (PID 1). │ │
999 ├─────────────┼─────────────────────┼───────────┤
1000 │unknown │ The operational │ > 0 │
1001 │ │ state could not be │ │
1002 │ │ determined, due to │ │
1003 │ │ lack of resources │ │
1004 │ │ or another error │ │
1005 │ │ cause. │ │
1006 └─────────────┴─────────────────────┴───────────┘
1007 default
1009 Enter default mode. This is equivalent to systemctl isolate
1010 default.target. This operation is blocking by default, use
1011 --no-block to request asynchronous behavior.
1012 rescue
1014 Enter rescue mode. This is equivalent to systemctl isolate
1015 rescue.target. This operation is blocking by default, use
1016 --no-block to request asynchronous behavior.
1017 emergency
1019 Enter emergency mode. This is equivalent to systemctl isolate
1020 emergency.target. This operation is blocking by default, use
1021 --no-block to request asynchronous behavior.
1022 halt
1024 Shut down and halt the system. This is mostly equivalent to
1025 systemctl start halt.target --job-mode=replace-irreversibly
1026 --no-block, but also prints a wall message to all users. This
1027 command is asynchronous; it will return after the halt operation is
1028 enqueued, without waiting for it to complete. Note that this
1029 operation will simply halt the OS kernel after shutting down,
1030 leaving the hardware powered on. Use systemctl poweroff for
1031 powering off the system (see below).
1032 If combined with --force, shutdown of all running services is
1034 skipped, however all processes are killed and all file systems are
1035 unmounted or mounted read-only, immediately followed by the system
1036 halt. If --force is specified twice, the operation is immediately
1037 executed without terminating any processes or unmounting any file
1038 systems. This may result in data loss. Note that when --force is
1039 specified twice the halt operation is executed by systemctl itself,
1040 and the system manager is not contacted. This means the command
1041 should succeed even when the system manager has crashed.
1042 poweroff
1044 Shut down and power-off the system. This is mostly equivalent to
1045 systemctl start poweroff.target --job-mode=replace-irreversibly
1046 --no-block, but also prints a wall message to all users. This
1047 command is asynchronous; it will return after the power-off
1048 operation is enqueued, without waiting for it to complete.
1049 If combined with --force, shutdown of all running services is
1051 skipped, however all processes are killed and all file systems are
1052 unmounted or mounted read-only, immediately followed by the
1053 powering off. If --force is specified twice, the operation is
1054 immediately executed without terminating any processes or
1055 unmounting any file systems. This may result in data loss. Note
1056 that when --force is specified twice the power-off operation is
1057 executed by systemctl itself, and the system manager is not
1058 contacted. This means the command should succeed even when the
1059 system manager has crashed.
1060 reboot
1062 Shut down and reboot the system. This is mostly equivalent to
1063 systemctl start reboot.target --job-mode=replace-irreversibly
1064 --no-block, but also prints a wall message to all users. This
1065 command is asynchronous; it will return after the reboot operation
1066 is enqueued, without waiting for it to complete.
1067 If combined with --force, shutdown of all running services is
1069 skipped, however all processes are killed and all file systems are
1070 unmounted or mounted read-only, immediately followed by the reboot.
1071 If --force is specified twice, the operation is immediately
1072 executed without terminating any processes or unmounting any file
1073 systems. This may result in data loss. Note that when --force is
1074 specified twice the reboot operation is executed by systemctl
1075 itself, and the system manager is not contacted. This means the
1076 command should succeed even when the system manager has crashed.
1077 If the switch --reboot-argument= is given, it will be passed as the
1079 optional argument to the reboot(2) system call.
1080 kexec
1082 Shut down and reboot the system via kexec. This is equivalent to
1083 systemctl start kexec.target --job-mode=replace-irreversibly
1084 --no-block. This command is asynchronous; it will return after the
1085 reboot operation is enqueued, without waiting for it to complete.
1086 If combined with --force, shutdown of all running services is
1088 skipped, however all processes are killed and all file systems are
1089 unmounted or mounted read-only, immediately followed by the reboot.
1090 exit [EXIT_CODE]
1092 Ask the service manager to quit. This is only supported for user
1093 service managers (i.e. in conjunction with the --user option) or in
1094 containers and is equivalent to poweroff otherwise. This command is
1095 asynchronous; it will return after the exit operation is enqueued,
1096 without waiting for it to complete.
1097 The service manager will exit with the specified exit code, if
1099 EXIT_CODE is passed.
1100 switch-root ROOT [INIT]
1102 Switches to a different root directory and executes a new system
1103 manager process below it. This is intended for usage in initial RAM
1104 disks ("initrd"), and will transition from the initrd's system
1105 manager process (a.k.a. "init" process) to the main system manager
1106 process which is loaded from the actual host volume. This call
1107 takes two arguments: the directory that is to become the new root
1108 directory, and the path to the new system manager binary below it
1109 to execute as PID 1. If the latter is omitted or the empty string,
1110 a systemd binary will automatically be searched for and used as
1111 init. If the system manager path is omitted, equal to the empty
1112 string or identical to the path to the systemd binary, the state of
1113 the initrd's system manager process is passed to the main system
1114 manager, which allows later introspection of the state of the
1115 services involved in the initrd boot phase.
1116 suspend
1118 Suspend the system. This will trigger activation of the special
1119 target unit suspend.target. This command is asynchronous, and will
1120 return after the suspend operation is successfully enqueued. It
1121 will not wait for the suspend/resume cycle to complete.
1122 hibernate
1124 Hibernate the system. This will trigger activation of the special
1125 target unit hibernate.target. This command is asynchronous, and
1126 will return after the hibernation operation is successfully
1127 enqueued. It will not wait for the hibernate/thaw cycle to
1128 complete.
1129 hybrid-sleep
1131 Hibernate and suspend the system. This will trigger activation of
1132 the special target unit hybrid-sleep.target. This command is
1133 asynchronous, and will return after the hybrid sleep operation is
1134 successfully enqueued. It will not wait for the sleep/wake-up cycle
1135 to complete.
1136 suspend-then-hibernate
1138 Suspend the system and hibernate it after the delay specified in
1139 systemd-sleep.conf. This will trigger activation of the special
1140 target unit suspend-then-hibernate.target. This command is
1141 asynchronous, and will return after the hybrid sleep operation is
1142 successfully enqueued. It will not wait for the sleep/wake-up or
1143 hibernate/thaw cycle to complete.
1144 Parameter Syntax
1146 Unit commands listed above take either a single unit name (designated
1147 as UNIT), or multiple unit specifications (designated as PATTERN...).
1148 In the first case, the unit name with or without a suffix must be
1149 given. If the suffix is not specified (unit name is "abbreviated"),
1150 systemctl will append a suitable suffix, ".service" by default, and a
1151 type-specific suffix in case of commands which operate only on specific
1152 unit types. For example,
1153 # systemctl start sshd
1155 and
1157 # systemctl start sshd.service
1159 are equivalent, as are
1161 # systemctl isolate default
1163 and
1165 # systemctl isolate default.target
1167 Note that (absolute) paths to device nodes are automatically converted
1169 to device unit names, and other (absolute) paths to mount unit names.
1170 # systemctl status /dev/sda
1172 # systemctl status /home
1173 are equivalent to:
1175 # systemctl status dev-sda.device
1177 # systemctl status home.mount
1178 In the second case, shell-style globs will be matched against the
1180 primary names of all units currently in memory; literal unit names,
1181 with or without a suffix, will be treated as in the first case. This
1182 means that literal unit names always refer to exactly one unit, but
1183 globs may match zero units and this is not considered an error.
1184 Glob patterns use fnmatch(3), so normal shell-style globbing rules are
1186 used, and "*", "?", "[]" may be used. See glob(7) for more details. The
1187 patterns are matched against the primary names of units currently in
1188 memory, and patterns which do not match anything are silently skipped.
1189 For example:
1190 # systemctl stop sshd@*.service
1192 will stop all sshd@.service instances. Note that alias names of units,
1194 and units that aren't in memory are not considered for glob expansion.
1195 For unit file commands, the specified UNIT should be the name of the
1197 unit file (possibly abbreviated, see above), or the absolute path to
1198 the unit file:
1199 # systemctl enable foo.service
1201 or
1203 # systemctl link /path/to/foo.service
1205
1208 The following options are understood:
1209 -t, --type=
1211 The argument should be a comma-separated list of unit types such as
1212 service and socket.
1213 If one of the arguments is a unit type, when listing units, limit
1215 display to certain unit types. Otherwise, units of all types will
1216 be shown.
1217 As a special case, if one of the arguments is help, a list of
1219 allowed values will be printed and the program will exit.
1220 --state=
1222 The argument should be a comma-separated list of unit LOAD, SUB, or
1223 ACTIVE states. When listing units, show only those in the specified
1224 states. Use --state=failed to show only failed units.
1225 As a special case, if one of the arguments is help, a list of
1227 allowed values will be printed and the program will exit.
1228 -p, --property=
1230 When showing unit/job/manager properties with the show command,
1231 limit display to properties specified in the argument. The argument
1232 should be a comma-separated list of property names, such as
1233 "MainPID". Unless specified, all known properties are shown. If
1234 specified more than once, all properties with the specified names
1235 are shown. Shell completion is implemented for property names.
1236 For the manager itself, systemctl show will show all available
1238 properties. Those properties are documented in systemd-
1239 system.conf(5).
1240 Properties for units vary by unit type, so showing any unit (even a
1242 non-existent one) is a way to list properties pertaining to this
1243 type. Similarly, showing any job will list properties pertaining to
1244 all jobs. Properties for units are documented in systemd.unit(5),
1245 and the pages for individual unit types systemd.service(5),
1246 systemd.socket(5), etc.
1247 -P
1249 Equivalent to --value --property=, i.e. shows the value of the
1250 property without the property name or "=". Note that using -P once
1251 will also affect all properties listed with -p/--property=.
1252 -a, --all
1254 When listing units with list-units, also show inactive units and
1255 units which are following other units. When showing
1256 unit/job/manager properties, show all properties regardless whether
1257 they are set or not.
1258 To list all units installed in the file system, use the
1260 list-unit-files command instead.
1261 When listing units with list-dependencies, recursively show
1263 dependencies of all dependent units (by default only dependencies
1264 of target units are shown).
1265 When used with status, show journal messages in full, even if they
1267 include unprintable characters or are very long. By default, fields
1268 with unprintable characters are abbreviated as "blob data". (Note
1269 that the pager may escape unprintable characters again.)
1270 -r, --recursive
1272 When listing units, also show units of local containers. Units of
1273 local containers will be prefixed with the container name,
1274 separated by a single colon character (":").
1275 --reverse
1277 Show reverse dependencies between units with list-dependencies,
1278 i.e. follow dependencies of type WantedBy=, RequiredBy=, PartOf=,
1279 BoundBy=, instead of Wants= and similar.
1280 --after
1282 With list-dependencies, show the units that are ordered before the
1283 specified unit. In other words, recursively list units following
1284 the After= dependency.
1285 Note that any After= dependency is automatically mirrored to create
1287 a Before= dependency. Temporal dependencies may be specified
1288 explicitly, but are also created implicitly for units which are
1289 WantedBy= targets (see systemd.target(5)), and as a result of other
1290 directives (for example RequiresMountsFor=). Both explicitly and
1291 implicitly introduced dependencies are shown with
1292 list-dependencies.
1293 When passed to the list-jobs command, for each printed job show
1295 which other jobs are waiting for it. May be combined with --before
1296 to show both the jobs waiting for each job as well as all jobs each
1297 job is waiting for.
1298 --before
1300 With list-dependencies, show the units that are ordered after the
1301 specified unit. In other words, recursively list units following
1302 the Before= dependency.
1303 When passed to the list-jobs command, for each printed job show
1305 which other jobs it is waiting for. May be combined with --after to
1306 show both the jobs waiting for each job as well as all jobs each
1307 job is waiting for.
1308 --with-dependencies
1310 When used with status, cat, list-units, and list-unit-files, those
1311 commands print all specified units and the dependencies of those
1312 units.
1313 Options --reverse, --after, --before may be used to change what
1315 types of dependencies are shown.
1316 -l, --full
1318 Do not ellipsize unit names, process tree entries, journal output,
1319 or truncate unit descriptions in the output of status, list-units,
1320 list-jobs, and list-timers.
1321 Also, show installation targets in the output of is-enabled.
1323 --value
1325 When printing properties with show, only print the value, and skip
1326 the property name and "=". Also see option -P above.
1327 --show-types
1329 When showing sockets, show the type of the socket.
1330 --job-mode=
1332 When queuing a new job, this option controls how to deal with
1333 already queued jobs. It takes one of "fail", "replace",
1334 "replace-irreversibly", "isolate", "ignore-dependencies",
1335 "ignore-requirements", "flush", or "triggering". Defaults to
1336 "replace", except when the isolate command is used which implies
1337 the "isolate" job mode.
1338 If "fail" is specified and a requested operation conflicts with a
1340 pending job (more specifically: causes an already pending start job
1341 to be reversed into a stop job or vice versa), cause the operation
1342 to fail.
1343 If "replace" (the default) is specified, any conflicting pending
1345 job will be replaced, as necessary.
1346 If "replace-irreversibly" is specified, operate like "replace", but
1348 also mark the new jobs as irreversible. This prevents future
1349 conflicting transactions from replacing these jobs (or even being
1350 enqueued while the irreversible jobs are still pending).
1351 Irreversible jobs can still be cancelled using the cancel command.
1352 This job mode should be used on any transaction which pulls in
1353 shutdown.target.
1354 "isolate" is only valid for start operations and causes all other
1356 units to be stopped when the specified unit is started. This mode
1357 is always used when the isolate command is used.
1358 "flush" will cause all queued jobs to be canceled when the new job
1360 is enqueued.
1361 If "ignore-dependencies" is specified, then all unit dependencies
1363 are ignored for this new job and the operation is executed
1364 immediately. If passed, no required units of the unit passed will
1365 be pulled in, and no ordering dependencies will be honored. This is
1366 mostly a debugging and rescue tool for the administrator and should
1367 not be used by applications.
1368 "ignore-requirements" is similar to "ignore-dependencies", but only
1370 causes the requirement dependencies to be ignored, the ordering
1371 dependencies will still be honored.
1372 "triggering" may only be used with systemctl stop. In this mode,
1374 the specified unit and any active units that trigger it are
1375 stopped. See the discussion of Triggers= in systemd.unit(5) for
1376 more information about triggering units.
1377 -T, --show-transaction
1379 When enqueuing a unit job (for example as effect of a systemctl
1380 start invocation or similar), show brief information about all jobs
1381 enqueued, covering both the requested job and any added because of
1382 unit dependencies. Note that the output will only include jobs
1383 immediately part of the transaction requested. It is possible that
1384 service start-up program code run as effect of the enqueued jobs
1385 might request further jobs to be pulled in. This means that
1386 completion of the listed jobs might ultimately entail more jobs
1387 than the listed ones.
1388 --fail
1390 Shorthand for --job-mode=fail.
1391 When used with the kill command, if no units were killed, the
1393 operation results in an error.
1394 --check-inhibitors=
1396 When system shutdown or sleep state is request, this option
1397 controls how to deal with inhibitor locks. It takes one of "auto",
1398 "yes" or "no". Defaults to "auto", which will behave like "yes" for
1399 interactive invocations (i.e. from a TTY) and "no" for
1400 non-interactive invocations. "yes" will let the request respect
1401 inhibitor locks. "no" will let the request ignore inhibitor locks.
1402 Applications can establish inhibitor locks to avoid that certain
1404 important operations (such as CD burning or suchlike) are
1405 interrupted by system shutdown or a sleep state. Any user may take
1406 these locks and privileged users may override these locks. If any
1407 locks are taken, shutdown and sleep state requests will normally
1408 fail (unless privileged) and a list of active locks is printed.
1409 However, if "no" is specified or "auto" is specified on a
1410 non-interactive requests, the established locks are ignored and not
1411 shown, and the operation attempted anyway, possibly requiring
1412 additional privileges. May be overridden by --force.
1413 -i
1415 Shortcut for --check-inhibitors=no.
1416 --dry-run
1418 Just print what would be done. Currently supported by verbs halt,
1419 poweroff, reboot, kexec, suspend, hibernate, hybrid-sleep,
1420 suspend-then-hibernate, default, rescue, emergency, and exit.
1421 -q, --quiet
1423 Suppress printing of the results of various commands and also the
1424 hints about truncated log lines. This does not suppress output of
1425 commands for which the printed output is the only result (like
1426 show). Errors are always printed.
1427 --no-block
1429 Do not synchronously wait for the requested operation to finish. If
1430 this is not specified, the job will be verified, enqueued and
1431 systemctl will wait until the unit's start-up is completed. By
1432 passing this argument, it is only verified and enqueued. This
1433 option may not be combined with --wait.
1434 --wait
1436 Synchronously wait for started units to terminate again. This
1437 option may not be combined with --no-block. Note that this will
1438 wait forever if any given unit never terminates (by itself or by
1439 getting stopped explicitly); particularly services which use
1440 "RemainAfterExit=yes".
1441 When used with is-system-running, wait until the boot process is
1443 completed before returning.
1444 --user
1446 Talk to the service manager of the calling user, rather than the
1447 service manager of the system.
1448 --system
1450 Talk to the service manager of the system. This is the implied
1451 default.
1452 --failed
1454 List units in failed state. This is equivalent to --state=failed.
1455 --no-wall
1457 Do not send wall message before halt, power-off and reboot.
1458 --global
1460 When used with enable and disable, operate on the global user
1461 configuration directory, thus enabling or disabling a unit file
1462 globally for all future logins of all users.
1463 --no-reload
1465 When used with enable and disable, do not implicitly reload daemon
1466 configuration after executing the changes.
1467 --no-ask-password
1469 When used with start and related commands, disables asking for
1470 passwords. Background services may require input of a password or
1471 passphrase string, for example to unlock system hard disks or
1472 cryptographic certificates. Unless this option is specified and the
1473 command is invoked from a terminal, systemctl will query the user
1474 on the terminal for the necessary secrets. Use this option to
1475 switch this behavior off. In this case, the password must be
1476 supplied by some other means (for example graphical password
1477 agents) or the service might fail. This also disables querying the
1478 user for authentication for privileged operations.
1479 --kill-who=
1481 When used with kill, choose which processes to send a signal to.
1482 Must be one of main, control or all to select whether to kill only
1483 the main process, the control process or all processes of the unit.
1484 The main process of the unit is the one that defines the life-time
1485 of it. A control process of a unit is one that is invoked by the
1486 manager to induce state changes of it. For example, all processes
1487 started due to the ExecStartPre=, ExecStop= or ExecReload= settings
1488 of service units are control processes. Note that there is only one
1489 control process per unit at a time, as only one state change is
1490 executed at a time. For services of type Type=forking, the initial
1491 process started by the manager for ExecStart= is a control process,
1492 while the process ultimately forked off by that one is then
1493 considered the main process of the unit (if it can be determined).
1494 This is different for service units of other types, where the
1495 process forked off by the manager for ExecStart= is always the main
1496 process itself. A service unit consists of zero or one main
1497 process, zero or one control process plus any number of additional
1498 processes. Not all unit types manage processes of these types
1499 however. For example, for mount units, control processes are
1500 defined (which are the invocations of /usr/bin/mount and
1501 /usr/bin/umount), but no main process is defined. If omitted,
1502 defaults to all.
1503 -s, --signal=
1505 When used with kill, choose which signal to send to selected
1506 processes. Must be one of the well-known signal specifiers such as
1507 SIGTERM, SIGINT or SIGSTOP. If omitted, defaults to SIGTERM.
1508 The special value "help" will list the known values and the program
1510 will exit immediately, and the special value "list" will list known
1511 values along with the numerical signal numbers and the program will
1512 exit immediately.
1513 --what=
1515 Select what type of per-unit resources to remove when the clean
1516 command is invoked, see below. Takes one of configuration, state,
1517 cache, logs, runtime to select the type of resource. This option
1518 may be specified more than once, in which case all specified
1519 resource types are removed. Also accepts the special value all as a
1520 shortcut for specifying all five resource types. If this option is
1521 not specified defaults to the combination of cache and runtime,
1522 i.e. the two kinds of resources that are generally considered to be
1523 redundant and can be reconstructed on next invocation.
1524 -f, --force
1526 When used with enable, overwrite any existing conflicting symlinks.
1527 When used with edit, create all of the specified units which do not
1529 already exist.
1530 When used with halt, poweroff, reboot or kexec, execute the
1532 selected operation without shutting down all units. However, all
1533 processes will be killed forcibly and all file systems are
1534 unmounted or remounted read-only. This is hence a drastic but
1535 relatively safe option to request an immediate reboot. If --force
1536 is specified twice for these operations (with the exception of
1537 kexec), they will be executed immediately, without terminating any
1538 processes or unmounting any file systems. Warning: specifying
1539 --force twice with any of these operations might result in data
1540 loss. Note that when --force is specified twice the selected
1541 operation is executed by systemctl itself, and the system manager
1542 is not contacted. This means the command should succeed even when
1543 the system manager has crashed.
1544 --message=
1546 When used with halt, poweroff or reboot, set a short message
1547 explaining the reason for the operation. The message will be logged
1548 together with the default shutdown message.
1549 --now
1551 When used with enable, the units will also be started. When used
1552 with disable or mask, the units will also be stopped. The start or
1553 stop operation is only carried out when the respective enable or
1554 disable operation has been successful.
1555 --root=
1557 When used with enable/disable/is-enabled (and related commands),
1558 use the specified root path when looking for unit files. If this
1559 option is present, systemctl will operate on the file system
1560 directly, instead of communicating with the systemd daemon to carry
1561 out changes.
1562 --runtime
1564 When used with enable, disable, edit, (and related commands), make
1565 changes only temporarily, so that they are lost on the next reboot.
1566 This will have the effect that changes are not made in
1567 subdirectories of /etc/ but in /run/, with identical immediate
1568 effects, however, since the latter is lost on reboot, the changes
1569 are lost too.
1570 Similarly, when used with set-property, make changes only
1572 temporarily, so that they are lost on the next reboot.
1573 --preset-mode=
1575 Takes one of "full" (the default), "enable-only", "disable-only".
1576 When used with the preset or preset-all commands, controls whether
1577 units shall be disabled and enabled according to the preset rules,
1578 or only enabled, or only disabled.
1579 -n, --lines=
1581 When used with status, controls the number of journal lines to
1582 show, counting from the most recent ones. Takes a positive integer
1583 argument, or 0 to disable journal output. Defaults to 10.
1584 -o, --output=
1586 When used with status, controls the formatting of the journal
1587 entries that are shown. For the available choices, see
1588 journalctl(1). Defaults to "short".
1589 --firmware-setup
1591 When used with the reboot command, indicate to the system's
1592 firmware to reboot into the firmware setup interface. Note that
1593 this functionality is not available on all systems.
1594 --boot-loader-menu=
1596 When used with the reboot command, indicate to the system's boot
1597 loader to show the boot loader menu on the following boot. Takes a
1598 time value as parameter — indicating the menu timeout. Pass zero in
1599 order to disable the menu timeout. Note that not all boot loaders
1600 support this functionality.
1601 --boot-loader-entry=
1603 When used with the reboot command, indicate to the system's boot
1604 loader to boot into a specific boot loader entry on the following
1605 boot. Takes a boot loader entry identifier as argument, or "help"
1606 in order to list available entries. Note that not all boot loaders
1607 support this functionality.
1608 --reboot-argument=
1610 This switch is used with reboot. The value is architecture and
1611 firmware specific. As an example, "recovery" might be used to
1612 trigger system recovery, and "fota" might be used to trigger a
1613 “firmware over the air” update.
1614 --plain
1616 When used with list-dependencies, list-units or list-machines, the
1617 output is printed as a list instead of a tree, and the bullet
1618 circles are omitted.
1619 --timestamp=
1621 Change the format of printed timestamps. The following values may
1622 be used:
1623 pretty (this is the default)
1625 "Day YYYY-MM-DD HH:MM:SS TZ"
1626 us, µs
1628 "Day YYYY-MM-DD HH:MM:SS.UUUUUU TZ"
1629 utc
1631 "Day YYYY-MM-DD HH:MM:SS UTC"
1632 us+utc, µs+utc
1634 "Day YYYY-MM-DD HH:MM:SS.UUUUUU UTC"
1635 --mkdir
1637 When used with bind, creates the destination file or directory
1638 before applying the bind mount. Note that even though the name of
1639 this option suggests that it is suitable only for directories, this
1640 option also creates the destination file node to mount over if the
1641 object to mount is not a directory, but a regular file, device
1642 node, socket or FIFO.
1643 --marked
1645 Only allowed with reload-or-restart. Enqueues restart jobs for all
1646 units that have the "needs-restart" mark, and reload jobs for units
1647 that have the "needs-reload" mark. When a unit marked for reload
1648 does not support reload, restart will be queued. Those properties
1649 can be set using set-property Marks.
1650 Unless --no-block is used, systemctl will wait for the queued jobs
1652 to finish.
1653 --read-only
1655 When used with bind, creates a read-only bind mount.
1656 -H, --host=
1658 Execute the operation remotely. Specify a hostname, or a username
1659 and hostname separated by "@", to connect to. The hostname may
1660 optionally be suffixed by a port ssh is listening on, separated by
1661 ":", and then a container name, separated by "/", which connects
1662 directly to a specific container on the specified host. This will
1663 use SSH to talk to the remote machine manager instance. Container
1664 names may be enumerated with machinectl -H HOST. Put IPv6 addresses
1665 in brackets.
1666 -M, --machine=
1668 Execute operation on a local container. Specify a container name to
1669 connect to, optionally prefixed by a user name to connect as and a
1670 separating "@" character. If the special string ".host" is used in
1671 place of the container name, a connection to the local system is
1672 made (which is useful to connect to a specific user's user bus:
1673 "--user --machine=lennart@.host"). If the "@" syntax is not used,
1674 the connection is made as root user. If the "@" syntax is used
1675 either the left hand side or the right hand side may be omitted
1676 (but not both) in which case the local user name and ".host" are
1677 implied.
1678 --no-pager
1680 Do not pipe output into a pager.
1681 --legend=BOOL
1683 Enable or disable printing of the legend, i.e. column headers and
1684 the footer with hints. The legend is printed by default, unless
1685 disabled with --quiet or similar.
1686 -h, --help
1688 Print a short help text and exit.
1689 --version
1691 Print a short version string and exit.
1692
1694 On success, 0 is returned, a non-zero failure code otherwise.
1695 systemctl uses the return codes defined by LSB, as defined in LSB
1697 3.0.0[1].
1698 Table 3. LSB return codes
1700 ┌──────┬─────────────────────┬─────────────────────┐
1701 │Value │ Description in LSB │ Use in systemd │
1702 ├──────┼─────────────────────┼─────────────────────┤
1703 │0 │ "program is running │ unit is active │
1704 │ │ or service is OK" │ │
1705 ├──────┼─────────────────────┼─────────────────────┤
1706 │1 │ "program is dead │ unit not failed │
1707 │ │ and /var/run pid │ (used by is-failed) │
1708 │ │ file exists" │ │
1709 ├──────┼─────────────────────┼─────────────────────┤
1710 │2 │ "program is dead │ unused │
1711 │ │ and /var/lock lock │ │
1712 │ │ file exists" │ │
1713 ├──────┼─────────────────────┼─────────────────────┤
1714 │3 │ "program is not │ unit is not active │
1715 │ │ running" │ │
1716 ├──────┼─────────────────────┼─────────────────────┤
1717 │4 │ "program or service │ no such unit │
1718 │ │ status is unknown" │ │
1719 └──────┴─────────────────────┴─────────────────────┘
1720 The mapping of LSB service states to systemd unit states is imperfect,
1722 so it is better to not rely on those return values but to look for
1723 specific unit states and substates instead.
1724
1726 $SYSTEMD_EDITOR
1727 Editor to use when editing units; overrides $EDITOR and $VISUAL. If
1728 neither $SYSTEMD_EDITOR nor $EDITOR nor $VISUAL are present or if
1729 it is set to an empty string or if their execution failed,
1730 systemctl will try to execute well known editors in this order:
1731 editor(1), nano(1), vim(1), vi(1).
1732 $SYSTEMD_LOG_LEVEL
1734 The maximum log level of emitted messages (messages with a higher
1735 log level, i.e. less important ones, will be suppressed). Either
1736 one of (in order of decreasing importance) emerg, alert, crit, err,
1737 warning, notice, info, debug, or an integer in the range 0...7. See
1738 syslog(3) for more information.
1739 $SYSTEMD_LOG_COLOR
1741 A boolean. If true, messages written to the tty will be colored
1742 according to priority.
1743 This setting is only useful when messages are written directly to
1745 the terminal, because journalctl(1) and other tools that display
1746 logs will color messages based on the log level on their own.
1747 $SYSTEMD_LOG_TIME
1749 A boolean. If true, log messages will be prefixed with a timestamp.
1750 This setting is only useful when messages are written directly to
1752 the terminal or a file, because journalctl(1) and other tools that
1753 display logs will attach timestamps based on the entry metadata on
1754 their own.
1755 $SYSTEMD_LOG_LOCATION
1757 A boolean. If true, messages will be prefixed with a filename and
1758 line number in the source code where the message originates.
1759 Note that the log location is often attached as metadata to journal
1761 entries anyway. Including it directly in the message text can
1762 nevertheless be convenient when debugging programs.
1763 $SYSTEMD_LOG_TARGET
1765 The destination for log messages. One of console (log to the
1766 attached tty), console-prefixed (log to the attached tty but with
1767 prefixes encoding the log level and "facility", see syslog(3), kmsg
1768 (log to the kernel circular log buffer), journal (log to the
1769 journal), journal-or-kmsg (log to the journal if available, and to
1770 kmsg otherwise), auto (determine the appropriate log target
1771 automatically, the default), null (disable log output).
1772 $SYSTEMD_PAGER
1774 Pager to use when --no-pager is not given; overrides $PAGER. If
1775 neither $SYSTEMD_PAGER nor $PAGER are set, a set of well-known
1776 pager implementations are tried in turn, including less(1) and
1777 more(1), until one is found. If no pager implementation is
1778 discovered no pager is invoked. Setting this environment variable
1779 to an empty string or the value "cat" is equivalent to passing
1780 --no-pager.
1781 $SYSTEMD_LESS
1783 Override the options passed to less (by default "FRSXMK").
1784 Users might want to change two options in particular:
1786 K
1788 This option instructs the pager to exit immediately when Ctrl+C
1789 is pressed. To allow less to handle Ctrl+C itself to switch
1790 back to the pager command prompt, unset this option.
1791 If the value of $SYSTEMD_LESS does not include "K", and the
1793 pager that is invoked is less, Ctrl+C will be ignored by the
1794 executable, and needs to be handled by the pager.
1795 X
1797 This option instructs the pager to not send termcap
1798 initialization and deinitialization strings to the terminal. It
1799 is set by default to allow command output to remain visible in
1800 the terminal even after the pager exits. Nevertheless, this
1801 prevents some pager functionality from working, in particular
1802 paged output cannot be scrolled with the mouse.
1803 See less(1) for more discussion.
1805 $SYSTEMD_LESSCHARSET
1807 Override the charset passed to less (by default "utf-8", if the
1808 invoking terminal is determined to be UTF-8 compatible).
1809 $SYSTEMD_PAGERSECURE
1811 Takes a boolean argument. When true, the "secure" mode of the pager
1812 is enabled; if false, disabled. If $SYSTEMD_PAGERSECURE is not set
1813 at all, secure mode is enabled if the effective UID is not the same
1814 as the owner of the login session, see geteuid(2) and
1815 sd_pid_get_owner_uid(3). In secure mode, LESSSECURE=1 will be set
1816 when invoking the pager, and the pager shall disable commands that
1817 open or create new files or start new subprocesses. When
1818 $SYSTEMD_PAGERSECURE is not set at all, pagers which are not known
1819 to implement secure mode will not be used. (Currently only less(1)
1820 implements secure mode.)
1821 Note: when commands are invoked with elevated privileges, for
1823 example under sudo(8) or pkexec(1), care must be taken to ensure
1824 that unintended interactive features are not enabled. "Secure" mode
1825 for the pager may be enabled automatically as describe above.
1826 Setting SYSTEMD_PAGERSECURE=0 or not removing it from the inherited
1827 environment allows the user to invoke arbitrary commands. Note that
1828 if the $SYSTEMD_PAGER or $PAGER variables are to be honoured,
1829 $SYSTEMD_PAGERSECURE must be set too. It might be reasonable to
1830 completely disable the pager using --no-pager instead.
1831 $SYSTEMD_COLORS
1833 Takes a boolean argument. When true, systemd and related utilities
1834 will use colors in their output, otherwise the output will be
1835 monochrome. Additionally, the variable can take one of the
1836 following special values: "16", "256" to restrict the use of colors
1837 to the base 16 or 256 ANSI colors, respectively. This can be
1838 specified to override the automatic decision based on $TERM and
1839 what the console is connected to.
1840 $SYSTEMD_URLIFY
1842 The value must be a boolean. Controls whether clickable links
1843 should be generated in the output for terminal emulators supporting
1844 this. This can be specified to override the decision that systemd
1845 makes based on $TERM and other conditions.
1846
1848 systemd(1), journalctl(1), loginctl(1), machinectl(1), systemd.unit(5),
1849 systemd.resource-control(5), systemd.special(7), wall(1),
1850 systemd.preset(5), systemd.generator(7), glob(7)
1851
1853 1. LSB 3.0.0
1854 http://refspecs.linuxbase.org/LSB_3.0.0/LSB-PDA/LSB-PDA/iniscrptact.html
1855systemd 248 SYSTEMCTL(1)