1SYSTEMD.SERVICE(5) systemd.service SYSTEMD.SERVICE(5)
2
6 systemd.service - Service unit configuration
7
9 service.service
10
12 A unit configuration file whose name ends in ".service" encodes
13 information about a process controlled and supervised by systemd.
14 This man page lists the configuration options specific to this unit
16 type. See systemd.unit(5) for the common options of all unit
17 configuration files. The common configuration items are configured in
18 the generic [Unit] and [Install] sections. The service specific
19 configuration options are configured in the [Service] section.
20 Additional options are listed in systemd.exec(5), which define the
22 execution environment the commands are executed in, and in
23 systemd.kill(5), which define the way the processes of the service are
24 terminated, and in systemd.resource-control(5), which configure
25 resource control settings for the processes of the service.
26 If a service is requested under a certain name but no unit
28 configuration file is found, systemd looks for a SysV init script by
29 the same name (with the .service suffix removed) and dynamically
30 creates a service unit from that script. This is useful for
31 compatibility with SysV. Note that this compatibility is quite
32 comprehensive but not 100%. For details about the incompatibilities,
33 see the Incompatibilities with SysV[1] document.
34 The systemd-run(1) command allows creating .service and .scope units
36 dynamically and transiently from the command line.
37
39 It is possible for systemd services to take a single argument via the
40 "service@argument.service" syntax. Such services are called
41 "instantiated" services, while the unit definition without the argument
42 parameter is called a "template". An example could be a dhcpcd@.service
43 service template which takes a network interface as a parameter to form
44 an instantiated service. Within the service file, this parameter or
45 "instance name" can be accessed with %-specifiers. See systemd.unit(5)
46 for details.
47
49 Implicit Dependencies
50 The following dependencies are implicitly added:
51 • Services with Type=dbus set automatically acquire dependencies of
53 type Requires= and After= on dbus.socket.
54 • Socket activated services are automatically ordered after their
56 activating .socket units via an automatic After= dependency.
57 Services also pull in all .socket units listed in Sockets= via
58 automatic Wants= and After= dependencies.
59 Additional implicit dependencies may be added as result of execution
61 and resource control parameters as documented in systemd.exec(5) and
62 systemd.resource-control(5).
63 Default Dependencies
65 The following dependencies are added unless DefaultDependencies=no is
66 set:
67 • Service units will have dependencies of type Requires= and After=
69 on sysinit.target, a dependency of type After= on basic.target as
70 well as dependencies of type Conflicts= and Before= on
71 shutdown.target. These ensure that normal service units pull in
72 basic system initialization, and are terminated cleanly prior to
73 system shutdown. Only services involved with early boot or late
74 system shutdown should disable this option.
75 • Instanced service units (i.e. service units with an "@" in their
77 name) are assigned by default a per-template slice unit (see
78 systemd.slice(5)), named after the template unit, containing all
79 instances of the specific template. This slice is normally stopped
80 at shutdown, together with all template instances. If that is not
81 desired, set DefaultDependencies=no in the template unit, and
82 either define your own per-template slice unit file that also sets
83 DefaultDependencies=no, or set Slice=system.slice (or another
84 suitable slice) in the template unit. Also see systemd.resource-
85 control(5).
86
88 Service files must include a [Service] section, which carries
89 information about the service and the process it supervises. A number
90 of options that may be used in this section are shared with other unit
91 types. These options are documented in systemd.exec(5), systemd.kill(5)
92 and systemd.resource-control(5). The options specific to the [Service]
93 section of service units are the following:
94 Type=
96 Configures the process start-up type for this service unit. One of
97 simple, exec, forking, oneshot, dbus, notify or idle:
98 • If set to simple (the default if ExecStart= is specified but
100 neither Type= nor BusName= are), the service manager will
101 consider the unit started immediately after the main service
102 process has been forked off. It is expected that the process
103 configured with ExecStart= is the main process of the service.
104 In this mode, if the process offers functionality to other
105 processes on the system, its communication channels should be
106 installed before the service is started up (e.g. sockets set up
107 by systemd, via socket activation), as the service manager will
108 immediately proceed starting follow-up units, right after
109 creating the main service process, and before executing the
110 service's binary. Note that this means systemctl start command
111 lines for simple services will report success even if the
112 service's binary cannot be invoked successfully (for example
113 because the selected User= doesn't exist, or the service binary
114 is missing).
115 • The exec type is similar to simple, but the service manager
117 will consider the unit started immediately after the main
118 service binary has been executed. The service manager will
119 delay starting of follow-up units until that point. (Or in
120 other words: simple proceeds with further jobs right after
121 fork() returns, while exec will not proceed before both fork()
122 and execve() in the service process succeeded.) Note that this
123 means systemctl start command lines for exec services will
124 report failure when the service's binary cannot be invoked
125 successfully (for example because the selected User= doesn't
126 exist, or the service binary is missing).
127 • If set to forking, it is expected that the process configured
129 with ExecStart= will call fork() as part of its start-up. The
130 parent process is expected to exit when start-up is complete
131 and all communication channels are set up. The child continues
132 to run as the main service process, and the service manager
133 will consider the unit started when the parent process exits.
134 This is the behavior of traditional UNIX services. If this
135 setting is used, it is recommended to also use the PIDFile=
136 option, so that systemd can reliably identify the main process
137 of the service. systemd will proceed with starting follow-up
138 units as soon as the parent process exits.
139 • Behavior of oneshot is similar to simple; however, the service
141 manager will consider the unit up after the main process exits.
142 It will then start follow-up units. RemainAfterExit= is
143 particularly useful for this type of service. Type=oneshot is
144 the implied default if neither Type= nor ExecStart= are
145 specified. Note that if this option is used without
146 RemainAfterExit= the service will never enter "active" unit
147 state, but directly transition from "activating" to
148 "deactivating" or "dead" since no process is configured that
149 shall run continuously. In particular this means that after a
150 service of this type ran (and which has RemainAfterExit= not
151 set) it will not show up as started afterwards, but as dead.
152 • Behavior of dbus is similar to simple; however, it is expected
154 that the service acquires a name on the D-Bus bus, as
155 configured by BusName=. systemd will proceed with starting
156 follow-up units after the D-Bus bus name has been acquired.
157 Service units with this option configured implicitly gain
158 dependencies on the dbus.socket unit. This type is the default
159 if BusName= is specified. A service unit of this type is
160 considered to be in the activating state until the specified
161 bus name is acquired. It is considered activated while the bus
162 name is taken. Once the bus name is released the service is
163 considered being no longer functional which has the effect that
164 the service manager attempts to terminate any remaining
165 processes belonging to the service. Services that drop their
166 bus name as part of their shutdown logic thus should be
167 prepared to receive a SIGTERM (or whichever signal is
168 configured in KillSignal=) as result.
169 • Behavior of notify is similar to exec; however, it is expected
171 that the service sends a notification message via sd_notify(3)
172 or an equivalent call when it has finished starting up. systemd
173 will proceed with starting follow-up units after this
174 notification message has been sent. If this option is used,
175 NotifyAccess= (see below) should be set to open access to the
176 notification socket provided by systemd. If NotifyAccess= is
177 missing or set to none, it will be forcibly set to main.
178 • Behavior of idle is very similar to simple; however, actual
180 execution of the service program is delayed until all active
181 jobs are dispatched. This may be used to avoid interleaving of
182 output of shell services with the status output on the console.
183 Note that this type is useful only to improve console output,
184 it is not useful as a general unit ordering tool, and the
185 effect of this service type is subject to a 5s timeout, after
186 which the service program is invoked anyway.
187 It is generally recommended to use Type=simple for long-running
189 services whenever possible, as it is the simplest and fastest
190 option. However, as this service type won't propagate service
191 start-up failures and doesn't allow ordering of other units against
192 completion of initialization of the service (which for example is
193 useful if clients need to connect to the service through some form
194 of IPC, and the IPC channel is only established by the service
195 itself — in contrast to doing this ahead of time through socket or
196 bus activation or similar), it might not be sufficient for many
197 cases. If so, notify or dbus (the latter only in case the service
198 provides a D-Bus interface) are the preferred options as they allow
199 service program code to precisely schedule when to consider the
200 service started up successfully and when to proceed with follow-up
201 units. The notify service type requires explicit support in the
202 service codebase (as sd_notify() or an equivalent API needs to be
203 invoked by the service at the appropriate time) — if it's not
204 supported, then forking is an alternative: it supports the
205 traditional UNIX service start-up protocol. Finally, exec might be
206 an option for cases where it is enough to ensure the service binary
207 is invoked, and where the service binary itself executes no or
208 little initialization on its own (and its initialization is
209 unlikely to fail). Note that using any type other than simple
210 possibly delays the boot process, as the service manager needs to
211 wait for service initialization to complete. It is hence
212 recommended not to needlessly use any types other than simple.
213 (Also note it is generally not recommended to use idle or oneshot
214 for long-running services.)
215 RemainAfterExit=
217 Takes a boolean value that specifies whether the service shall be
218 considered active even when all its processes exited. Defaults to
219 no.
220 GuessMainPID=
222 Takes a boolean value that specifies whether systemd should try to
223 guess the main PID of a service if it cannot be determined
224 reliably. This option is ignored unless Type=forking is set and
225 PIDFile= is unset because for the other types or with an explicitly
226 configured PID file, the main PID is always known. The guessing
227 algorithm might come to incorrect conclusions if a daemon consists
228 of more than one process. If the main PID cannot be determined,
229 failure detection and automatic restarting of a service will not
230 work reliably. Defaults to yes.
231 PIDFile=
233 Takes a path referring to the PID file of the service. Usage of
234 this option is recommended for services where Type= is set to
235 forking. The path specified typically points to a file below /run/.
236 If a relative path is specified it is hence prefixed with /run/.
237 The service manager will read the PID of the main process of the
238 service from this file after start-up of the service. The service
239 manager will not write to the file configured here, although it
240 will remove the file after the service has shut down if it still
241 exists. The PID file does not need to be owned by a privileged
242 user, but if it is owned by an unprivileged user additional safety
243 restrictions are enforced: the file may not be a symlink to a file
244 owned by a different user (neither directly nor indirectly), and
245 the PID file must refer to a process already belonging to the
246 service.
247 Note that PID files should be avoided in modern projects. Use
249 Type=notify or Type=simple where possible, which does not require
250 use of PID files to determine the main process of a service and
251 avoids needless forking.
252 BusName=
254 Takes a D-Bus destination name that this service shall use. This
255 option is mandatory for services where Type= is set to dbus. It is
256 recommended to always set this property if known to make it easy to
257 map the service name to the D-Bus destination. In particular,
258 systemctl service-log-level/service-log-target verbs make use of
259 this.
260 ExecStart=
262 Commands with their arguments that are executed when this service
263 is started. The value is split into zero or more command lines
264 according to the rules described below (see section "Command Lines"
265 below).
266 Unless Type= is oneshot, exactly one command must be given. When
268 Type=oneshot is used, zero or more commands may be specified.
269 Commands may be specified by providing multiple command lines in
270 the same directive, or alternatively, this directive may be
271 specified more than once with the same effect. If the empty string
272 is assigned to this option, the list of commands to start is reset,
273 prior assignments of this option will have no effect. If no
274 ExecStart= is specified, then the service must have
275 RemainAfterExit=yes and at least one ExecStop= line set. (Services
276 lacking both ExecStart= and ExecStop= are not valid.)
277 For each of the specified commands, the first argument must be
279 either an absolute path to an executable or a simple file name
280 without any slashes. Optionally, this filename may be prefixed with
281 a number of special characters:
282 Table 1. Special executable prefixes
284 ┌───────┬────────────────────────────┐
285 │Prefix │ Effect │
286 ├───────┼────────────────────────────┤
287 │"@" │ If the executable path is │
288 │ │ prefixed with "@", the │
289 │ │ second specified token │
290 │ │ will be passed as │
291 │ │ "argv[0]" to the executed │
292 │ │ process (instead of the │
293 │ │ actual filename), followed │
294 │ │ by the further arguments │
295 │ │ specified. │
296 ├───────┼────────────────────────────┤
297 │"-" │ If the executable path is │
298 │ │ prefixed with "-", an exit │
299 │ │ code of the command │
300 │ │ normally considered a │
301 │ │ failure (i.e. non-zero │
302 │ │ exit status or abnormal │
303 │ │ exit due to signal) is │
304 │ │ recorded, but has no │
305 │ │ further effect and is │
306 │ │ considered equivalent to │
307 │ │ success. │
308 ├───────┼────────────────────────────┤
309 │":" │ If the executable path is │
310 │ │ prefixed with ":", │
311 │ │ environment variable │
312 │ │ substitution (as described │
313 │ │ by the "Command Lines" │
314 │ │ section below) is not │
315 │ │ applied. │
316 ├───────┼────────────────────────────┤
317 │"+" │ If the executable path is │
318 │ │ prefixed with "+" then the │
319 │ │ process is executed with │
320 │ │ full privileges. In this │
321 │ │ mode privilege │
322 │ │ restrictions configured │
323 │ │ with User=, Group=, │
324 │ │ CapabilityBoundingSet= or │
325 │ │ the various file system │
326 │ │ namespacing options (such │
327 │ │ as PrivateDevices=, │
328 │ │ PrivateTmp=) are not │
329 │ │ applied to the invoked │
330 │ │ command line (but still │
331 │ │ affect any other │
332 │ │ ExecStart=, ExecStop=, ... │
333 │ │ lines). │
334 ├───────┼────────────────────────────┤
335 │"!" │ Similar to the "+" │
336 │ │ character discussed above │
337 │ │ this permits invoking │
338 │ │ command lines with │
339 │ │ elevated privileges. │
340 │ │ However, unlike "+" the │
341 │ │ "!" character exclusively │
342 │ │ alters the effect of │
343 │ │ User=, Group= and │
344 │ │ SupplementaryGroups=, i.e. │
345 │ │ only the stanzas that │
346 │ │ affect user and group │
347 │ │ credentials. Note that │
348 │ │ this setting may be │
349 │ │ combined with │
350 │ │ DynamicUser=, in which │
351 │ │ case a dynamic user/group │
352 │ │ pair is allocated before │
353 │ │ the command is invoked, │
354 │ │ but credential changing is │
355 │ │ left to the executed │
356 │ │ process itself. │
357 ├───────┼────────────────────────────┤
358 │"!!" │ This prefix is very │
359 │ │ similar to "!", however it │
360 │ │ only has an effect on │
361 │ │ systems lacking support │
362 │ │ for ambient process │
363 │ │ capabilities, i.e. without │
364 │ │ support for │
365 │ │ AmbientCapabilities=. It's │
366 │ │ intended to be used for │
367 │ │ unit files that take │
368 │ │ benefit of ambient │
369 │ │ capabilities to run │
370 │ │ processes with minimal │
371 │ │ privileges wherever │
372 │ │ possible while remaining │
373 │ │ compatible with systems │
374 │ │ that lack ambient │
375 │ │ capabilities support. Note │
376 │ │ that when "!!" is used, │
377 │ │ and a system lacking │
378 │ │ ambient capability support │
379 │ │ is detected any configured │
380 │ │ SystemCallFilter= and │
381 │ │ CapabilityBoundingSet= │
382 │ │ stanzas are implicitly │
383 │ │ modified, in order to │
384 │ │ permit spawned processes │
385 │ │ to drop credentials and │
386 │ │ capabilities themselves, │
387 │ │ even if this is configured │
388 │ │ to not be allowed. │
389 │ │ Moreover, if this prefix │
390 │ │ is used and a system │
391 │ │ lacking ambient capability │
392 │ │ support is detected │
393 │ │ AmbientCapabilities= will │
394 │ │ be skipped and not be │
395 │ │ applied. On systems │
396 │ │ supporting ambient │
397 │ │ capabilities, "!!" has no │
398 │ │ effect and is redundant. │
399 └───────┴────────────────────────────┘
400 "@", "-", ":", and one of "+"/"!"/"!!" may be used together and
401 they can appear in any order. However, only one of "+", "!", "!!"
402 may be used at a time. Note that these prefixes are also supported
403 for the other command line settings, i.e. ExecStartPre=,
404 ExecStartPost=, ExecReload=, ExecStop= and ExecStopPost=.
405 If more than one command is specified, the commands are invoked
407 sequentially in the order they appear in the unit file. If one of
408 the commands fails (and is not prefixed with "-"), other lines are
409 not executed, and the unit is considered failed.
410 Unless Type=forking is set, the process started via this command
412 line will be considered the main process of the daemon.
413 ExecStartPre=, ExecStartPost=
415 Additional commands that are executed before or after the command
416 in ExecStart=, respectively. Syntax is the same as for ExecStart=,
417 except that multiple command lines are allowed and the commands are
418 executed one after the other, serially.
419 If any of those commands (not prefixed with "-") fail, the rest are
421 not executed and the unit is considered failed.
422 ExecStart= commands are only run after all ExecStartPre= commands
424 that were not prefixed with a "-" exit successfully.
425 ExecStartPost= commands are only run after the commands specified
427 in ExecStart= have been invoked successfully, as determined by
428 Type= (i.e. the process has been started for Type=simple or
429 Type=idle, the last ExecStart= process exited successfully for
430 Type=oneshot, the initial process exited successfully for
431 Type=forking, "READY=1" is sent for Type=notify, or the BusName=
432 has been taken for Type=dbus).
433 Note that ExecStartPre= may not be used to start long-running
435 processes. All processes forked off by processes invoked via
436 ExecStartPre= will be killed before the next service process is
437 run.
438 Note that if any of the commands specified in ExecStartPre=,
440 ExecStart=, or ExecStartPost= fail (and are not prefixed with "-",
441 see above) or time out before the service is fully up, execution
442 continues with commands specified in ExecStopPost=, the commands in
443 ExecStop= are skipped.
444 Note that the execution of ExecStartPost= is taken into account for
446 the purpose of Before=/After= ordering constraints.
447 ExecCondition=
449 Optional commands that are executed before the command(s) in
450 ExecStartPre=. Syntax is the same as for ExecStart=, except that
451 multiple command lines are allowed and the commands are executed
452 one after the other, serially.
453 The behavior is like an ExecStartPre= and condition check hybrid:
455 when an ExecCondition= command exits with exit code 1 through 254
456 (inclusive), the remaining commands are skipped and the unit is not
457 marked as failed. However, if an ExecCondition= command exits with
458 255 or abnormally (e.g. timeout, killed by a signal, etc.), the
459 unit will be considered failed (and remaining commands will be
460 skipped). Exit code of 0 or those matching SuccessExitStatus= will
461 continue execution to the next command(s).
462 The same recommendations about not running long-running processes
464 in ExecStartPre= also applies to ExecCondition=. ExecCondition=
465 will also run the commands in ExecStopPost=, as part of stopping
466 the service, in the case of any non-zero or abnormal exits, like
467 the ones described above.
468 ExecReload=
470 Commands to execute to trigger a configuration reload in the
471 service. This argument takes multiple command lines, following the
472 same scheme as described for ExecStart= above. Use of this setting
473 is optional. Specifier and environment variable substitution is
474 supported here following the same scheme as for ExecStart=.
475 One additional, special environment variable is set: if known,
477 $MAINPID is set to the main process of the daemon, and may be used
478 for command lines like the following:
479 ExecReload=kill -HUP $MAINPID
481 Note however that reloading a daemon by sending a signal (as with
483 the example line above) is usually not a good choice, because this
484 is an asynchronous operation and hence not suitable to order
485 reloads of multiple services against each other. It is strongly
486 recommended to set ExecReload= to a command that not only triggers
487 a configuration reload of the daemon, but also synchronously waits
488 for it to complete. For example, dbus-broker(1) uses the following:
489 ExecReload=busctl call org.freedesktop.DBus \
491 /org/freedesktop/DBus org.freedesktop.DBus \
492 ReloadConfig
493 ExecStop=
495 Commands to execute to stop the service started via ExecStart=.
496 This argument takes multiple command lines, following the same
497 scheme as described for ExecStart= above. Use of this setting is
498 optional. After the commands configured in this option are run, it
499 is implied that the service is stopped, and any processes remaining
500 for it are terminated according to the KillMode= setting (see
501 systemd.kill(5)). If this option is not specified, the process is
502 terminated by sending the signal specified in KillSignal= or
503 RestartKillSignal= when service stop is requested. Specifier and
504 environment variable substitution is supported (including $MAINPID,
505 see above).
506 Note that it is usually not sufficient to specify a command for
508 this setting that only asks the service to terminate (for example,
509 by sending some form of termination signal to it), but does not
510 wait for it to do so. Since the remaining processes of the services
511 are killed according to KillMode= and KillSignal= or
512 RestartKillSignal= as described above immediately after the command
513 exited, this may not result in a clean stop. The specified command
514 should hence be a synchronous operation, not an asynchronous one.
515 Note that the commands specified in ExecStop= are only executed
517 when the service started successfully first. They are not invoked
518 if the service was never started at all, or in case its start-up
519 failed, for example because any of the commands specified in
520 ExecStart=, ExecStartPre= or ExecStartPost= failed (and weren't
521 prefixed with "-", see above) or timed out. Use ExecStopPost= to
522 invoke commands when a service failed to start up correctly and is
523 shut down again. Also note that the stop operation is always
524 performed if the service started successfully, even if the
525 processes in the service terminated on their own or were killed.
526 The stop commands must be prepared to deal with that case.
527 $MAINPID will be unset if systemd knows that the main process
528 exited by the time the stop commands are called.
529 Service restart requests are implemented as stop operations
531 followed by start operations. This means that ExecStop= and
532 ExecStopPost= are executed during a service restart operation.
533 It is recommended to use this setting for commands that communicate
535 with the service requesting clean termination. For post-mortem
536 clean-up steps use ExecStopPost= instead.
537 ExecStopPost=
539 Additional commands that are executed after the service is stopped.
540 This includes cases where the commands configured in ExecStop= were
541 used, where the service does not have any ExecStop= defined, or
542 where the service exited unexpectedly. This argument takes multiple
543 command lines, following the same scheme as described for
544 ExecStart=. Use of these settings is optional. Specifier and
545 environment variable substitution is supported. Note that – unlike
546 ExecStop= – commands specified with this setting are invoked when a
547 service failed to start up correctly and is shut down again.
548 It is recommended to use this setting for clean-up operations that
550 shall be executed even when the service failed to start up
551 correctly. Commands configured with this setting need to be able to
552 operate even if the service failed starting up half-way and left
553 incompletely initialized data around. As the service's processes
554 have been terminated already when the commands specified with this
555 setting are executed they should not attempt to communicate with
556 them.
557 Note that all commands that are configured with this setting are
559 invoked with the result code of the service, as well as the main
560 process' exit code and status, set in the $SERVICE_RESULT,
561 $EXIT_CODE and $EXIT_STATUS environment variables, see
562 systemd.exec(5) for details.
563 Note that the execution of ExecStopPost= is taken into account for
565 the purpose of Before=/After= ordering constraints.
566 RestartSec=
568 Configures the time to sleep before restarting a service (as
569 configured with Restart=). Takes a unit-less value in seconds, or a
570 time span value such as "5min 20s". Defaults to 100ms.
571 TimeoutStartSec=
573 Configures the time to wait for start-up. If a daemon service does
574 not signal start-up completion within the configured time, the
575 service will be considered failed and will be shut down again. The
576 precise action depends on the TimeoutStartFailureMode= option.
577 Takes a unit-less value in seconds, or a time span value such as
578 "5min 20s". Pass "infinity" to disable the timeout logic. Defaults
579 to DefaultTimeoutStartSec= from the manager configuration file,
580 except when Type=oneshot is used, in which case the timeout is
581 disabled by default (see systemd-system.conf(5)).
582 If a service of Type=notify sends "EXTEND_TIMEOUT_USEC=...", this
584 may cause the start time to be extended beyond TimeoutStartSec=.
585 The first receipt of this message must occur before
586 TimeoutStartSec= is exceeded, and once the start time has extended
587 beyond TimeoutStartSec=, the service manager will allow the service
588 to continue to start, provided the service repeats
589 "EXTEND_TIMEOUT_USEC=..." within the interval specified until the
590 service startup status is finished by "READY=1". (see
591 sd_notify(3)).
592 TimeoutStopSec=
594 This option serves two purposes. First, it configures the time to
595 wait for each ExecStop= command. If any of them times out,
596 subsequent ExecStop= commands are skipped and the service will be
597 terminated by SIGTERM. If no ExecStop= commands are specified, the
598 service gets the SIGTERM immediately. This default behavior can be
599 changed by the TimeoutStopFailureMode= option. Second, it
600 configures the time to wait for the service itself to stop. If it
601 doesn't terminate in the specified time, it will be forcibly
602 terminated by SIGKILL (see KillMode= in systemd.kill(5)). Takes a
603 unit-less value in seconds, or a time span value such as "5min
604 20s". Pass "infinity" to disable the timeout logic. Defaults to
605 DefaultTimeoutStopSec= from the manager configuration file (see
606 systemd-system.conf(5)).
607 If a service of Type=notify sends "EXTEND_TIMEOUT_USEC=...", this
609 may cause the stop time to be extended beyond TimeoutStopSec=. The
610 first receipt of this message must occur before TimeoutStopSec= is
611 exceeded, and once the stop time has extended beyond
612 TimeoutStopSec=, the service manager will allow the service to
613 continue to stop, provided the service repeats
614 "EXTEND_TIMEOUT_USEC=..." within the interval specified, or
615 terminates itself (see sd_notify(3)).
616 TimeoutAbortSec=
618 This option configures the time to wait for the service to
619 terminate when it was aborted due to a watchdog timeout (see
620 WatchdogSec=). If the service has a short TimeoutStopSec= this
621 option can be used to give the system more time to write a core
622 dump of the service. Upon expiration the service will be forcibly
623 terminated by SIGKILL (see KillMode= in systemd.kill(5)). The core
624 file will be truncated in this case. Use TimeoutAbortSec= to set a
625 sensible timeout for the core dumping per service that is large
626 enough to write all expected data while also being short enough to
627 handle the service failure in due time.
628 Takes a unit-less value in seconds, or a time span value such as
630 "5min 20s". Pass an empty value to skip the dedicated watchdog
631 abort timeout handling and fall back TimeoutStopSec=. Pass
632 "infinity" to disable the timeout logic. Defaults to
633 DefaultTimeoutAbortSec= from the manager configuration file (see
634 systemd-system.conf(5)).
635 If a service of Type=notify handles SIGABRT itself (instead of
637 relying on the kernel to write a core dump) it can send
638 "EXTEND_TIMEOUT_USEC=..." to extended the abort time beyond
639 TimeoutAbortSec=. The first receipt of this message must occur
640 before TimeoutAbortSec= is exceeded, and once the abort time has
641 extended beyond TimeoutAbortSec=, the service manager will allow
642 the service to continue to abort, provided the service repeats
643 "EXTEND_TIMEOUT_USEC=..." within the interval specified, or
644 terminates itself (see sd_notify(3)).
645 TimeoutSec=
647 A shorthand for configuring both TimeoutStartSec= and
648 TimeoutStopSec= to the specified value.
649 TimeoutStartFailureMode=, TimeoutStopFailureMode=
651 These options configure the action that is taken in case a daemon
652 service does not signal start-up within its configured
653 TimeoutStartSec=, respectively if it does not stop within
654 TimeoutStopSec=. Takes one of terminate, abort and kill. Both
655 options default to terminate.
656 If terminate is set the service will be gracefully terminated by
658 sending the signal specified in KillSignal= (defaults to SIGTERM,
659 see systemd.kill(5)). If the service does not terminate the
660 FinalKillSignal= is sent after TimeoutStopSec=. If abort is set,
661 WatchdogSignal= is sent instead and TimeoutAbortSec= applies before
662 sending FinalKillSignal=. This setting may be used to analyze
663 services that fail to start-up or shut-down intermittently. By
664 using kill the service is immediately terminated by sending
665 FinalKillSignal= without any further timeout. This setting can be
666 used to expedite the shutdown of failing services.
667 RuntimeMaxSec=
669 Configures a maximum time for the service to run. If this is used
670 and the service has been active for longer than the specified time
671 it is terminated and put into a failure state. Note that this
672 setting does not have any effect on Type=oneshot services, as they
673 terminate immediately after activation completed. Pass "infinity"
674 (the default) to configure no runtime limit.
675 If a service of Type=notify sends "EXTEND_TIMEOUT_USEC=...", this
677 may cause the runtime to be extended beyond RuntimeMaxSec=. The
678 first receipt of this message must occur before RuntimeMaxSec= is
679 exceeded, and once the runtime has extended beyond RuntimeMaxSec=,
680 the service manager will allow the service to continue to run,
681 provided the service repeats "EXTEND_TIMEOUT_USEC=..." within the
682 interval specified until the service shutdown is achieved by
683 "STOPPING=1" (or termination). (see sd_notify(3)).
684 WatchdogSec=
686 Configures the watchdog timeout for a service. The watchdog is
687 activated when the start-up is completed. The service must call
688 sd_notify(3) regularly with "WATCHDOG=1" (i.e. the "keep-alive
689 ping"). If the time between two such calls is larger than the
690 configured time, then the service is placed in a failed state and
691 it will be terminated with SIGABRT (or the signal specified by
692 WatchdogSignal=). By setting Restart= to on-failure, on-watchdog,
693 on-abnormal or always, the service will be automatically restarted.
694 The time configured here will be passed to the executed service
695 process in the WATCHDOG_USEC= environment variable. This allows
696 daemons to automatically enable the keep-alive pinging logic if
697 watchdog support is enabled for the service. If this option is
698 used, NotifyAccess= (see below) should be set to open access to the
699 notification socket provided by systemd. If NotifyAccess= is not
700 set, it will be implicitly set to main. Defaults to 0, which
701 disables this feature. The service can check whether the service
702 manager expects watchdog keep-alive notifications. See
703 sd_watchdog_enabled(3) for details. sd_event_set_watchdog(3) may
704 be used to enable automatic watchdog notification support.
705 Restart=
707 Configures whether the service shall be restarted when the service
708 process exits, is killed, or a timeout is reached. The service
709 process may be the main service process, but it may also be one of
710 the processes specified with ExecStartPre=, ExecStartPost=,
711 ExecStop=, ExecStopPost=, or ExecReload=. When the death of the
712 process is a result of systemd operation (e.g. service stop or
713 restart), the service will not be restarted. Timeouts include
714 missing the watchdog "keep-alive ping" deadline and a service
715 start, reload, and stop operation timeouts.
716 Takes one of no, on-success, on-failure, on-abnormal, on-watchdog,
718 on-abort, or always. If set to no (the default), the service will
719 not be restarted. If set to on-success, it will be restarted only
720 when the service process exits cleanly. In this context, a clean
721 exit means any of the following:
722 • exit code of 0;
724 • for types other than Type=oneshot, one of the signals SIGHUP,
726 SIGINT, SIGTERM, or SIGPIPE;
727 • exit statuses and signals specified in SuccessExitStatus=.
729 If set to on-failure, the service will be restarted when the
731 process exits with a non-zero exit code, is terminated by a signal
732 (including on core dump, but excluding the aforementioned four
733 signals), when an operation (such as service reload) times out, and
734 when the configured watchdog timeout is triggered. If set to
735 on-abnormal, the service will be restarted when the process is
736 terminated by a signal (including on core dump, excluding the
737 aforementioned four signals), when an operation times out, or when
738 the watchdog timeout is triggered. If set to on-abort, the service
739 will be restarted only if the service process exits due to an
740 uncaught signal not specified as a clean exit status. If set to
741 on-watchdog, the service will be restarted only if the watchdog
742 timeout for the service expires. If set to always, the service will
743 be restarted regardless of whether it exited cleanly or not, got
744 terminated abnormally by a signal, or hit a timeout.
745 Table 2. Exit causes and the effect of the Restart= settings
747 ┌──────────────┬────┬────────┬────────────┬────────────┬─────────────┬──────────┬─────────────┐
748 │Restart │ no │ always │ on-success │ on-failure │ on-abnormal │ on-abort │ on-watchdog │
749 │settings/Exit │ │ │ │ │ │ │ │
750 │causes │ │ │ │ │ │ │ │
751 ├──────────────┼────┼────────┼────────────┼────────────┼─────────────┼──────────┼─────────────┤
752 │Clean exit │ │ X │ X │ │ │ │ │
753 │code or │ │ │ │ │ │ │ │
754 │signal │ │ │ │ │ │ │ │
755 ├──────────────┼────┼────────┼────────────┼────────────┼─────────────┼──────────┼─────────────┤
756 │Unclean exit │ │ X │ │ X │ │ │ │
757 │code │ │ │ │ │ │ │ │
758 ├──────────────┼────┼────────┼────────────┼────────────┼─────────────┼──────────┼─────────────┤
759 │Unclean │ │ X │ │ X │ X │ X │ │
760 │signal │ │ │ │ │ │ │ │
761 ├──────────────┼────┼────────┼────────────┼────────────┼─────────────┼──────────┼─────────────┤
762 │Timeout │ │ X │ │ X │ X │ │ │
763 ├──────────────┼────┼────────┼────────────┼────────────┼─────────────┼──────────┼─────────────┤
764 │Watchdog │ │ X │ │ X │ X │ │ X │
765 └──────────────┴────┴────────┴────────────┴────────────┴─────────────┴──────────┴─────────────┘
766 As exceptions to the setting above, the service will not be
767 restarted if the exit code or signal is specified in
768 RestartPreventExitStatus= (see below) or the service is stopped
769 with systemctl stop or an equivalent operation. Also, the services
770 will always be restarted if the exit code or signal is specified in
771 RestartForceExitStatus= (see below).
772 Note that service restart is subject to unit start rate limiting
774 configured with StartLimitIntervalSec= and StartLimitBurst=, see
775 systemd.unit(5) for details. A restarted service enters the failed
776 state only after the start limits are reached.
777 Setting this to on-failure is the recommended choice for
779 long-running services, in order to increase reliability by
780 attempting automatic recovery from errors. For services that shall
781 be able to terminate on their own choice (and avoid immediate
782 restarting), on-abnormal is an alternative choice.
783 SuccessExitStatus=
785 Takes a list of exit status definitions that, when returned by the
786 main service process, will be considered successful termination, in
787 addition to the normal successful exit status 0 and, except for
788 Type=oneshot, the signals SIGHUP, SIGINT, SIGTERM, and SIGPIPE.
789 Exit status definitions can be numeric termination statuses,
790 termination status names, or termination signal names, separated by
791 spaces. See the Process Exit Codes section in systemd.exec(5) for a
792 list of termination status names (for this setting only the part
793 without the "EXIT_" or "EX_" prefix should be used). See signal(7)
794 for a list of signal names.
795 Note that this setting does not change the mapping between numeric
797 exit statuses and their names, i.e. regardless how this setting is
798 used 0 will still be mapped to "SUCCESS" (and thus typically shown
799 as "0/SUCCESS" in tool outputs) and 1 to "FAILURE" (and thus
800 typically shown as "1/FAILURE"), and so on. It only controls what
801 happens as effect of these exit statuses, and how it propagates to
802 the state of the service as a whole.
803 This option may appear more than once, in which case the list of
805 successful exit statuses is merged. If the empty string is assigned
806 to this option, the list is reset, all prior assignments of this
807 option will have no effect.
808 Example 1. A service with the SuccessExitStatus= setting
810 SuccessExitStatus=TEMPFAIL 250 SIGKILL
812 Exit status 75 (TEMPFAIL), 250, and the termination signal SIGKILL
814 are considered clean service terminations.
815 Note: systemd-analyze exit-status may be used to list exit statuses
817 and translate between numerical status values and names.
818 RestartPreventExitStatus=
820 Takes a list of exit status definitions that, when returned by the
821 main service process, will prevent automatic service restarts,
822 regardless of the restart setting configured with Restart=. Exit
823 status definitions can either be numeric exit codes or termination
824 signal names, and are separated by spaces. Defaults to the empty
825 list, so that, by default, no exit status is excluded from the
826 configured restart logic. For example:
827 RestartPreventExitStatus=1 6 SIGABRT
829 ensures that exit codes 1 and 6 and the termination signal SIGABRT
831 will not result in automatic service restarting. This option may
832 appear more than once, in which case the list of restart-preventing
833 statuses is merged. If the empty string is assigned to this option,
834 the list is reset and all prior assignments of this option will
835 have no effect.
836 Note that this setting has no effect on processes configured via
838 ExecStartPre=, ExecStartPost=, ExecStop=, ExecStopPost= or
839 ExecReload=, but only on the main service process, i.e. either the
840 one invoked by ExecStart= or (depending on Type=, PIDFile=, ...)
841 the otherwise configured main process.
842 RestartForceExitStatus=
844 Takes a list of exit status definitions that, when returned by the
845 main service process, will force automatic service restarts,
846 regardless of the restart setting configured with Restart=. The
847 argument format is similar to RestartPreventExitStatus=.
848 RootDirectoryStartOnly=
850 Takes a boolean argument. If true, the root directory, as
851 configured with the RootDirectory= option (see systemd.exec(5) for
852 more information), is only applied to the process started with
853 ExecStart=, and not to the various other ExecStartPre=,
854 ExecStartPost=, ExecReload=, ExecStop=, and ExecStopPost= commands.
855 If false, the setting is applied to all configured commands the
856 same way. Defaults to false.
857 NonBlocking=
859 Set the O_NONBLOCK flag for all file descriptors passed via
860 socket-based activation. If true, all file descriptors >= 3 (i.e.
861 all except stdin, stdout, stderr), excluding those passed in via
862 the file descriptor storage logic (see FileDescriptorStoreMax= for
863 details), will have the O_NONBLOCK flag set and hence are in
864 non-blocking mode. This option is only useful in conjunction with a
865 socket unit, as described in systemd.socket(5) and has no effect on
866 file descriptors which were previously saved in the file-descriptor
867 store for example. Defaults to false.
868 NotifyAccess=
870 Controls access to the service status notification socket, as
871 accessible via the sd_notify(3) call. Takes one of none (the
872 default), main, exec or all. If none, no daemon status updates are
873 accepted from the service processes, all status update messages are
874 ignored. If main, only service updates sent from the main process
875 of the service are accepted. If exec, only service updates sent
876 from any of the main or control processes originating from one of
877 the Exec*= commands are accepted. If all, all services updates from
878 all members of the service's control group are accepted. This
879 option should be set to open access to the notification socket when
880 using Type=notify or WatchdogSec= (see above). If those options are
881 used but NotifyAccess= is not configured, it will be implicitly set
882 to main.
883 Note that sd_notify() notifications may be attributed to units
885 correctly only if either the sending process is still around at the
886 time PID 1 processes the message, or if the sending process is
887 explicitly runtime-tracked by the service manager. The latter is
888 the case if the service manager originally forked off the process,
889 i.e. on all processes that match main or exec. Conversely, if an
890 auxiliary process of the unit sends an sd_notify() message and
891 immediately exits, the service manager might not be able to
892 properly attribute the message to the unit, and thus will ignore
893 it, even if NotifyAccess=all is set for it.
894 Hence, to eliminate all race conditions involving lookup of the
896 client's unit and attribution of notifications to units correctly,
897 sd_notify_barrier() may be used. This call acts as a
898 synchronization point and ensures all notifications sent before
899 this call have been picked up by the service manager when it
900 returns successfully. Use of sd_notify_barrier() is needed for
901 clients which are not invoked by the service manager, otherwise
902 this synchronization mechanism is unnecessary for attribution of
903 notifications to the unit.
904 Sockets=
906 Specifies the name of the socket units this service shall inherit
907 socket file descriptors from when the service is started. Normally,
908 it should not be necessary to use this setting, as all socket file
909 descriptors whose unit shares the same name as the service (subject
910 to the different unit name suffix of course) are passed to the
911 spawned process.
912 Note that the same socket file descriptors may be passed to
914 multiple processes simultaneously. Also note that a different
915 service may be activated on incoming socket traffic than the one
916 which is ultimately configured to inherit the socket file
917 descriptors. Or, in other words: the Service= setting of .socket
918 units does not have to match the inverse of the Sockets= setting of
919 the .service it refers to.
920 This option may appear more than once, in which case the list of
922 socket units is merged. Note that once set, clearing the list of
923 sockets again (for example, by assigning the empty string to this
924 option) is not supported.
925 FileDescriptorStoreMax=
927 Configure how many file descriptors may be stored in the service
928 manager for the service using sd_pid_notify_with_fds(3)'s
929 "FDSTORE=1" messages. This is useful for implementing services that
930 can restart after an explicit request or a crash without losing
931 state. Any open sockets and other file descriptors which should not
932 be closed during the restart may be stored this way. Application
933 state can either be serialized to a file in /run/, or better,
934 stored in a memfd_create(2) memory file descriptor. Defaults to 0,
935 i.e. no file descriptors may be stored in the service manager. All
936 file descriptors passed to the service manager from a specific
937 service are passed back to the service's main process on the next
938 service restart (see sd_listen_fds(3) for details about the precise
939 protocol used and the order in which the file descriptors are
940 passed). Any file descriptors passed to the service manager are
941 automatically closed when POLLHUP or POLLERR is seen on them, or
942 when the service is fully stopped and no job is queued or being
943 executed for it. If this option is used, NotifyAccess= (see above)
944 should be set to open access to the notification socket provided by
945 systemd. If NotifyAccess= is not set, it will be implicitly set to
946 main.
947 USBFunctionDescriptors=
949 Configure the location of a file containing USB FunctionFS[2]
950 descriptors, for implementation of USB gadget functions. This is
951 used only in conjunction with a socket unit with ListenUSBFunction=
952 configured. The contents of this file are written to the ep0 file
953 after it is opened.
954 USBFunctionStrings=
956 Configure the location of a file containing USB FunctionFS strings.
957 Behavior is similar to USBFunctionDescriptors= above.
958 OOMPolicy=
960 Configure the Out-Of-Memory (OOM) killer policy. On Linux, when
961 memory becomes scarce the kernel might decide to kill a running
962 process in order to free up memory and reduce memory pressure. This
963 setting takes one of continue, stop or kill. If set to continue and
964 a process of the service is killed by the kernel's OOM killer this
965 is logged but the service continues running. If set to stop the
966 event is logged but the service is terminated cleanly by the
967 service manager. If set to kill and one of the service's processes
968 is killed by the OOM killer the kernel is instructed to kill all
969 remaining processes of the service, too. Defaults to the setting
970 DefaultOOMPolicy= in systemd-system.conf(5) is set to, except for
971 services where Delegate= is turned on, where it defaults to
972 continue.
973 Use the OOMScoreAdjust= setting to configure whether processes of
975 the unit shall be considered preferred or less preferred candidates
976 for process termination by the Linux OOM killer logic. See
977 systemd.exec(5) for details.
978 Check systemd.exec(5) and systemd.kill(5) for more settings.
980
982 This section describes command line parsing and variable and specifier
983 substitutions for ExecStart=, ExecStartPre=, ExecStartPost=,
984 ExecReload=, ExecStop=, and ExecStopPost= options.
985 Multiple command lines may be concatenated in a single directive by
987 separating them with semicolons (these semicolons must be passed as
988 separate words). Lone semicolons may be escaped as "\;".
989 Each command line is unquoted using the rules described in "Quoting"
991 section in systemd.syntax(7). The first item becomes the command to
992 execute, and the subsequent items the arguments.
993 This syntax is inspired by shell syntax, but only the meta-characters
995 and expansions described in the following paragraphs are understood,
996 and the expansion of variables is different. Specifically, redirection
997 using "<", "<<", ">", and ">>", pipes using "|", running programs in
998 the background using "&", and other elements of shell syntax are not
999 supported.
1000 The command to execute may contain spaces, but control characters are
1002 not allowed.
1003 The command line accepts "%" specifiers as described in
1005 systemd.unit(5).
1006 Basic environment variable substitution is supported. Use "${FOO}" as
1008 part of a word, or as a word of its own, on the command line, in which
1009 case it will be erased and replaced by the exact value of the
1010 environment variable (if any) including all whitespace it contains,
1011 always resulting in exactly a single argument. Use "$FOO" as a separate
1012 word on the command line, in which case it will be replaced by the
1013 value of the environment variable split at whitespace, resulting in
1014 zero or more arguments. For this type of expansion, quotes are
1015 respected when splitting into words, and afterwards removed.
1016 If the command is not a full (absolute) path, it will be resolved to a
1018 full path using a fixed search path determined at compilation time.
1019 Searched directories include /usr/local/bin/, /usr/bin/, /bin/ on
1020 systems using split /usr/bin/ and /bin/ directories, and their sbin/
1021 counterparts on systems using split bin/ and sbin/. It is thus safe to
1022 use just the executable name in case of executables located in any of
1023 the "standard" directories, and an absolute path must be used in other
1024 cases. Using an absolute path is recommended to avoid ambiguity. Hint:
1025 this search path may be queried using systemd-path
1026 search-binaries-default.
1027 Example:
1029 Environment="ONE=one" 'TWO=two two'
1031 ExecStart=echo $ONE $TWO ${TWO}
1032 This will execute /bin/echo with four arguments: "one", "two", "two",
1034 and "two two".
1035 Example:
1037 Environment=ONE='one' "TWO='two two' too" THREE=
1039 ExecStart=/bin/echo ${ONE} ${TWO} ${THREE}
1040 ExecStart=/bin/echo $ONE $TWO $THREE
1041 This results in /bin/echo being called twice, the first time with
1043 arguments "'one'", "'two two' too", "", and the second time with
1044 arguments "one", "two two", "too".
1045 To pass a literal dollar sign, use "$$". Variables whose value is not
1047 known at expansion time are treated as empty strings. Note that the
1048 first argument (i.e. the program to execute) may not be a variable.
1049 Variables to be used in this fashion may be defined through
1051 Environment= and EnvironmentFile=. In addition, variables listed in the
1052 section "Environment variables in spawned processes" in
1053 systemd.exec(5), which are considered "static configuration", may be
1054 used (this includes e.g. $USER, but not $TERM).
1055 Note that shell command lines are not directly supported. If shell
1057 command lines are to be used, they need to be passed explicitly to a
1058 shell implementation of some kind. Example:
1059 ExecStart=sh -c 'dmesg | tac'
1061 Example:
1063 ExecStart=echo one ; echo "two two"
1065 This will execute echo two times, each time with one argument: "one"
1067 and "two two", respectively. Because two commands are specified,
1068 Type=oneshot must be used.
1069 Example:
1071 ExecStart=echo / >/dev/null & \; \
1073 ls
1074 This will execute echo with five arguments: "/", ">/dev/null", "&",
1076 ";", and "ls".
1077
1079 Example 2. Simple service
1080 The following unit file creates a service that will execute
1082 /usr/sbin/foo-daemon. Since no Type= is specified, the default
1083 Type=simple will be assumed. systemd will assume the unit to be started
1084 immediately after the program has begun executing.
1085 [Unit]
1087 Description=Foo
1088 [Service]
1090 ExecStart=/usr/sbin/foo-daemon
1091 [Install]
1093 WantedBy=multi-user.target
1094 Note that systemd assumes here that the process started by systemd will
1096 continue running until the service terminates. If the program
1097 daemonizes itself (i.e. forks), please use Type=forking instead.
1098 Since no ExecStop= was specified, systemd will send SIGTERM to all
1100 processes started from this service, and after a timeout also SIGKILL.
1101 This behavior can be modified, see systemd.kill(5) for details.
1102 Note that this unit type does not include any type of notification when
1104 a service has completed initialization. For this, you should use other
1105 unit types, such as Type=notify if the service understands systemd's
1106 notification protocol, Type=forking if the service can background
1107 itself or Type=dbus if the unit acquires a DBus name once
1108 initialization is complete. See below.
1109 Example 3. Oneshot service
1111 Sometimes, units should just execute an action without keeping active
1113 processes, such as a filesystem check or a cleanup action on boot. For
1114 this, Type=oneshot exists. Units of this type will wait until the
1115 process specified terminates and then fall back to being inactive. The
1116 following unit will perform a cleanup action:
1117 [Unit]
1119 Description=Cleanup old Foo data
1120 [Service]
1122 Type=oneshot
1123 ExecStart=/usr/sbin/foo-cleanup
1124 [Install]
1126 WantedBy=multi-user.target
1127 Note that systemd will consider the unit to be in the state "starting"
1129 until the program has terminated, so ordered dependencies will wait for
1130 the program to finish before starting themselves. The unit will revert
1131 to the "inactive" state after the execution is done, never reaching the
1132 "active" state. That means another request to start the unit will
1133 perform the action again.
1134 Type=oneshot are the only service units that may have more than one
1136 ExecStart= specified. For units with multiple commands (Type=oneshot),
1137 all commands will be run again.
1138 For Type=oneshot, Restart=always and Restart=on-success are not
1140 allowed.
1141 Example 4. Stoppable oneshot service
1143 Similarly to the oneshot services, there are sometimes units that need
1145 to execute a program to set up something and then execute another to
1146 shut it down, but no process remains active while they are considered
1147 "started". Network configuration can sometimes fall into this category.
1148 Another use case is if a oneshot service shall not be executed each
1149 time when they are pulled in as a dependency, but only the first time.
1150 For this, systemd knows the setting RemainAfterExit=yes, which causes
1152 systemd to consider the unit to be active if the start action exited
1153 successfully. This directive can be used with all types, but is most
1154 useful with Type=oneshot and Type=simple. With Type=oneshot, systemd
1155 waits until the start action has completed before it considers the unit
1156 to be active, so dependencies start only after the start action has
1157 succeeded. With Type=simple, dependencies will start immediately after
1158 the start action has been dispatched. The following unit provides an
1159 example for a simple static firewall.
1160 [Unit]
1162 Description=Simple firewall
1163 [Service]
1165 Type=oneshot
1166 RemainAfterExit=yes
1167 ExecStart=/usr/local/sbin/simple-firewall-start
1168 ExecStop=/usr/local/sbin/simple-firewall-stop
1169 [Install]
1171 WantedBy=multi-user.target
1172 Since the unit is considered to be running after the start action has
1174 exited, invoking systemctl start on that unit again will cause no
1175 action to be taken.
1176 Example 5. Traditional forking services
1178 Many traditional daemons/services background (i.e. fork, daemonize)
1180 themselves when starting. Set Type=forking in the service's unit file
1181 to support this mode of operation. systemd will consider the service to
1182 be in the process of initialization while the original program is still
1183 running. Once it exits successfully and at least a process remains (and
1184 RemainAfterExit=no), the service is considered started.
1185 Often, a traditional daemon only consists of one process. Therefore, if
1187 only one process is left after the original process terminates, systemd
1188 will consider that process the main process of the service. In that
1189 case, the $MAINPID variable will be available in ExecReload=,
1190 ExecStop=, etc.
1191 In case more than one process remains, systemd will be unable to
1193 determine the main process, so it will not assume there is one. In that
1194 case, $MAINPID will not expand to anything. However, if the process
1195 decides to write a traditional PID file, systemd will be able to read
1196 the main PID from there. Please set PIDFile= accordingly. Note that the
1197 daemon should write that file before finishing with its initialization.
1198 Otherwise, systemd might try to read the file before it exists.
1199 The following example shows a simple daemon that forks and just starts
1201 one process in the background:
1202 [Unit]
1204 Description=Some simple daemon
1205 [Service]
1207 Type=forking
1208 ExecStart=/usr/sbin/my-simple-daemon -d
1209 [Install]
1211 WantedBy=multi-user.target
1212 Please see systemd.kill(5) for details on how you can influence the way
1214 systemd terminates the service.
1215 Example 6. DBus services
1217 For services that acquire a name on the DBus system bus, use Type=dbus
1219 and set BusName= accordingly. The service should not fork (daemonize).
1220 systemd will consider the service to be initialized once the name has
1221 been acquired on the system bus. The following example shows a typical
1222 DBus service:
1223 [Unit]
1225 Description=Simple DBus service
1226 [Service]
1228 Type=dbus
1229 BusName=org.example.simple-dbus-service
1230 ExecStart=/usr/sbin/simple-dbus-service
1231 [Install]
1233 WantedBy=multi-user.target
1234 For bus-activatable services, do not include a [Install] section in the
1236 systemd service file, but use the SystemdService= option in the
1237 corresponding DBus service file, for example
1238 (/usr/share/dbus-1/system-services/org.example.simple-dbus-service.service):
1239 [D-BUS Service]
1241 Name=org.example.simple-dbus-service
1242 Exec=/usr/sbin/simple-dbus-service
1243 User=root
1244 SystemdService=simple-dbus-service.service
1245 Please see systemd.kill(5) for details on how you can influence the way
1247 systemd terminates the service.
1248 Example 7. Services that notify systemd about their initialization
1250 Type=simple services are really easy to write, but have the major
1252 disadvantage of systemd not being able to tell when initialization of
1253 the given service is complete. For this reason, systemd supports a
1254 simple notification protocol that allows daemons to make systemd aware
1255 that they are done initializing. Use Type=notify for this. A typical
1256 service file for such a daemon would look like this:
1257 [Unit]
1259 Description=Simple notifying service
1260 [Service]
1262 Type=notify
1263 ExecStart=/usr/sbin/simple-notifying-service
1264 [Install]
1266 WantedBy=multi-user.target
1267 Note that the daemon has to support systemd's notification protocol,
1269 else systemd will think the service has not started yet and kill it
1270 after a timeout. For an example of how to update daemons to support
1271 this protocol transparently, take a look at sd_notify(3). systemd will
1272 consider the unit to be in the 'starting' state until a readiness
1273 notification has arrived.
1274 Please see systemd.kill(5) for details on how you can influence the way
1276 systemd terminates the service.
1277
1279 systemd(1), systemctl(1), systemd-system.conf(5), systemd.unit(5),
1280 systemd.exec(5), systemd.resource-control(5), systemd.kill(5),
1281 systemd.directives(7), systemd-run(1)
1282
1284 1. Incompatibilities with SysV
1285 https://www.freedesktop.org/wiki/Software/systemd/Incompatibilities
1286 2. USB FunctionFS
1288 https://www.kernel.org/doc/Documentation/usb/functionfs.txt
1289systemd 249 SYSTEMD.SERVICE(5)