1 <?xml version='1.0'?> <!--*-nxml-*-->
2 <?xml-stylesheet type="text/xsl" href="http://docbook.sourceforge.net/release/xsl/current/xhtml/docbook.xsl"?>
3 <!DOCTYPE refentry PUBLIC "-//OASIS//DTD DocBook XML V4.2//EN"
4 "http://www.oasis-open.org/docbook/xml/4.2/docbookx.dtd">
7 This file is part of systemd.
9 Copyright 2010 Lennart Poettering
11 systemd is free software; you can redistribute it and/or modify it
12 under the terms of the GNU Lesser General Public License as published by
13 the Free Software Foundation; either version 2.1 of the License, or
14 (at your option) any later version.
16 systemd is distributed in the hope that it will be useful, but
17 WITHOUT ANY WARRANTY; without even the implied warranty of
18 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
19 Lesser General Public License for more details.
21 You should have received a copy of the GNU Lesser General Public License
22 along with systemd; If not, see <http://www.gnu.org/licenses/>.
25 <refentry id="systemd.service">
27 <title>systemd.service</title>
28 <productname>systemd</productname>
32 <contrib>Developer</contrib>
33 <firstname>Lennart</firstname>
34 <surname>Poettering</surname>
35 <email>lennart@poettering.net</email>
41 <refentrytitle>systemd.service</refentrytitle>
42 <manvolnum>5</manvolnum>
46 <refname>systemd.service</refname>
47 <refpurpose>Service unit configuration</refpurpose>
51 <para><filename><replaceable>service</replaceable>.service</filename></para>
55 <title>Description</title>
57 <para>A unit configuration file whose name ends in
58 <filename>.service</filename> encodes information
59 about a process controlled and supervised by
62 <para>This man page lists the configuration options
63 specific to this unit type. See
64 <citerefentry><refentrytitle>systemd.unit</refentrytitle><manvolnum>5</manvolnum></citerefentry>
65 for the common options of all unit configuration
66 files. The common configuration items are configured
67 in the generic <literal>[Unit]</literal> and
68 <literal>[Install]</literal> sections. The service
69 specific configuration options are configured in the
70 <literal>[Service]</literal> section.</para>
72 <para>Additional options are listed in
73 <citerefentry><refentrytitle>systemd.exec</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
74 which define the execution environment the commands
75 are executed in, and in
76 <citerefentry><refentrytitle>systemd.kill</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
77 which define the way the processes of the service are
79 <citerefentry><refentrytitle>systemd.cgroup</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
80 which configure control group settings for the
81 processes of the service.</para>
83 <para>Unless <varname>DefaultDependencies=</varname>
84 is set to <option>false</option>, service units will
85 implicitly have dependencies of type
86 <varname>Requires=</varname> and
87 <varname>After=</varname> on
88 <filename>basic.target</filename> as well as
89 dependencies of type <varname>Conflicts=</varname> and
90 <varname>Before=</varname> on
91 <filename>shutdown.target</filename>. These ensure
92 that normal service units pull in basic system
93 initialization, and are terminated cleanly prior to
94 system shutdown. Only services involved with early
95 boot or late system shutdown should disable this
98 <para>If a service is requested under a certain name
99 but no unit configuration file is found, systemd looks
100 for a SysV init script by the same name (with the
101 <filename>.service</filename> suffix removed) and
102 dynamically creates a service unit from that
103 script. This is useful for compatibility with
104 SysV. Note that this compatibility is quite
105 comprehensive but not 100%. For details about the
106 incompatibilities see the <ulink
107 url="http://www.freedesktop.org/wiki/Software/systemd/Incompatibilities">Incompatibilities
108 with SysV</ulink> document.
113 <title>Options</title>
115 <para>Service files must include a
116 <literal>[Service]</literal> section, which carries
117 information about the service and the process it
118 supervises. A number of options that may be used in
119 this section are shared with other unit types. These
120 options are documented in
121 <citerefentry><refentrytitle>systemd.exec</refentrytitle><manvolnum>5</manvolnum></citerefentry>
123 <citerefentry><refentrytitle>systemd.kill</refentrytitle><manvolnum>5</manvolnum></citerefentry>. The
124 options specific to the <literal>[Service]</literal>
125 section of service units are the following:</para>
127 <variablelist class='unit-directives'>
129 <term><varname>Type=</varname></term>
131 <listitem><para>Configures the process
132 start-up type for this service
133 unit. One of <option>simple</option>,
134 <option>forking</option>,
135 <option>oneshot</option>,
136 <option>dbus</option>,
137 <option>notify</option> or
138 <option>idle</option>.</para>
141 <option>simple</option> (the default
142 value if <varname>BusName=</varname>
143 is not specified), it is expected that
144 the process configured with
145 <varname>ExecStart=</varname> is the
146 main process of the service. In this
147 mode, if the process offers
148 functionality to other processes on
149 the system, its communication channels
150 should be installed before the daemon
151 is started up (e.g. sockets set up by
152 systemd, via socket activation), as
153 systemd will immediately proceed
154 starting follow-up units.</para>
157 <option>forking</option>, it is
158 expected that the process configured
159 with <varname>ExecStart=</varname>
160 will call <function>fork()</function>
161 as part of its start-up. The parent process is
162 expected to exit when start-up is
163 complete and all communication
164 channels are set up. The child continues
165 to run as the main daemon
166 process. This is the behavior of
167 traditional UNIX daemons. If this
168 setting is used, it is recommended to
170 <varname>PIDFile=</varname> option, so
171 that systemd can identify the main
172 process of the daemon. systemd will
173 proceed starting follow-up units as
174 soon as the parent process
178 <option>oneshot</option> is similar
179 to <option>simple</option>, however
180 it is expected that the process has to
181 exit before systemd starts follow-up
182 units. <varname>RemainAfterExit=</varname>
183 is particularly useful for this type
187 <option>dbus</option> is similar to
188 <option>simple</option>, however it is
189 expected that the daemon acquires a
190 name on the D-Bus bus, as configured
192 <varname>BusName=</varname>. systemd
193 will proceed starting follow-up units
194 after the D-Bus bus name has been
195 acquired. Service units with this
196 option configured implicitly gain
198 <filename>dbus.socket</filename>
199 unit. This type is the default if
200 <varname>BusName=</varname> is
204 <option>notify</option> is similar to
205 <option>simple</option>, however it is
206 expected that the daemon sends a
207 notification message via
208 <citerefentry><refentrytitle>sd_notify</refentrytitle><manvolnum>3</manvolnum></citerefentry>
209 or an equivalent call when it finished
210 starting up. systemd will proceed
211 starting follow-up units after this
212 notification message has been sent. If
214 <varname>NotifyAccess=</varname> (see
215 below) should be set to open access to
216 the notification socket provided by
218 <varname>NotifyAccess=</varname> is
219 not set, it will be implicitly set to
220 <option>main</option>. Note that
222 <varname>Type=</varname><option>notify</option>
223 will not work if used in combination with
224 <varname>PrivateNetwork=</varname><option>yes</option>.</para>
227 <option>idle</option> is very similar
228 to <option>simple</option>, however
229 actual execution of the service
230 binary is delayed until all jobs are
231 dispatched. This may be used to avoid
232 interleaving of output of shell
233 services with the status output on the
239 <term><varname>RemainAfterExit=</varname></term>
241 <listitem><para>Takes a boolean value
242 that specifies whether the service
243 shall be considered active even when
244 all its processes exited. Defaults to
245 <option>no</option>.</para>
250 <term><varname>GuessMainPID=</varname></term>
252 <listitem><para>Takes a boolean value
253 that specifies whether systemd should
254 try to guess the main PID of a service
255 if it cannot be determined
256 reliably. This option is ignored
257 unless <option>Type=forking</option>
258 is set and <option>PIDFile=</option>
259 is unset because for the other types
260 or with an explicitly configured PID
261 file the main PID is always known. The
262 guessing algorithm might come to
263 incorrect conclusions if a daemon
264 consists of more than one process. If
265 the main PID cannot be determined,
266 failure detection and automatic
267 restarting of a service will not work
268 reliably. Defaults to
269 <option>yes</option>.</para>
274 <term><varname>PIDFile=</varname></term>
276 <listitem><para>Takes an absolute file
277 name pointing to the PID file of this
278 daemon. Use of this option is
279 recommended for services where
280 <varname>Type=</varname> is set to
281 <option>forking</option>. systemd will
282 read the PID of the main process of
283 the daemon after start-up of the
284 service. systemd will not write to the
285 file configured here.</para>
290 <term><varname>BusName=</varname></term>
292 <listitem><para>Takes a D-Bus bus
293 name, that this service is reachable
294 as. This option is mandatory for
296 <varname>Type=</varname> is set to
297 <option>dbus</option>, but its use
298 is otherwise recommended as well if
299 the process takes a name on the D-Bus
305 <term><varname>ExecStart=</varname></term>
306 <listitem><para>Commands with their
307 arguments that are executed when this
308 service is started. The first
309 argument must be an absolute path
312 <para>When <varname>Type</varname> is
313 not <option>oneshot</option>, only one
314 command may be given. When
315 <varname>Type=oneshot</varname> is
316 used, more than one command may be
317 specified. Multiple command lines may
318 be concatenated in a single directive,
319 by separating them with semicolons
320 (these semicolons must be passed as
321 separate words). Alternatively, this
322 directive may be specified more than
323 once with the same effect. However,
324 the latter syntax is not recommended
325 for compatibility with parsers
327 <filename>.desktop</filename> files.
328 Lone semicolons may be escaped as
329 <literal>\;</literal>. If the empty
330 string is assigned to this option, the
331 list of commands to start is reset,
332 prior assignments of this option will
333 have no effect.</para>
335 <para>If more than one command is
336 specified, the commands are invoked
337 one by one sequentially in the order
338 they appear in the unit file. If one
339 of the commands fails (and is not
340 prefixed with <literal>-</literal>),
341 other lines are not executed and the
342 unit is considered failed.</para>
345 <varname>Type=forking</varname> is
346 set, the process started via this
347 command line will be considered the
348 main process of the daemon.</para>
350 <para>The command line accepts
351 <literal>%</literal> specifiers as
353 <citerefentry><refentrytitle>systemd.unit</refentrytitle><manvolnum>5</manvolnum></citerefentry>. Note
354 that the first argument of the command
355 line (i.e. the program to execute) may
356 not include specifiers.</para>
358 <para>Basic environment variable
359 substitution is supported. Use
360 <literal>${FOO}</literal> as part of a
361 word, or as a word of its own on the
362 command line, in which case it will be
363 replaced by the value of the
364 environment variable including all
365 whitespace it contains, resulting in a
367 <literal>$FOO</literal> as a separate
368 word on the command line, in which
369 case it will be replaced by the value
370 of the environment variable split up
371 at whitespace, resulting in zero or
372 more arguments. To pass literal dollar sign
373 use <literal>$$</literal>. Note that the first
374 argument (i.e. the program to execute)
375 may not be a variable, since it must
376 be a literal and absolute path
379 <para>Optionally, if the absolute file
380 name is prefixed with
381 <literal>@</literal>, the second token
383 <literal>argv[0]</literal> to the
384 executed process, followed by the
385 further arguments specified. If the
386 absolute filename is prefixed with
387 <literal>-</literal>, an exit code of
388 the command normally considered a
389 failure (i.e. non-zero exit status or
390 abnormal exit due to signal) is ignored
391 and considered success. If both
392 <literal>-</literal> and
393 <literal>@</literal> are used, they
394 can appear in either order.</para>
396 <para>Note that this setting does not
397 directly support shell command
398 lines. If shell command lines are to
399 be used, they need to be passed
400 explicitly to a shell implementation
401 of some kind. Example:</para>
402 <programlisting>ExecStart=/bin/sh -c 'dmesg | tac'
405 <para>For services run by a user
406 instance of systemd the special
408 <varname>$MANAGERPID</varname> is set
409 to the PID of the systemd
415 <term><varname>ExecStartPre=</varname></term>
416 <term><varname>ExecStartPost=</varname></term>
417 <listitem><para>Additional commands
418 that are executed before or after
420 <varname>ExecStart=</varname>, respectively.
421 Syntax is the same as for
422 <varname>ExecStart=</varname>, except
423 that multiple command lines are allowed
424 and the commands are executed one
425 after the other, serially.</para>
427 <para>If any of those commands (not
428 prefixed with <literal>-</literal>)
429 fail, the rest are not executed and
430 the unit is considered failed.</para>
435 <term><varname>ExecReload=</varname></term>
436 <listitem><para>Commands to execute to
437 trigger a configuration reload in the
438 service. This argument takes multiple
439 command lines, following the same
440 scheme as described for
441 <varname>ExecStart=</varname>
442 above. Use of this setting is
443 optional. Specifier and environment
444 variable substitution is supported
445 here following the same scheme as for
446 <varname>ExecStart=</varname>.</para>
448 <para>One additional special
449 environment variables is set: if known
450 <varname>$MAINPID</varname> is set to
451 the main process of the daemon, and
452 may be used for command lines like the
455 <programlisting>/bin/kill -HUP $MAINPID</programlisting>
460 <term><varname>ExecStop=</varname></term>
461 <listitem><para>Commands to execute to
462 stop the service started via
463 <varname>ExecStart=</varname>. This
464 argument takes multiple command lines,
465 following the same scheme as described
466 for <varname>ExecStart=</varname>
467 above. Use of this setting is
468 optional. All processes remaining for
469 a service after the commands
470 configured in this option are run are
471 terminated according to the
472 <varname>KillMode=</varname> setting
474 <citerefentry><refentrytitle>systemd.kill</refentrytitle><manvolnum>5</manvolnum></citerefentry>). If
475 this option is not specified, the
476 process is terminated right-away when
477 service stop is requested. Specifier
478 and environment variable substitution
479 is supported (including
480 <varname>$MAINPID</varname>, see
481 above).</para></listitem>
485 <term><varname>ExecStopPost=</varname></term>
486 <listitem><para>Additional commands
487 that are executed after the service
488 was stopped. This includes cases where
489 the commands configured in
490 <varname>ExecStop=</varname> were used,
491 where the service does not have any
492 <varname>ExecStop=</varname> defined, or
493 where the service exited unexpectedly. This
494 argument takes multiple command lines,
495 following the same scheme as described
496 for <varname>ExecStart</varname>. Use
498 optional. Specifier and environment
499 variable substitution is
500 supported.</para></listitem>
504 <term><varname>RestartSec=</varname></term>
505 <listitem><para>Configures the time to
506 sleep before restarting a service (as
508 <varname>Restart=</varname>). Takes a
509 unit-less value in seconds, or a time
510 span value such as "5min
512 100ms.</para></listitem>
516 <term><varname>TimeoutStartSec=</varname></term>
517 <listitem><para>Configures the time to
518 wait for start-up. If a
519 daemon service does not signal
520 start-up completion within the
521 configured time, the service will be
522 considered failed and be shut down
524 Takes a unit-less value in seconds, or a
525 time span value such as "5min
526 20s". Pass 0 to disable the timeout
527 logic. Defaults to 90s, except when
528 <varname>Type=oneshot</varname> is
529 used in which case the timeout
530 is disabled by default.
535 <term><varname>TimeoutStopSec=</varname></term>
536 <listitem><para>Configures the time to
537 wait for stop. If a service is asked
538 to stop but does not terminate in the
539 specified time, it will be terminated
540 forcibly via <constant>SIGTERM</constant>, and after
541 another delay of this time with
542 <constant>SIGKILL</constant> (See
543 <varname>KillMode=</varname>
544 in <citerefentry><refentrytitle>systemd.kill</refentrytitle><manvolnum>5</manvolnum></citerefentry>).
545 Takes a unit-less value in seconds, or a
546 time span value such as "5min
547 20s". Pass 0 to disable the timeout
548 logic. Defaults to 90s.
553 <term><varname>TimeoutSec=</varname></term>
554 <listitem><para>A shorthand for configuring
555 both <varname>TimeoutStartSec=</varname>
556 and <varname>TimeoutStopSec=</varname>
557 to the specified value.
562 <term><varname>WatchdogSec=</varname></term>
563 <listitem><para>Configures the
564 watchdog timeout for a service. The
565 watchdog is activated when the start-up is
566 completed. The service must call
567 <citerefentry><refentrytitle>sd_notify</refentrytitle><manvolnum>3</manvolnum></citerefentry>
568 regularly with "WATCHDOG=1" (i.e. the
569 "keep-alive ping"). If the time
570 between two such calls is larger than
571 the configured time, then the service
572 is placed in a failure state. By
573 setting <varname>Restart=</varname> to
574 <option>on-failure</option> or
575 <option>always</option>, the service
576 will be automatically restarted. The
577 time configured here will be passed to
578 the executed service process in the
579 <varname>WATCHDOG_USEC=</varname>
580 environment variable. This allows
581 daemons to automatically enable the
582 keep-alive pinging logic if watchdog
583 support is enabled for the service. If
585 <varname>NotifyAccess=</varname> (see
586 below) should be set to open access to
587 the notification socket provided by
589 <varname>NotifyAccess=</varname> is
590 not set, it will be implicitly set to
591 <option>main</option>. Defaults to 0,
593 feature.</para></listitem>
597 <term><varname>Restart=</varname></term>
598 <listitem><para>Configures whether the
599 service shall be restarted when the
600 service process exits, is killed,
601 or a timeout is reached. The service
602 process may be the main service
603 process, but also one of the processes
605 <varname>ExecStartPre=</varname>,
606 <varname>ExecStartPost=</varname>,
607 <varname>ExecStopPre=</varname>,
608 <varname>ExecStopPost=</varname>, or
609 <varname>ExecReload=</varname>.
610 When the death of the process is a
611 result of systemd operation (e.g. service
612 stop or restart), the service will not be
613 restarted. Timeouts include missing
614 the watchdog "keep-alive ping"
615 deadline and a service start, reload,
616 and stop operation timeouts.</para>
620 <option>on-success</option>,
621 <option>on-failure</option>,
622 <option>on-watchdog</option>,
623 <option>on-abort</option>, or
624 <option>always</option>. If set to
625 <option>no</option> (the default), the
626 service will not be restarted. If set to
627 <option>on-success</option>, it will be
628 restarted only when the service process
630 In this context, a clean exit means
631 an exit code of 0, or one of the signals
632 <constant>SIGHUP</constant>, <constant>SIGINT</constant>, <constant>SIGTERM</constant>, or <constant>SIGPIPE</constant>, and
633 additionally, exit statuses and signals
634 specified in <varname>SuccessExitStatus=</varname>.
635 If set to <option>on-failure</option>,
636 the service will be restarted when the
637 process exits with an nonzero exit code,
638 is terminated by a signal (including on
639 core dump), when an operation (such as
640 service reload) times out, and when the
641 configured watchdog timeout is triggered.
643 <option>on-abort</option>, the service
644 will be restarted only if the service
645 process exits due to an uncaught
646 signal not specified as a clean exit
649 <option>on-watchdog</option>, the service
650 will be restarted only if the watchdog
651 timeout for the service expires.
653 <option>always</option>, the service
654 will be restarted regardless whether
655 it exited cleanly or not, got
656 terminated abnormally by a signal or
657 hit a timeout.</para>
659 <para>In addition to the above settings,
660 the service will not be restarted if the
661 exit code or signal is specified in
662 <varname>RestartPreventExitStatus=</varname>
663 (see below).</para></listitem>
667 <term><varname>SuccessExitStatus=</varname></term>
668 <listitem><para>Takes a list of exit
669 status definitions that when returned
670 by the main service process will be
671 considered successful termination, in
672 addition to the normal successful exit
673 code 0 and the signals <constant>SIGHUP</constant>, <constant>SIGINT</constant>,
674 <constant>SIGTERM</constant> and <constant>SIGPIPE</constant>. Exit status
675 definitions can either be numeric exit
676 codes or termination signal names,
677 separated by spaces. Example:
678 <literal>SuccessExitStatus=1 2 8
679 <constant>SIGKILL</constant></literal>, ensures that exit
680 codes 1, 2, 8 and the termination
681 signal <constant>SIGKILL</constant> are considered clean
682 service terminations. This option may
683 appear more than once in which case
684 the list of successful exit statuses
685 is merged. If the empty string is
686 assigned to this option, the list is
687 reset, all prior assignments of this
689 effect.</para></listitem>
693 <term><varname>RestartPreventExitStatus=</varname></term>
694 <listitem><para>Takes a list of exit
695 status definitions that when returned
696 by the main service process will
697 prevent automatic service restarts
698 regardless of the restart setting
700 <varname>Restart=</varname>. Exit
701 status definitions can either be
702 numeric exit codes or termination
703 signal names, and are separated by
704 spaces. Defaults to the empty list, so
705 that by default no exit status is
706 excluded from the configured restart
708 <literal>RestartPreventExitStatus=1 6
709 SIGABRT</literal>, ensures that exit
710 codes 1 and 6 and the termination
711 signal SIGABRT will not result in
712 automatic service restarting. This
713 option may appear more than once in
714 which case the list of restart preventing
715 statuses is merged. If the empty
716 string is assigned to this option, the
717 list is reset, all prior assignments
718 of this option will have no
719 effect.</para></listitem>
723 <term><varname>PermissionsStartOnly=</varname></term>
724 <listitem><para>Takes a boolean
725 argument. If true, the permission
726 related execution options as
728 <varname>User=</varname> and similar
730 <citerefentry><refentrytitle>systemd.exec</refentrytitle><manvolnum>5</manvolnum></citerefentry>
731 for more information) are only applied
732 to the process started with
733 <varname>ExecStart=</varname>, and not
735 <varname>ExecStartPre=</varname>,
736 <varname>ExecStartPost=</varname>,
737 <varname>ExecReload=</varname>,
738 <varname>ExecStop=</varname>,
739 <varname>ExecStopPost=</varname>
740 commands. If false, the setting is
741 applied to all configured commands the
742 same way. Defaults to
743 false.</para></listitem>
747 <term><varname>RootDirectoryStartOnly=</varname></term>
748 <listitem><para>Takes a boolean
749 argument. If true, the root directory
750 as configured with the
751 <varname>RootDirectory=</varname>
753 <citerefentry><refentrytitle>systemd.exec</refentrytitle><manvolnum>5</manvolnum></citerefentry>
754 for more information) is only applied
755 to the process started with
756 <varname>ExecStart=</varname>, and not
758 <varname>ExecStartPre=</varname>,
759 <varname>ExecStartPost=</varname>,
760 <varname>ExecReload=</varname>,
761 <varname>ExecStop=</varname>,
762 <varname>ExecStopPost=</varname>
763 commands. If false, the setting is
764 applied to all configured commands the
765 same way. Defaults to
766 false.</para></listitem>
770 <term><varname>NonBlocking=</varname></term>
771 <listitem><para>Set O_NONBLOCK flag
772 for all file descriptors passed via
773 socket-based activation. If true, all
774 file descriptors >= 3 (i.e. all except
775 STDIN/STDOUT/STDERR) will have
776 the O_NONBLOCK flag set and hence are in
777 non-blocking mode. This option is only
778 useful in conjunction with a socket
779 unit, as described in
780 <citerefentry><refentrytitle>systemd.socket</refentrytitle><manvolnum>5</manvolnum></citerefentry>. Defaults
781 to false.</para></listitem>
785 <term><varname>NotifyAccess=</varname></term>
786 <listitem><para>Controls access to the
787 service status notification socket, as
789 <citerefentry><refentrytitle>sd_notify</refentrytitle><manvolnum>3</manvolnum></citerefentry>
791 <option>none</option> (the default),
792 <option>main</option> or
793 <option>all</option>. If
794 <option>none</option>, no daemon status
795 updates are accepted from the service
796 processes, all status update messages
797 are ignored. If <option>main</option>,
798 only service updates sent from the
799 main process of the service are
800 accepted. If <option>all</option>, all
801 services updates from all members of
802 the service's control group are
803 accepted. This option should be set to
804 open access to the notification socket
806 <varname>Type=notify</varname> or
807 <varname>WatchdogSec=</varname> (see
808 above). If those options are used but
809 <varname>NotifyAccess=</varname> is not
810 configured, it will be implicitly set
812 <option>main</option>.</para></listitem>
816 <term><varname>Sockets=</varname></term>
817 <listitem><para>Specifies the name of
818 the socket units this service shall
819 inherit the sockets from when the
820 service is started. Normally it
821 should not be necessary to use this
822 setting as all sockets whose unit
823 shares the same name as the service
824 (ignoring the different suffix of course)
825 are passed to the spawned
828 <para>Note that the same socket may be
829 passed to multiple processes at the
830 same time. Also note that a different
831 service may be activated on incoming
832 traffic than inherits the sockets. Or
834 <varname>Service=</varname> setting of
835 <filename>.socket</filename> units
836 does not have to match the inverse of
837 the <varname>Sockets=</varname>
839 <filename>.service</filename> it
842 <para>This option may appear more than
843 once, in which case the list of socket
844 units is merged. If the empty string
845 is assigned to this option, the list of
846 sockets is reset, all prior uses of
847 this setting will have no
848 effect.</para></listitem>
852 <term><varname>StartLimitInterval=</varname></term>
853 <term><varname>StartLimitBurst=</varname></term>
855 <listitem><para>Configure service
856 start rate limiting. By default,
857 services which are started more often
858 than 5 times within 10s are not
859 permitted to start any more times
860 until the 10s interval ends. With
861 these two options, this rate limiting
863 <varname>StartLimitInterval=</varname>
864 to configure the checking interval
865 (defaults to 10s, set to 0 to disable
866 any kind of rate limiting). Use
867 <varname>StartLimitBurst=</varname> to
868 configure how many starts per interval
869 are allowed (defaults to 5). These
870 configuration options are particularly
871 useful in conjunction with
872 <varname>Restart=</varname>, however
873 apply to all kinds of starts
874 (including manual), not just those
876 <varname>Restart=</varname> logic.
877 Note that units which are configured
878 for <varname>Restart=</varname> and
879 which reach the start limit are not
880 attempted to be restarted anymore,
881 however they may still be restarted
882 manually at a later point from which
883 point on the restart logic is again
886 reset-failed</command> will cause the
887 restart rate counter for a service to
888 be flushed, which is useful if the
889 administrator wants to manually start
890 a service and the start limit
892 that.</para></listitem>
896 <term><varname>StartLimitAction=</varname></term>
898 <listitem><para>Configure the action
899 to take if the rate limit configured
901 <varname>StartLimitInterval=</varname>
903 <varname>StartLimitBurst=</varname> is
905 <option>none</option>,
906 <option>reboot</option>,
907 <option>reboot-force</option> or
908 <option>reboot-immediate</option>. If
909 <option>none</option> is set,
910 hitting the rate limit will trigger no
911 action besides that the start will not
913 permitted. <option>reboot</option>
914 causes a reboot following the normal
915 shutdown procedure (i.e. equivalent to
916 <command>systemctl reboot</command>),
917 <option>reboot-force</option> causes
918 an forced reboot which will terminate
919 all processes forcibly but should
920 cause no dirty file systems on reboot
921 (i.e. equivalent to <command>systemctl
922 reboot -f</command>) and
923 <option>reboot-immediate</option>
924 causes immediate execution of the
925 <citerefentry><refentrytitle>reboot</refentrytitle><manvolnum>2</manvolnum></citerefentry>
926 system call, which might result in
927 data loss. Defaults to
928 <option>none</option>.</para></listitem>
934 <citerefentry><refentrytitle>systemd.exec</refentrytitle><manvolnum>5</manvolnum></citerefentry>
936 <citerefentry><refentrytitle>systemd.kill</refentrytitle><manvolnum>5</manvolnum></citerefentry>
937 for more settings.</para>
942 <title>Compatibility Options</title>
944 <para>The following options are also available in the
945 <literal>[Service]</literal> section, but exist purely
946 for compatibility reasons and should not be used in
947 newly written service files.</para>
949 <variablelist class='unit-directives'>
951 <term><varname>SysVStartPriority=</varname></term>
952 <listitem><para>Set the SysV start
953 priority to use to order this service
954 in relation to SysV services lacking
955 LSB headers. This option is only
956 necessary to fix ordering in relation
957 to legacy SysV services, that have no
958 ordering information encoded in the
959 script headers. As such it should only
960 be used as temporary compatibility
961 option, and not be used in new unit
962 files. Almost always it is a better
963 choice to add explicit ordering
965 <varname>After=</varname> or
966 <varname>Before=</varname>,
967 instead. For more details see
968 <citerefentry><refentrytitle>systemd.unit</refentrytitle><manvolnum>5</manvolnum></citerefentry>. If
969 used, pass an integer value in the
970 range 0-99.</para></listitem>
974 <term><varname>FsckPassNo=</varname></term>
975 <listitem><para>Set the fsck passno
976 priority to use to order this service
977 in relation to other file system
978 checking services. This option is only
979 necessary to fix ordering in relation
980 to fsck jobs automatically created for
981 all <filename>/etc/fstab</filename>
982 entries with a value in the fs_passno
983 column > 0. As such it should only be
984 used as option for fsck
985 services. Almost always it is a better
986 choice to add explicit ordering
988 <varname>After=</varname> or
989 <varname>Before=</varname>,
990 instead. For more details see
991 <citerefentry><refentrytitle>systemd.unit</refentrytitle><manvolnum>5</manvolnum></citerefentry>. If
992 used, pass an integer value in the
994 <filename>/etc/fstab</filename>'s
995 fs_passno column. See
996 <citerefentry><refentrytitle>fstab</refentrytitle><manvolnum>5</manvolnum></citerefentry>
997 for details.</para></listitem>
1004 <title>See Also</title>
1006 <citerefentry><refentrytitle>systemd</refentrytitle><manvolnum>1</manvolnum></citerefentry>,
1007 <citerefentry><refentrytitle>systemctl</refentrytitle><manvolnum>8</manvolnum></citerefentry>,
1008 <citerefentry><refentrytitle>systemd.unit</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
1009 <citerefentry><refentrytitle>systemd.exec</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
1010 <citerefentry><refentrytitle>systemd.cgroup</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
1011 <citerefentry><refentrytitle>systemd.kill</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
1012 <citerefentry><refentrytitle>systemd.directives</refentrytitle><manvolnum>7</manvolnum></citerefentry>