1 <?xml version='
1.0'
?> <!--*- Mode: nxml; nxml-child-indent: 2; indent-tabs-mode: nil -*-->
2 <!DOCTYPE refentry PUBLIC
"-//OASIS//DTD DocBook XML V4.2//EN"
3 "http://www.oasis-open.org/docbook/xml/4.2/docbookx.dtd">
6 SPDX-License-Identifier: LGPL-2.1+
8 This file is part of systemd.
10 Copyright 2010 Lennart Poettering
12 systemd is free software; you can redistribute it and/or modify it
13 under the terms of the GNU Lesser General Public License as published by
14 the Free Software Foundation; either version 2.1 of the License, or
15 (at your option) any later version.
17 systemd is distributed in the hope that it will be useful, but
18 WITHOUT ANY WARRANTY; without even the implied warranty of
19 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
20 Lesser General Public License for more details.
22 You should have received a copy of the GNU Lesser General Public License
23 along with systemd; If not, see <http://www.gnu.org/licenses/>.
26 <refentry id=
"systemd"
27 xmlns:
xi=
"http://www.w3.org/2001/XInclude">
30 <title>systemd
</title>
31 <productname>systemd
</productname>
35 <contrib>Developer
</contrib>
36 <firstname>Lennart
</firstname>
37 <surname>Poettering
</surname>
38 <email>lennart@poettering.net
</email>
44 <refentrytitle>systemd
</refentrytitle>
45 <manvolnum>1</manvolnum>
49 <refname>systemd
</refname>
50 <refname>init
</refname>
51 <refpurpose>systemd system and service manager
</refpurpose>
56 <command>/usr/lib/systemd/systemd
</command>
57 <arg choice=
"opt" rep=
"repeat">OPTIONS
</arg>
60 <command>init
</command>
61 <arg choice=
"opt" rep=
"repeat">OPTIONS
</arg>
62 <arg choice=
"req">COMMAND
</arg>
67 <title>Description
</title>
69 <para>systemd is a system and service manager for Linux operating
70 systems. When run as first process on boot (as PID
1), it acts as
71 init system that brings up and maintains userspace
74 <para>For compatibility with SysV, if systemd is called as
75 <command>init
</command> and a PID that is not
1, it will execute
76 <command>telinit
</command> and pass all command line arguments
77 unmodified. That means
<command>init
</command> and
78 <command>telinit
</command> are mostly equivalent when invoked from
79 normal login sessions. See
80 <citerefentry><refentrytitle>telinit
</refentrytitle><manvolnum>8</manvolnum></citerefentry>
81 for more information.
</para>
83 <para>When run as a system instance, systemd interprets the
84 configuration file
<filename>system.conf
</filename> and the files
85 in
<filename>system.conf.d
</filename> directories; when run as a
86 user instance, systemd interprets the configuration file
87 <filename>user.conf
</filename> and the files in
88 <filename>user.conf.d
</filename> directories. See
89 <citerefentry><refentrytitle>systemd-system.conf
</refentrytitle><manvolnum>5</manvolnum></citerefentry>
90 for more information.
</para>
94 <title>Options
</title>
96 <para>The following options are understood:
</para>
100 <term><option>--test
</option></term>
102 <listitem><para>Determine startup sequence, dump it and exit.
103 This is an option useful for debugging only.
</para></listitem>
106 <term><option>--dump-configuration-items
</option></term>
108 <listitem><para>Dump understood unit configuration items. This
109 outputs a terse but complete list of configuration items
110 understood in unit definition files.
</para></listitem>
113 <term><option>--unit=
</option></term>
115 <listitem><para>Set default unit to activate on startup. If
116 not specified, defaults to
117 <filename>default.target
</filename>.
</para></listitem>
120 <term><option>--system
</option></term>
121 <term><option>--user
</option></term>
123 <listitem><para>For
<option>--system
</option>, tell systemd to
124 run a system instance, even if the process ID is not
1, i.e.
125 systemd is not run as init process.
<option>--user
</option>
126 does the opposite, running a user instance even if the process
127 ID is
1. Normally, it should not be necessary to pass these
128 options, as systemd automatically detects the mode it is
129 started in. These options are hence of little use except for
130 debugging. Note that it is not supported booting and
131 maintaining a full system with systemd running in
132 <option>--system
</option> mode, but PID not
1. In practice,
133 passing
<option>--system
</option> explicitly is only useful in
134 conjunction with
<option>--test
</option>.
</para></listitem>
137 <term><option>--dump-core
</option></term>
139 <listitem><para>Enable core dumping on crash. This switch has
140 no effect when running as user instance. This setting may also
141 be enabled during boot on the kernel command line via the
142 <varname>systemd.dump_core=
</varname> option, see
143 below.
</para></listitem>
147 <term><option>--crash-vt=
</option><replaceable>VT
</replaceable></term>
149 <listitem><para>Switch to a specific virtual console (VT) on
150 crash. Takes a positive integer in the range
1–
63, or a
151 boolean argument. If an integer is passed, selects which VT to
152 switch to. If
<constant>yes
</constant>, the VT kernel messages
153 are written to is selected. If
<constant>no
</constant>, no VT
154 switch is attempted. This switch has no effect when running as
155 user instance. This setting may also be enabled during boot,
156 on the kernel command line via the
157 <varname>systemd.crash_vt=
</varname> option, see
158 <!-- FIXME: there is no crash_vt command line option? -->
159 below.
</para></listitem>
163 <term><option>--crash-shell
</option></term>
165 <listitem><para>Run a shell on crash. This switch has no
166 effect when running as user instance. This setting may also be
167 enabled during boot, on the kernel command line via the
168 <varname>systemd.crash_shell=
</varname> option, see
169 below.
</para></listitem>
173 <term><option>--crash-reboot
</option></term>
175 <listitem><para>Automatically reboot the system on crash. This
176 switch has no effect when running as user instance. This
177 setting may also be enabled during boot, on the kernel command
178 line via the
<varname>systemd.crash_reboot=
</varname> option,
179 see below.
</para></listitem>
183 <term><option>--confirm-spawn
</option></term>
185 <listitem><para>Ask for confirmation when spawning processes.
186 This switch has no effect when run as user
187 instance.
</para></listitem>
190 <term><option>--show-status=
</option></term>
192 <listitem><para>Takes a boolean argument or the special value
<constant>auto
</constant>. If on, terse unit
193 status information is shown on the console during boot-up and shutdown. If off, no such status information is
194 shown. If set to
<constant>auto
</constant> behavior is similar to off, except that it is automatically switched
195 to on, as soon as the first unit failure or significant boot delay is encountered. This switch has no effect
196 when invoked as user instance. If specified, overrides both the kernel command line setting
197 <varname>systemd.show_status=
</varname> (see below) and the configuration file option
198 <option>ShowStatus=
</option>, see
199 <citerefentry><refentrytitle>systemd-system.conf
</refentrytitle><manvolnum>5</manvolnum></citerefentry>.
</para></listitem>
202 <term><option>--log-target=
</option></term>
204 <listitem><para>Set log target. Argument must be one of
205 <option>console
</option>,
206 <option>journal
</option>,
207 <option>kmsg
</option>,
208 <option>journal-or-kmsg
</option>,
209 <option>null
</option>.
</para></listitem>
212 <term><option>--log-level=
</option></term>
214 <listitem><para>Set log level. As
215 argument this accepts a numerical log
216 level or the well-known
<citerefentry project='man-pages'
><refentrytitle>syslog
</refentrytitle><manvolnum>3</manvolnum></citerefentry>
217 symbolic names (lowercase):
218 <option>emerg
</option>,
219 <option>alert
</option>,
220 <option>crit
</option>,
221 <option>err
</option>,
222 <option>warning
</option>,
223 <option>notice
</option>,
224 <option>info
</option>,
225 <option>debug
</option>.
</para></listitem>
228 <term><option>--log-color=
</option></term>
230 <listitem><para>Highlight important log messages. Argument is
231 a boolean value. If the argument is omitted, it defaults to
232 <option>true
</option>.
</para></listitem>
235 <term><option>--log-location=
</option></term>
237 <listitem><para>Include code location in log messages. This is
238 mostly relevant for debugging purposes. Argument is a boolean
239 value. If the argument is omitted it defaults to
240 <option>true
</option>.
</para></listitem>
243 <term><option>--default-standard-output=
</option></term>
244 <term><option>--default-standard-error=
</option></term>
246 <listitem><para>Sets the default output or error output for
247 all services and sockets, respectively. That is, controls the
248 default for
<option>StandardOutput=
</option> and
249 <option>StandardError=
</option> (see
250 <citerefentry><refentrytitle>systemd.exec
</refentrytitle><manvolnum>5</manvolnum></citerefentry>
251 for details). Takes one of
252 <option>inherit
</option>,
253 <option>null
</option>,
254 <option>tty
</option>,
255 <option>journal
</option>,
256 <option>journal+console
</option>,
257 <option>syslog
</option>,
258 <option>syslog+console
</option>,
259 <option>kmsg
</option>,
260 <option>kmsg+console
</option>. If the
262 <option>--default-standard-output=
</option> defaults to
263 <option>journal
</option> and
264 <option>--default-standard-error=
</option> to
265 <option>inherit
</option>.
</para></listitem>
269 <term><option>--machine-id=
</option></term>
271 <listitem><para>Override the machine-id set on the hard drive,
272 useful for network booting or for containers. May not be set
273 to all zeros.
</para></listitem>
277 <term><option>--service-watchdogs=
</option></term>
279 <listitem><para>Globally enable/disable all service watchdog timeouts and emergency
280 actions. This setting may also be specified during boot, on the kernel
281 command line via the
<varname>systemd.service_watchdogs=
</varname>
282 option, see below. Defaults to enabled.
</para></listitem>
285 <xi:include href=
"standard-options.xml" xpointer=
"help" />
286 <xi:include href=
"standard-options.xml" xpointer=
"version" />
291 <title>Concepts
</title>
293 <para>systemd provides a dependency system between various
294 entities called
"units" of
11 different types. Units encapsulate
295 various objects that are relevant for system boot-up and
296 maintenance. The majority of units are configured in unit
297 configuration files, whose syntax and basic set of options is
299 <citerefentry><refentrytitle>systemd.unit
</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
300 however some are created automatically from other configuration,
301 dynamically from system state or programmatically at runtime.
302 Units may be
"active" (meaning started, bound, plugged in, …,
303 depending on the unit type, see below), or
"inactive" (meaning
304 stopped, unbound, unplugged, …), as well as in the process of
305 being activated or deactivated, i.e. between the two states (these
306 states are called
"activating",
"deactivating"). A special
307 "failed" state is available as well, which is very similar to
308 "inactive" and is entered when the service failed in some way
309 (process returned error code on exit, or crashed, an operation
310 timed out, or after too many restarts). If this state is entered,
311 the cause will be logged, for later reference. Note that the
312 various unit types may have a number of additional substates,
313 which are mapped to the five generalized unit states described
316 <para>The following unit types are available:
</para>
319 <listitem><para>Service units, which start and control daemons
320 and the processes they consist of. For details, see
321 <citerefentry><refentrytitle>systemd.service
</refentrytitle><manvolnum>5</manvolnum></citerefentry>.
</para></listitem>
323 <listitem><para>Socket units, which encapsulate local IPC or
324 network sockets in the system, useful for socket-based
325 activation. For details about socket units, see
326 <citerefentry><refentrytitle>systemd.socket
</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
327 for details on socket-based activation and other forms of
329 <citerefentry><refentrytitle>daemon
</refentrytitle><manvolnum>7</manvolnum></citerefentry>.
</para></listitem>
331 <listitem><para>Target units are useful to group units, or
332 provide well-known synchronization points during boot-up, see
333 <citerefentry><refentrytitle>systemd.target
</refentrytitle><manvolnum>5</manvolnum></citerefentry>.
</para></listitem>
335 <listitem><para>Device units expose kernel devices in systemd
336 and may be used to implement device-based activation. For
338 <citerefentry><refentrytitle>systemd.device
</refentrytitle><manvolnum>5</manvolnum></citerefentry>.
</para></listitem>
340 <listitem><para>Mount units control mount points in the file
341 system, for details see
342 <citerefentry><refentrytitle>systemd.mount
</refentrytitle><manvolnum>5</manvolnum></citerefentry>.
</para></listitem>
344 <listitem><para>Automount units provide automount capabilities,
345 for on-demand mounting of file systems as well as parallelized
347 <citerefentry><refentrytitle>systemd.automount
</refentrytitle><manvolnum>5</manvolnum></citerefentry>.
</para></listitem>
349 <listitem><para>Timer units are useful for triggering activation
350 of other units based on timers. You may find details in
351 <citerefentry><refentrytitle>systemd.timer
</refentrytitle><manvolnum>5</manvolnum></citerefentry>.
</para></listitem>
353 <listitem><para>Swap units are very similar to mount units and
354 encapsulate memory swap partitions or files of the operating
355 system. They are described in
356 <citerefentry><refentrytitle>systemd.swap
</refentrytitle><manvolnum>5</manvolnum></citerefentry>.
</para></listitem>
358 <listitem><para>Path units may be used to activate other
359 services when file system objects change or are modified. See
360 <citerefentry><refentrytitle>systemd.path
</refentrytitle><manvolnum>5</manvolnum></citerefentry>.
</para></listitem>
362 <listitem><para>Slice units may be used to group units which
363 manage system processes (such as service and scope units) in a
364 hierarchical tree for resource management purposes. See
365 <citerefentry><refentrytitle>systemd.slice
</refentrytitle><manvolnum>5</manvolnum></citerefentry>.
</para></listitem>
367 <listitem><para>Scope units are similar to service units, but
368 manage foreign processes instead of starting them as well. See
369 <citerefentry><refentrytitle>systemd.scope
</refentrytitle><manvolnum>5</manvolnum></citerefentry>.
</para></listitem>
373 <para>Units are named as their configuration files. Some units
374 have special semantics. A detailed list is available in
375 <citerefentry><refentrytitle>systemd.special
</refentrytitle><manvolnum>7</manvolnum></citerefentry>.
</para>
377 <para>systemd knows various kinds of dependencies, including
378 positive and negative requirement dependencies (i.e.
379 <varname>Requires=
</varname> and
<varname>Conflicts=
</varname>) as
380 well as ordering dependencies (
<varname>After=
</varname> and
381 <varname>Before=
</varname>). NB: ordering and requirement
382 dependencies are orthogonal. If only a requirement dependency
383 exists between two units (e.g.
<filename>foo.service
</filename>
384 requires
<filename>bar.service
</filename>), but no ordering
385 dependency (e.g.
<filename>foo.service
</filename> after
386 <filename>bar.service
</filename>) and both are requested to start,
387 they will be started in parallel. It is a common pattern that both
388 requirement and ordering dependencies are placed between two
389 units. Also note that the majority of dependencies are implicitly
390 created and maintained by systemd. In most cases, it should be
391 unnecessary to declare additional dependencies manually, however
392 it is possible to do this.
</para>
394 <para>Application programs and units (via dependencies) may
395 request state changes of units. In systemd, these requests are
396 encapsulated as 'jobs' and maintained in a job queue. Jobs may
397 succeed or can fail, their execution is ordered based on the
398 ordering dependencies of the units they have been scheduled
401 <para>On boot systemd activates the target unit
402 <filename>default.target
</filename> whose job is to activate
403 on-boot services and other on-boot units by pulling them in via
404 dependencies. Usually, the unit name is just an alias (symlink) for
405 either
<filename>graphical.target
</filename> (for fully-featured
406 boots into the UI) or
<filename>multi-user.target
</filename> (for
407 limited console-only boots for use in embedded or server
408 environments, or similar; a subset of graphical.target). However,
409 it is at the discretion of the administrator to configure it as an
410 alias to any other target unit. See
411 <citerefentry><refentrytitle>systemd.special
</refentrytitle><manvolnum>7</manvolnum></citerefentry>
412 for details about these target units.
</para>
414 <para>Processes systemd spawns are placed in individual Linux
415 control groups named after the unit which they belong to in the
416 private systemd hierarchy. (see
<ulink
417 url=
"https://www.kernel.org/doc/Documentation/cgroup-v1/cgroups.txt">cgroups.txt
</ulink>
418 for more information about control groups, or short
"cgroups").
419 systemd uses this to effectively keep track of processes. Control
420 group information is maintained in the kernel, and is accessible
421 via the file system hierarchy (beneath
422 <filename>/sys/fs/cgroup/systemd/
</filename>), or in tools such as
423 <citerefentry project='man-pages'
><refentrytitle>systemd-cgls
</refentrytitle><manvolnum>1</manvolnum></citerefentry>
425 <citerefentry project='man-pages'
><refentrytitle>ps
</refentrytitle><manvolnum>1</manvolnum></citerefentry>
426 (
<command>ps xawf -eo pid,user,cgroup,args
</command> is
427 particularly useful to list all processes and the systemd units
428 they belong to.).
</para>
430 <para>systemd is compatible with the SysV init system to a large
431 degree: SysV init scripts are supported and simply read as an
432 alternative (though limited) configuration file format. The SysV
433 <filename>/dev/initctl
</filename> interface is provided, and
434 compatibility implementations of the various SysV client tools are
435 available. In addition to that, various established Unix
436 functionality such as
<filename>/etc/fstab
</filename> or the
437 <filename>utmp
</filename> database are supported.
</para>
439 <para>systemd has a minimal transaction system: if a unit is
440 requested to start up or shut down it will add it and all its
441 dependencies to a temporary transaction. Then, it will verify if
442 the transaction is consistent (i.e. whether the ordering of all
443 units is cycle-free). If it is not, systemd will try to fix it up,
444 and removes non-essential jobs from the transaction that might
445 remove the loop. Also, systemd tries to suppress non-essential
446 jobs in the transaction that would stop a running service. Finally
447 it is checked whether the jobs of the transaction contradict jobs
448 that have already been queued, and optionally the transaction is
449 aborted then. If all worked out and the transaction is consistent
450 and minimized in its impact it is merged with all already
451 outstanding jobs and added to the run queue. Effectively this
452 means that before executing a requested operation, systemd will
453 verify that it makes sense, fixing it if possible, and only
454 failing if it really cannot work.
</para>
456 <para>systemd contains native implementations of various tasks
457 that need to be executed as part of the boot process. For example,
458 it sets the hostname or configures the loopback network device. It
459 also sets up and mounts various API file systems, such as
460 <filename>/sys
</filename> or
<filename>/proc
</filename>.
</para>
462 <para>For more information about the concepts and
463 ideas behind systemd, please refer to the
464 <ulink url=
"http://0pointer.de/blog/projects/systemd.html">Original Design Document
</ulink>.
</para>
466 <para>Note that some but not all interfaces provided
467 by systemd are covered by the
468 <ulink url=
"https://www.freedesktop.org/wiki/Software/systemd/InterfaceStabilityPromise">Interface
469 Stability Promise
</ulink>.
</para>
471 <para>Units may be generated dynamically at boot and system
472 manager reload time, for example based on other configuration
473 files or parameters passed on the kernel command line. For details, see
474 <citerefentry><refentrytitle>systemd.generator
</refentrytitle><manvolnum>7</manvolnum></citerefentry>.
</para>
476 <para>Systems which invoke systemd in a container or initrd
477 environment should implement the
478 <ulink url=
"https://www.freedesktop.org/wiki/Software/systemd/ContainerInterface">Container Interface
</ulink> or
479 <ulink url=
"https://www.freedesktop.org/wiki/Software/systemd/InitrdInterface">initrd Interface
</ulink>
480 specifications, respectively.
</para>
484 <title>Directories
</title>
488 <term>System unit directories
</term>
490 <listitem><para>The systemd system manager reads unit
491 configuration from various directories. Packages that want to
492 install unit files shall place them in the directory returned
493 by
<command>pkg-config systemd
494 --variable=systemdsystemunitdir
</command>. Other directories
495 checked are
<filename>/usr/local/lib/systemd/system
</filename>
496 and
<filename>/usr/lib/systemd/system
</filename>. User
497 configuration always takes precedence.
<command>pkg-config
498 systemd --variable=systemdsystemconfdir
</command> returns the
499 path of the system configuration directory. Packages should
500 alter the content of these directories only with the
501 <command>enable
</command> and
<command>disable
</command>
503 <citerefentry><refentrytitle>systemctl
</refentrytitle><manvolnum>1</manvolnum></citerefentry>
504 tool. Full list of directories is provided in
505 <citerefentry><refentrytitle>systemd.unit
</refentrytitle><manvolnum>5</manvolnum></citerefentry>.
512 <term>User unit directories
</term>
514 <listitem><para>Similar rules apply for the user unit
515 directories. However, here the
516 <ulink url=
"http://standards.freedesktop.org/basedir-spec/basedir-spec-latest.html">XDG
517 Base Directory specification
</ulink> is followed to find
518 units. Applications should place their unit files in the
519 directory returned by
<command>pkg-config systemd
520 --variable=systemduserunitdir
</command>. Global configuration
521 is done in the directory reported by
<command>pkg-config
522 systemd --variable=systemduserconfdir
</command>. The
523 <command>enable
</command> and
<command>disable
</command>
525 <citerefentry><refentrytitle>systemctl
</refentrytitle><manvolnum>1</manvolnum></citerefentry>
526 tool can handle both global (i.e. for all users) and private
527 (for one user) enabling/disabling of units. Full list of
528 directories is provided in
529 <citerefentry><refentrytitle>systemd.unit
</refentrytitle><manvolnum>5</manvolnum></citerefentry>.
536 <term>SysV init scripts directory
</term>
538 <listitem><para>The location of the SysV init script directory
539 varies between distributions. If systemd cannot find a native
540 unit file for a requested service, it will look for a SysV
541 init script of the same name (with the
542 <filename>.service
</filename> suffix
543 removed).
</para></listitem>
549 <term>SysV runlevel link farm directory
</term>
551 <listitem><para>The location of the SysV runlevel link farm
552 directory varies between distributions. systemd will take the
553 link farm into account when figuring out whether a service
554 shall be enabled. Note that a service unit with a native unit
555 configuration file cannot be started by activating it in the
556 SysV runlevel link farm.
</para></listitem>
562 <title>Signals
</title>
566 <term><constant>SIGTERM
</constant></term>
568 <listitem><para>Upon receiving this signal the systemd system
569 manager serializes its state, reexecutes itself and
570 deserializes the saved state again. This is mostly equivalent
571 to
<command>systemctl daemon-reexec
</command>.
</para>
573 <para>systemd user managers will start the
574 <filename>exit.target
</filename> unit when this signal is
575 received. This is mostly equivalent to
<command>systemctl
576 --user start exit.target
577 --job-mode=replace-irreversible
</command>.
</para></listitem>
581 <term><constant>SIGINT
</constant></term>
583 <listitem><para>Upon receiving this signal the systemd system
584 manager will start the
585 <filename>ctrl-alt-del.target
</filename> unit. This is mostly
586 equivalent to
<command>systemctl start ctrl-alt-del.target
587 --job-mode=replace-irreversible
</command>. If this signal is
588 received more than
7 times per
2s, an immediate reboot is
589 triggered. Note that pressing Ctrl-Alt-Del on the console
590 will trigger this signal. Hence, if a reboot is hanging,
591 pressing Ctrl-Alt-Del more than
7 times in
2s is a relatively
592 safe way to trigger an immediate reboot.
</para>
594 <para>systemd user managers treat this signal the same way as
595 <constant>SIGTERM
</constant>.
</para></listitem>
599 <term><constant>SIGWINCH
</constant></term>
601 <listitem><para>When this signal is received the systemd
602 system manager will start the
603 <filename>kbrequest.target
</filename> unit. This is mostly
604 equivalent to
<command>systemctl start
605 kbrequest.target
</command>.
</para>
607 <para>This signal is ignored by systemd user
608 managers.
</para></listitem>
612 <term><constant>SIGPWR
</constant></term>
614 <listitem><para>When this signal is received the systemd
615 manager will start the
<filename>sigpwr.target
</filename>
616 unit. This is mostly equivalent to
<command>systemctl start
617 sigpwr.target
</command>.
</para></listitem>
621 <term><constant>SIGUSR1
</constant></term>
623 <listitem><para>When this signal is received the systemd
624 manager will try to reconnect to the D-Bus
625 bus.
</para></listitem>
629 <term><constant>SIGUSR2
</constant></term>
631 <listitem><para>When this signal is received the systemd
632 manager will log its complete state in human-readable form.
633 The data logged is the same as printed by
634 <command>systemd-analyze dump
</command>.
</para></listitem>
638 <term><constant>SIGHUP
</constant></term>
640 <listitem><para>Reloads the complete daemon configuration.
641 This is mostly equivalent to
<command>systemctl
642 daemon-reload
</command>.
</para></listitem>
646 <term><constant>SIGRTMIN+
0</constant></term>
648 <listitem><para>Enters default mode, starts the
649 <filename>default.target
</filename> unit. This is mostly
650 equivalent to
<command>systemctl isolate
651 default.target
</command>.
</para></listitem>
655 <term><constant>SIGRTMIN+
1</constant></term>
657 <listitem><para>Enters rescue mode, starts the
658 <filename>rescue.target
</filename> unit. This is mostly
659 equivalent to
<command>systemctl isolate
660 rescue.target
</command>.
</para></listitem>
664 <term><constant>SIGRTMIN+
2</constant></term>
666 <listitem><para>Enters emergency mode, starts the
667 <filename>emergency.service
</filename> unit. This is mostly
668 equivalent to
<command>systemctl isolate
669 emergency.service
</command>.
</para></listitem>
673 <term><constant>SIGRTMIN+
3</constant></term>
675 <listitem><para>Halts the machine, starts the
676 <filename>halt.target
</filename> unit. This is mostly
677 equivalent to
<command>systemctl start halt.target
678 --job-mode=replace-irreversible
</command>.
</para>
683 <term><constant>SIGRTMIN+
4</constant></term>
685 <listitem><para>Powers off the machine, starts the
686 <filename>poweroff.target
</filename> unit. This is mostly
687 equivalent to
<command>systemctl start poweroff.target
688 --job-mode=replace-irreversible
</command>.
</para>
693 <term><constant>SIGRTMIN+
5</constant></term>
695 <listitem><para>Reboots the machine, starts the
696 <filename>reboot.target
</filename> unit. This is mostly
697 equivalent to
<command>systemctl start reboot.target
698 --job-mode=replace-irreversible
</command>.
</para>
703 <term><constant>SIGRTMIN+
6</constant></term>
705 <listitem><para>Reboots the machine via kexec, starts the
706 <filename>kexec.target
</filename> unit. This is mostly
707 equivalent to
<command>systemctl start kexec.target
708 --job-mode=replace-irreversible
</command>.
</para>
713 <term><constant>SIGRTMIN+
13</constant></term>
715 <listitem><para>Immediately halts the machine.
</para></listitem>
719 <term><constant>SIGRTMIN+
14</constant></term>
721 <listitem><para>Immediately powers off the machine.
</para></listitem>
725 <term><constant>SIGRTMIN+
15</constant></term>
727 <listitem><para>Immediately reboots the machine.
</para></listitem>
731 <term><constant>SIGRTMIN+
16</constant></term>
733 <listitem><para>Immediately reboots the machine with kexec.
</para></listitem>
737 <term><constant>SIGRTMIN+
20</constant></term>
739 <listitem><para>Enables display of status messages on the
740 console, as controlled via
741 <varname>systemd.show_status=
1</varname> on the kernel command
742 line.
</para></listitem>
746 <term><constant>SIGRTMIN+
21</constant></term>
748 <listitem><para>Disables display of
749 status messages on the console, as
751 <varname>systemd.show_status=
0</varname>
752 on the kernel command
753 line.
</para></listitem>
757 <term><constant>SIGRTMIN+
22</constant></term>
758 <term><constant>SIGRTMIN+
23</constant></term>
760 <listitem><para>Sets the log level to
<literal>debug
</literal>
761 (or
<literal>info
</literal> on
762 <constant>SIGRTMIN+
23</constant>), as controlled via
763 <varname>systemd.log_level=debug
</varname> (or
764 <varname>systemd.log_level=info
</varname> on
765 <constant>SIGRTMIN+
23</constant>) on the kernel command
766 line.
</para></listitem>
770 <term><constant>SIGRTMIN+
24</constant></term>
772 <listitem><para>Immediately exits the manager (only available
773 for --user instances).
</para></listitem>
777 <term><constant>SIGRTMIN+
26</constant></term>
778 <term><constant>SIGRTMIN+
27</constant></term>
779 <term><constant>SIGRTMIN+
28</constant></term>
781 <listitem><para>Sets the log target to
782 <literal>journal-or-kmsg
</literal> (or
783 <literal>console
</literal> on
784 <constant>SIGRTMIN+
27</constant>,
<literal>kmsg
</literal> on
785 <constant>SIGRTMIN+
28</constant>), as controlled via
786 <varname>systemd.log_target=journal-or-kmsg
</varname> (or
787 <varname>systemd.log_target=console
</varname> on
788 <constant>SIGRTMIN+
27</constant> or
789 <varname>systemd.log_target=kmsg
</varname> on
790 <constant>SIGRTMIN+
28</constant>) on the kernel command
791 line.
</para></listitem>
797 <title>Environment
</title>
799 <variablelist class='environment-variables'
>
801 <term><varname>$SYSTEMD_LOG_LEVEL
</varname></term>
802 <listitem><para>systemd reads the log level from this
803 environment variable. This can be overridden with
804 <option>--log-level=
</option>.
</para></listitem>
808 <term><varname>$SYSTEMD_LOG_TARGET
</varname></term>
809 <listitem><para>systemd reads the log target from this
810 environment variable. This can be overridden with
811 <option>--log-target=
</option>.
</para></listitem>
815 <term><varname>$SYSTEMD_LOG_COLOR
</varname></term>
816 <listitem><para>Controls whether systemd highlights important
817 log messages. This can be overridden with
818 <option>--log-color=
</option>.
</para></listitem>
822 <term><varname>$SYSTEMD_LOG_LOCATION
</varname></term>
823 <listitem><para>Controls whether systemd prints the code
824 location along with log messages. This can be overridden with
825 <option>--log-location=
</option>.
</para></listitem>
829 <term><varname>$XDG_CONFIG_HOME
</varname></term>
830 <term><varname>$XDG_CONFIG_DIRS
</varname></term>
831 <term><varname>$XDG_DATA_HOME
</varname></term>
832 <term><varname>$XDG_DATA_DIRS
</varname></term>
834 <listitem><para>The systemd user manager uses these variables
835 in accordance to the
<ulink
836 url=
"http://standards.freedesktop.org/basedir-spec/basedir-spec-latest.html">XDG
837 Base Directory specification
</ulink> to find its
838 configuration.
</para></listitem>
842 <term><varname>$SYSTEMD_UNIT_PATH
</varname></term>
844 <listitem><para>Controls where systemd looks for unit
845 files.
</para></listitem>
849 <term><varname>$SYSTEMD_SYSVINIT_PATH
</varname></term>
851 <listitem><para>Controls where systemd looks for SysV init
852 scripts.
</para></listitem>
856 <term><varname>$SYSTEMD_SYSVRCND_PATH
</varname></term>
858 <listitem><para>Controls where systemd looks for SysV init
859 script runlevel link farms.
</para></listitem>
863 <term><varname>$SYSTEMD_COLORS
</varname></term>
865 <listitem><para>The value must be a boolean. Controls whether colorized output should be
866 generated. This can be specified to override the decision that
<command>systemd
</command>
867 makes based on
<varname>$TERM
</varname> and what the console is connected to.
</para>
872 <term><varname>$LISTEN_PID
</varname></term>
873 <term><varname>$LISTEN_FDS
</varname></term>
874 <term><varname>$LISTEN_FDNAMES
</varname></term>
876 <listitem><para>Set by systemd for supervised processes during
877 socket-based activation. See
878 <citerefentry><refentrytitle>sd_listen_fds
</refentrytitle><manvolnum>3</manvolnum></citerefentry>
879 for more information.
</para></listitem>
883 <term><varname>$NOTIFY_SOCKET
</varname></term>
885 <listitem><para>Set by systemd for supervised processes for
886 status and start-up completion notification. See
887 <citerefentry><refentrytitle>sd_notify
</refentrytitle><manvolnum>3</manvolnum></citerefentry>
888 for more information.
</para></listitem>
894 <title>Kernel Command Line
</title>
896 <para>When run as system instance systemd parses a number of
897 kernel command line arguments
<footnote><para>If run inside a Linux
898 container these arguments may be passed as command line arguments
899 to systemd itself, next to any of the command line options listed
900 in the Options section above. If run outside of Linux containers,
901 these arguments are parsed from
<filename>/proc/cmdline
</filename>
902 instead.
</para></footnote>:
</para>
904 <variablelist class='kernel-commandline-options'
>
906 <term><varname>systemd.unit=
</varname></term>
907 <term><varname>rd.systemd.unit=
</varname></term>
909 <listitem><para>Overrides the unit to activate on boot.
910 Defaults to
<filename>default.target
</filename>. This may be
911 used to temporarily boot into a different boot unit, for
912 example
<filename>rescue.target
</filename> or
913 <filename>emergency.service
</filename>. See
914 <citerefentry><refentrytitle>systemd.special
</refentrytitle><manvolnum>7</manvolnum></citerefentry>
915 for details about these units. The option prefixed with
916 <literal>rd.
</literal> is honored only in the initial RAM disk
917 (initrd), while the one that is not prefixed only in the main
918 system.
</para></listitem>
922 <term><varname>systemd.dump_core
</varname></term>
924 <listitem><para>Takes a boolean argument or enables the option if specified
925 without an argument. If enabled, the systemd manager (PID
1) dumps core when
926 it crashes. Otherwise, no core dump is created. Defaults to enabled.
</para>
931 <term><varname>systemd.crash_chvt
</varname></term>
933 <listitem><para>Takes a positive integer, or a boolean argument. Can be also
934 specified without an argument, with the same effect as a positive boolean. If
935 a positive integer (in the range
1–
63) is specified, the system manager (PID
936 1) will activate the specified virtual terminal (VT) when it
937 crashes. Defaults to disabled, meaning that no such switch is attempted. If
938 set to enabled, the VT the kernel messages are written to is selected.
943 <term><varname>systemd.crash_shell
</varname></term>
945 <listitem><para>Takes a boolean argument or enables the option if specified
946 without an argument. If enabled, the system manager (PID
1) spawns a shell
947 when it crashes, after a
10s delay. Otherwise, no shell is spawned. Defaults
948 to disabled, for security reasons, as the shell is not protected by password
949 authentication.
</para></listitem>
953 <term><varname>systemd.crash_reboot
</varname></term>
955 <listitem><para>Takes a boolean argument or enables the option if specified
956 without an argument. If enabled, the system manager (PID
1) will reboot the
957 machine automatically when it crashes, after a
10s delay. Otherwise, the
958 system will hang indefinitely. Defaults to disabled, in order to avoid a
959 reboot loop. If combined with
<varname>systemd.crash_shell
</varname>, the
960 system is rebooted after the shell exits.
</para></listitem>
964 <term><varname>systemd.confirm_spawn
</varname></term>
966 <listitem><para>Takes a boolean argument or a path to the virtual console
967 where the confirmation messages should be emitted. Can be also specified
968 without an argument, with the same effect as a positive boolean. If enabled,
969 the system manager (PID
1) asks for confirmation when spawning processes
970 using
<option>/dev/console
</option>. If a path or a console name (such as
971 <literal>ttyS0
</literal>) is provided, the virtual console pointed to by this
972 path or described by the give name will be used instead. Defaults to disabled.
977 <term><varname>systemd.service_watchdogs=
</varname></term>
979 <listitem><para>Takes a boolean argument. If disabled, all service runtime
980 watchdogs (
<option>WatchdogSec=
</option>) and emergency actions (e.g.
981 <option>OnFailure=
</option> or
<option>StartLimitAction=
</option>) are
982 ignored by the system manager (PID
1); see
983 <citerefentry><refentrytitle>systemd.service
</refentrytitle><manvolnum>5</manvolnum></citerefentry>.
984 Defaults to enabled, i.e. watchdogs and failure actions are processed
985 normally. The hardware watchdog is not affected by this
986 option.
</para></listitem>
990 <term><varname>systemd.show_status
</varname></term>
992 <listitem><para>Takes a boolean argument or the constant
993 <constant>auto
</constant>. Can be also specified without an argument, with
994 the same effect as a positive boolean. If enabled, the systemd manager (PID
995 1) shows terse service status updates on the console during bootup.
996 <constant>auto
</constant> behaves like
<option>false
</option> until a unit
997 fails or there is a significant delay in boot. Defaults to enabled, unless
998 <option>quiet
</option> is passed as kernel command line option, in which case
999 it defaults to
<constant>auto
</constant>. If specified overrides the system
1000 manager configuration file option
<option>ShowStatus=
</option>, see
1001 <citerefentry><refentrytitle>systemd-system.conf
</refentrytitle><manvolnum>5</manvolnum></citerefentry>.
1002 However, the process command line option
<option>--show-status=
</option>
1003 takes precedence over both this kernel command line option and the
1004 configuration file option.
</para></listitem>
1008 <term><varname>systemd.log_target=
</varname></term>
1009 <term><varname>systemd.log_level=
</varname></term>
1010 <term><varname>systemd.log_location=
</varname></term>
1011 <term><varname>systemd.log_color
</varname></term>
1013 <listitem><para>Controls log output, with the same effect as the
1014 <varname>$SYSTEMD_LOG_TARGET
</varname>,
1015 <varname>$SYSTEMD_LOG_LEVEL
</varname>,
1016 <varname>$SYSTEMD_LOG_LOCATION
</varname>,
1017 <varname>$SYSTEMD_LOG_COLOR
</varname> environment variables described above.
1018 <varname>systemd.log_color
</varname> can be specified without an argument,
1019 with the same effect as a positive boolean.
</para></listitem>
1023 <term><varname>systemd.default_standard_output=
</varname></term>
1024 <term><varname>systemd.default_standard_error=
</varname></term>
1025 <listitem><para>Controls default standard output and error
1026 output for services, with the same effect as the
1027 <option>--default-standard-output=
</option> and
1028 <option>--default-standard-error=
</option> command line
1029 arguments described above, respectively.
</para></listitem>
1033 <term><varname>systemd.setenv=
</varname></term>
1035 <listitem><para>Takes a string argument in the form
1036 VARIABLE=VALUE. May be used to set default environment
1037 variables to add to forked child processes. May be used more
1038 than once to set multiple variables.
</para></listitem>
1042 <term><varname>systemd.machine_id=
</varname></term>
1044 <listitem><para>Takes a
32 character hex value to be
1045 used for setting the machine-id. Intended mostly for
1046 network booting where the same machine-id is desired
1047 for every boot.
</para></listitem>
1051 <term><varname>systemd.unified_cgroup_hierarchy
</varname></term>
1053 <listitem><para>When specified without an argument or with a true argument,
1054 enables the usage of
1055 <ulink url=
"https://www.kernel.org/doc/Documentation/cgroup-v2.txt">unified cgroup hierarchy
</ulink>
1056 (a.k.a. cgroups-v2). When specified with a false argument, fall back to
1057 hybrid or full legacy cgroup hierarchy.
</para>
1059 <para>If this option is not specified, the default behaviour is determined
1060 during compilation (the
<option>--with-default-hierarchy=
</option>
1061 option). If the kernel does not support unified cgroup hierarchy, the legacy
1062 hierarchy will be used even if this option is specified.
</para>
1067 <term><varname>systemd.legacy_systemd_cgroup_controller
</varname></term>
1069 <listitem><para>Takes effect if the full unified cgroup hierarchy is not used
1070 (see previous option). When specified without an argument or with a true
1071 argument, disables the use of
"hybrid" cgroup hierarchy (i.e. a cgroups-v2
1072 tree used for systemd, and
1073 <ulink url=
"https://www.kernel.org/doc/Documentation/cgroup-v1/">legacy
1074 cgroup hierarchy
</ulink>, a.k.a. cgroups-v1, for other controllers), and
1075 forces a full
"legacy" mode. When specified with a false argument, enables
1076 the use of
"hybrid" hierarchy.
</para>
1078 <para>If this option is not specified, the default behaviour is determined
1079 during compilation (the
<option>--with-default-hierarchy=
</option>
1080 option). If the kernel does not support unified cgroup hierarchy, the legacy
1081 hierarchy will be used even if this option is specified.
</para>
1086 <term><varname>quiet
</varname></term>
1088 <listitem><para>Turn off status output at boot, much like
1089 <varname>systemd.show_status=false
</varname> would. Note that
1090 this option is also read by the kernel itself and disables
1091 kernel log output. Passing this option hence turns off the
1092 usual output from both the system manager and the kernel.
1097 <term><varname>debug
</varname></term>
1099 <listitem><para>Turn on debugging output. This is equivalent
1100 to
<varname>systemd.log_level=debug
</varname>. Note that this
1101 option is also read by the kernel itself and enables kernel
1102 debug output. Passing this option hence turns on the debug
1103 output from both the system manager and the
1104 kernel.
</para></listitem>
1108 <term><varname>emergency
</varname></term>
1109 <term><varname>rd.emergency
</varname></term>
1110 <term><varname>-b
</varname></term>
1112 <listitem><para>Boot into emergency mode. This is equivalent
1113 to
<varname>systemd.unit=emergency.target
</varname> or
1114 <varname>rd.systemd.unit=emergency.target
</varname>, respectively, and
1115 provided for compatibility reasons and to be easier to type.
</para></listitem>
1119 <term><varname>rescue
</varname></term>
1120 <term><varname>rd.rescue
</varname></term>
1121 <term><varname>single
</varname></term>
1122 <term><varname>s
</varname></term>
1123 <term><varname>S
</varname></term>
1124 <term><varname>1</varname></term>
1126 <listitem><para>Boot into rescue mode. This is equivalent to
1127 <varname>systemd.unit=rescue.target
</varname> or
1128 <varname>rd.systemd.unit=rescue.target
</varname>, respectively, and
1129 provided for compatibility reasons and to be easier to type.
</para></listitem>
1133 <term><varname>2</varname></term>
1134 <term><varname>3</varname></term>
1135 <term><varname>4</varname></term>
1136 <term><varname>5</varname></term>
1138 <listitem><para>Boot into the specified legacy SysV runlevel.
1139 These are equivalent to
1140 <varname>systemd.unit=runlevel2.target
</varname>,
1141 <varname>systemd.unit=runlevel3.target
</varname>,
1142 <varname>systemd.unit=runlevel4.target
</varname>, and
1143 <varname>systemd.unit=runlevel5.target
</varname>,
1144 respectively, and provided for compatibility reasons and to be
1145 easier to type.
</para></listitem>
1149 <term><varname>locale.LANG=
</varname></term>
1150 <term><varname>locale.LANGUAGE=
</varname></term>
1151 <term><varname>locale.LC_CTYPE=
</varname></term>
1152 <term><varname>locale.LC_NUMERIC=
</varname></term>
1153 <term><varname>locale.LC_TIME=
</varname></term>
1154 <term><varname>locale.LC_COLLATE=
</varname></term>
1155 <term><varname>locale.LC_MONETARY=
</varname></term>
1156 <term><varname>locale.LC_MESSAGES=
</varname></term>
1157 <term><varname>locale.LC_PAPER=
</varname></term>
1158 <term><varname>locale.LC_NAME=
</varname></term>
1159 <term><varname>locale.LC_ADDRESS=
</varname></term>
1160 <term><varname>locale.LC_TELEPHONE=
</varname></term>
1161 <term><varname>locale.LC_MEASUREMENT=
</varname></term>
1162 <term><varname>locale.LC_IDENTIFICATION=
</varname></term>
1164 <listitem><para>Set the system locale to use. This overrides
1165 the settings in
<filename>/etc/locale.conf
</filename>. For
1166 more information, see
1167 <citerefentry project='man-pages'
><refentrytitle>locale.conf
</refentrytitle><manvolnum>5</manvolnum></citerefentry>
1169 <citerefentry project='man-pages'
><refentrytitle>locale
</refentrytitle><manvolnum>7</manvolnum></citerefentry>.
1174 <para>For other kernel command line parameters understood by
1175 components of the core OS, please refer to
1176 <citerefentry><refentrytitle>kernel-command-line
</refentrytitle><manvolnum>7</manvolnum></citerefentry>.
</para>
1180 <title>Sockets and FIFOs
</title>
1184 <term><filename>/run/systemd/notify
</filename></term>
1186 <listitem><para>Daemon status notification socket. This is an
1187 <constant>AF_UNIX
</constant> datagram socket and is used to
1188 implement the daemon notification logic as implemented by
1189 <citerefentry><refentrytitle>sd_notify
</refentrytitle><manvolnum>3</manvolnum></citerefentry>.
</para></listitem>
1194 <term><filename>/run/systemd/private
</filename></term>
1196 <listitem><para>Used internally as communication channel
1198 <citerefentry><refentrytitle>systemctl
</refentrytitle><manvolnum>1</manvolnum></citerefentry>
1199 and the systemd process. This is an
1200 <constant>AF_UNIX
</constant> stream socket. This interface is
1201 private to systemd and should not be used in external
1202 projects.
</para></listitem>
1206 <term><filename>/dev/initctl
</filename></term>
1208 <listitem><para>Limited compatibility support for the SysV
1209 client interface, as implemented by the
1210 <filename>systemd-initctl.service
</filename> unit. This is a
1211 named pipe in the file system. This interface is obsolete and
1212 should not be used in new applications.
</para></listitem>
1218 <title>See Also
</title>
1220 The
<ulink url=
"https://www.freedesktop.org/wiki/Software/systemd/">systemd Homepage
</ulink>,
1221 <citerefentry><refentrytitle>systemd-system.conf
</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
1222 <citerefentry project='man-pages'
><refentrytitle>locale.conf
</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
1223 <citerefentry><refentrytitle>systemctl
</refentrytitle><manvolnum>1</manvolnum></citerefentry>,
1224 <citerefentry><refentrytitle>journalctl
</refentrytitle><manvolnum>1</manvolnum></citerefentry>,
1225 <citerefentry><refentrytitle>systemd-notify
</refentrytitle><manvolnum>1</manvolnum></citerefentry>,
1226 <citerefentry><refentrytitle>daemon
</refentrytitle><manvolnum>7</manvolnum></citerefentry>,
1227 <citerefentry><refentrytitle>sd-daemon
</refentrytitle><manvolnum>3</manvolnum></citerefentry>,
1228 <citerefentry><refentrytitle>systemd.unit
</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
1229 <citerefentry><refentrytitle>systemd.special
</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
1230 <citerefentry project='die-net'
><refentrytitle>pkg-config
</refentrytitle><manvolnum>1</manvolnum></citerefentry>,
1231 <citerefentry><refentrytitle>kernel-command-line
</refentrytitle><manvolnum>7</manvolnum></citerefentry>,
1232 <citerefentry project='man-pages'
><refentrytitle>bootup
</refentrytitle><manvolnum>7</manvolnum></citerefentry>,
1233 <citerefentry><refentrytitle>systemd.directives
</refentrytitle><manvolnum>7</manvolnum></citerefentry>