Copyright 2010 Lennart Poettering
systemd is free software; you can redistribute it and/or modify it
- under the terms of the GNU General Public License as published by
- the Free Software Foundation; either version 2 of the License, or
+ under the terms of the GNU Lesser General Public License as published by
+ the Free Software Foundation; either version 2.1 of the License, or
(at your option) any later version.
systemd is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- General Public License for more details.
+ Lesser General Public License for more details.
- You should have received a copy of the GNU General Public License
+ You should have received a copy of the GNU Lesser General Public License
along with systemd; If not, see <http://www.gnu.org/licenses/>.
-->
<refnamediv>
<refname>systemd</refname>
<refname>init</refname>
- <refpurpose>systemd System and Session Manager</refpurpose>
+ <refpurpose>systemd System and Service Manager</refpurpose>
</refnamediv>
<refsynopsisdiv>
<refsect1>
<title>Description</title>
- <para>systemd is a system and session manager for
+ <para>systemd is a system and service manager for
Linux operating systems. When run as first process on
boot (as PID 1), it acts as init system that brings
up and maintains userspace services.</para>
<para>When run as system instance, systemd interprets
the configuration file
<filename>system.conf</filename>, otherwise
- <filename>session.conf</filename>. See
+ <filename>user.conf</filename>. See
<citerefentry><refentrytitle>systemd.conf</refentrytitle><manvolnum>5</manvolnum></citerefentry>
for more information.</para>
</refsect1>
</varlistentry>
<varlistentry>
<term><option>--system</option></term>
- <term><option>--session</option></term>
+ <term><option>--user</option></term>
<listitem><para>Tell systemd to run a
- system instance (resp. session
+ system instance (resp. user
instance), even if the process ID is
- not 1 (resp. is 1), i.e. system is not
- (resp. is) run as init process.
+ not 1 (resp. is 1), i.e. systemd is
+ not (resp. is) run as init process.
Normally it should not be necessary to
pass these options, as systemd
automatically detects the mode it is
started in. These options are hence of
- little use except for
- debugging.</para></listitem>
+ little use except for debugging. Note
+ that it is not supported booting and
+ maintaining a full system with systemd
+ running in <option>--system</option>
+ mode, but PID not 1. In practice,
+ passing <option>--system</option> explicitly is
+ only useful in conjunction with
+ <option>--test</option>.</para></listitem>
</varlistentry>
<varlistentry>
<term><option>--dump-core</option></term>
- <listitem><para>Dump core on crash. This switch has no effect when run as session instance.</para></listitem>
+ <listitem><para>Dump core on
+ crash. This switch has no effect when
+ run as user
+ instance.</para></listitem>
</varlistentry>
<varlistentry>
<term><option>--crash-shell</option></term>
- <listitem><para>Run shell on crash. This switch has no effect when run as session instance.</para></listitem>
+ <listitem><para>Run shell on
+ crash. This switch has no effect when
+ run as user
+ instance.</para></listitem>
</varlistentry>
<varlistentry>
<term><option>--confirm-spawn</option></term>
- <listitem><para>Ask for confirmation when spawning processes. This switch has no effect when run as session instance.</para></listitem>
+ <listitem><para>Ask for confirmation
+ when spawning processes. This switch
+ has no effect when run as user
+ instance.</para></listitem>
</varlistentry>
<varlistentry>
- <term><option>--show-status</option></term>
+ <term><option>--show-status=</option></term>
+
+ <listitem><para>Show terse service
+ status information while booting. This
+ switch has no effect when run as user
+ instance. Takes a boolean argument
+ which may be omitted which is
+ interpreted as
+ <option>true</option>.</para></listitem>
+ </varlistentry>
+ <varlistentry>
+ <term><option>--sysv-console=</option></term>
- <listitem><para>Show terse service status information while booting. This switch has no effect when run as session instance.</para></listitem>
+ <listitem><para>Controls whether
+ output of SysV init scripts will be
+ directed to the console. This switch
+ has no effect when run as user
+ instance. Takes a boolean argument
+ which may be omitted which is
+ interpreted as
+ <option>true</option>.</para></listitem>
</varlistentry>
<varlistentry>
<term><option>--log-target=</option></term>
<listitem><para>Set log
target. Argument must be one of
<option>console</option>,
+ <option>journal</option>,
<option>syslog</option>,
<option>kmsg</option>,
+ <option>journal-or-kmsg</option>,
<option>syslog-or-kmsg</option>,
<option>null</option>.</para></listitem>
</varlistentry>
it defaults to
<option>true</option>.</para></listitem>
</varlistentry>
+ <varlistentry>
+ <term><option>--default-standard-output=</option></term>
+ <term><option>--default-standard-error=</option></term>
+
+ <listitem><para>Sets the default
+ output resp. error output for all
+ services and sockets, i.e. controls
+ the default for
+ <option>StandardOutput=</option>
+ resp. <option>StandardError=</option>
+ (see
+ <citerefentry><refentrytitle>systemd.exec</refentrytitle><manvolnum>5</manvolnum></citerefentry>
+ for details). Takes one of
+ <option>inherit</option>,
+ <option>null</option>,
+ <option>tty</option>,
+ <option>journal</option>,
+ <option>journal+console</option>,
+ <option>syslog</option>,
+ <option>syslog+console</option>,
+ <option>kmsg</option>,
+ <option>kmsg+console</option>. If the
+ argument is omitted
+ <option>--default-standard-output=</option>
+ defaults to <option>journal</option>
+ and
+ <option>--default-standard-error=</option>
+ to
+ <option>inherit</option>.</para></listitem>
+ </varlistentry>
</variablelist>
</refsect1>
<para>systemd provides a dependency system between
various entities called "units". Units encapsulate
various objects that are relevant for system boot-up
- and maintainance. The majority of units are configured
+ and maintenance. The majority of units are configured
in unit configuration files, whose syntax and basic
set of options is described in
<citerefentry><refentrytitle>systemd.unit</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
however some are created automatically from other
configuration or dynamically from system state. Units
- may be active (meaning started, bound, plugged in, ...
- depending on the unit type), or inactive (meaning
- stopped, unbound, unplugged, ...), as well as in the
- process of being activated or deactivated,
- i.e. between the two states. The following unit types
- are available:</para>
+ may be 'active' (meaning started, bound, plugged in,
+ ... depending on the unit type, see below), or
+ 'inactive' (meaning stopped, unbound, unplugged, ...),
+ as well as in the process of being activated or
+ deactivated, i.e. between the two states (these states
+ are called 'activating', 'deactivating'). A special
+ 'failed' state is available as well which is very
+ similar to 'inactive' and is entered when the service
+ failed in some way (process returned error code on
+ exit, or crashed, or an operation timed out). If this
+ state is entered the cause will be logged, for later
+ reference. Note that the various unit types may have a
+ number of additional substates, which are mapped to
+ the five generalized unit states described
+ here.</para>
+
+ <para>The following unit types are available:</para>
<orderedlist>
<listitem><para>Service units, which control
systemd units, which later may be restored by
activating the saved snapshot unit. For more
information see
- <citerefentry><refentrytitle>systemd.automount</refentrytitle><manvolnum>5</manvolnum></citerefentry>.</para></listitem>
+ <citerefentry><refentrytitle>systemd.snapshot</refentrytitle><manvolnum>5</manvolnum></citerefentry>.</para></listitem>
<listitem><para>Timer units are useful for
triggering activation of other units based on
<citerefentry><refentrytitle>systemd.timer</refentrytitle><manvolnum>5</manvolnum></citerefentry>.</para></listitem>
<listitem><para>Swap units are very similar to
- mount units and encapsulated memory swap
+ mount units and encapsulate memory swap
partitions or files of the operating
- systemd. They are described in <citerefentry><refentrytitle>systemd.swap</refentrytitle><manvolnum>5</manvolnum></citerefentry>.</para></listitem>
+ system. They are described in <citerefentry><refentrytitle>systemd.swap</refentrytitle><manvolnum>5</manvolnum></citerefentry>.</para></listitem>
<listitem><para>Path units may be used
to activate other services when file system
<para>Units are named as their configuration
files. Some units have special semantics. A detailed
- list you may find in
+ list is available in
<citerefentry><refentrytitle>systemd.special</refentrytitle><manvolnum>7</manvolnum></citerefentry>.</para>
+ <para>systemd knows various kinds of dependencies,
+ including positive and negative requirement
+ dependencies (i.e. <varname>Requires=</varname> and
+ <varname>Conflicts=</varname>) as well as ordering
+ dependencies (<varname>After=</varname> and
+ <varname>Before=</varname>). NB: ordering and
+ requirement dependencies are orthogonal. If only a
+ requirement dependency exists between two units
+ (e.g. <filename>foo.service</filename> requires
+ <filename>bar.service</filename>), but no ordering
+ dependency (e.g. <filename>foo.service</filename>
+ after <filename>bar.service</filename>) and both are
+ requested to start, they will be started in
+ parallel. It is a common pattern that both requirement
+ and ordering dependencies are placed between two
+ units. Also note that the majority of dependencies are
+ implicitly created and maintained by systemd. In most
+ cases it should be unnecessary to declare additional
+ dependencies manually, however it is possible to do
+ this.</para>
+
+ <para>Application programs and units (via
+ dependencies) may request state changes of units. In
+ systemd, these requests are encapsulated as 'jobs' and
+ maintained in a job queue. Jobs may succeed or can
+ fail, their execution is ordered based on the ordering
+ dependencies of the units they have been scheduled
+ for.</para>
+
<para>On boot systemd activates the target unit
<filename>default.target</filename> whose job is to
activate on-boot services and other on-boot units by
track of processes. Control group information is
maintained in the kernel, and is accessible via the
file system hierarchy (beneath
- <filename>/cgroup/systemd/</filename>), or in tools
+ <filename>/sys/fs/cgroup/systemd/</filename>), or in tools
such as
<citerefentry><refentrytitle>ps</refentrytitle><manvolnum>1</manvolnum></citerefentry>
(<command>ps xawf -eo pid,user,cgroup,args</command>
ideas behind systemd please refer to the <ulink
url="http://0pointer.de/blog/projects/systemd.html">Original
Design Document</ulink>.</para>
+
+ <para>Note that some but not all interfaces provided
+ by systemd are covered by the <ulink
+ url="http://www.freedesktop.org/wiki/Software/systemd/InterfaceStabilityPromise">Interface
+ Stability Promise</ulink>.</para>
</refsect1>
<refsect1>
<command>pkg-config systemd
--variable=systemdsystemunitdir</command>. Other
directories checked are
- <filename>/usr/local/share/systemd/system</filename>
+ <filename>/usr/local/lib/systemd/system</filename>
and
- <filename>/usr/share/systemd/system</filename>. User
+ <filename>/usr/lib/systemd/system</filename>. User
configuration always takes
precedence. <command>pkg-config
systemd
--variable=systemdsystemconfdir</command>
returns the path of the system
configuration directory. Packages
- should alter the content of these directories
- only with the
- <citerefentry><refentrytitle>systemd-install</refentrytitle><manvolnum>1</manvolnum></citerefentry>
+ should alter the content of these
+ directories only with the
+ <command>enable</command> and
+ <command>disable</command> commands of
+ the
+ <citerefentry><refentrytitle>systemctl</refentrytitle><manvolnum>1</manvolnum></citerefentry>
tool.</para></listitem>
</varlistentry>
</variablelist>
<variablelist>
<varlistentry>
- <term>Session unit directories</term>
+ <term>User unit directories</term>
<listitem><para>Similar rules apply
- for the session unit
+ for the user unit
directories. However, here the <ulink
url="http://standards.freedesktop.org/basedir-spec/basedir-spec-latest.html">XDG
Base Directory specification</ulink>
units. Applications should place their
unit files in the directory returned
by <command>pkg-config systemd
- --variable=systemdsessionunitdir</command>. Global
- configuration is done in the
- directory reported by
- <command>pkg-config systemd
- --variable=systemdsessionconfdir</command>. The
- <citerefentry><refentrytitle>systemd-install</refentrytitle><manvolnum>1</manvolnum></citerefentry>
+ --variable=systemduserunitdir</command>. Global
+ configuration is done in the directory
+ reported by <command>pkg-config
+ systemd
+ --variable=systemduserconfdir</command>. The
+ <command>enable</command> and
+ <command>disable</command> commands of
+ the
+ <citerefentry><refentrytitle>systemctl</refentrytitle><manvolnum>1</manvolnum></citerefentry>
tool can handle both global (i.e. for
all users) and private (for one user)
enabling/disabling of
to <command>systemctl
daemon-reexec</command>.</para>
- <para>systemd session managers will
+ <para>systemd user managers will
start the
<filename>exit.target</filename> unit
when this signal is received. This is
mostly equivalent to
- <command>systemctl --session start
+ <command>systemctl --user start
exit.target</command>.</para></listitem>
</varlistentry>
<command>systemctl start
ctl-alt-del.target</command>.</para>
- <para>systemd session managers
+ <para>systemd user managers
treat this signal the same way as
SIGTERM.</para></listitem>
</varlistentry>
kbrequest.target</command>.</para>
<para>This signal is ignored by
- systemd session
+ systemd user
managers.</para></listitem>
</varlistentry>
<command>systemctl start
reboot.target</command>.</para></listitem>
</varlistentry>
+
+ <varlistentry>
+ <term>SIGRTMIN+6</term>
+
+ <listitem><para>Reboots the machine via kexec,
+ starts the
+ <filename>kexec.target</filename>
+ unit. This is mostly equivalent to
+ <command>systemctl start
+ kexec.target</command>.</para></listitem>
+ </varlistentry>
+
+ <varlistentry>
+ <term>SIGRTMIN+13</term>
+
+ <listitem><para>Immediately halts the machine.</para></listitem>
+ </varlistentry>
+
+ <varlistentry>
+ <term>SIGRTMIN+14</term>
+
+ <listitem><para>Immediately powers off the machine.</para></listitem>
+ </varlistentry>
+
+ <varlistentry>
+ <term>SIGRTMIN+15</term>
+
+ <listitem><para>Immediately reboots the machine.</para></listitem>
+ </varlistentry>
+
+ <varlistentry>
+ <term>SIGRTMIN+16</term>
+
+ <listitem><para>Immediately reboots the machine with kexec.</para></listitem>
+ </varlistentry>
+
+ <varlistentry>
+ <term>SIGRTMIN+20</term>
+
+ <listitem><para>Enables display of
+ status messages on the console, as
+ controlled via
+ <varname>systemd.show_status=1</varname>
+ on the kernel command
+ line.</para></listitem>
+ </varlistentry>
+
+ <varlistentry>
+ <term>SIGRTMIN+21</term>
+
+ <listitem><para>Disables display of
+ status messages on the console, as
+ controlled via
+ <varname>systemd.show_status=0</varname>
+ on the kernel command
+ line.</para></listitem>
+ </varlistentry>
+
+ <varlistentry>
+ <term>SIGRTMIN+22</term>
+ <term>SIGRTMIN+23</term>
+
+ <listitem><para>Sets the log level to
+ <literal>debug</literal>
+ (resp. <literal>info</literal> on
+ <literal>SIGRTMIN+23</literal>), as
+ controlled via
+ <varname>systemd.log_level=debug</varname>
+ (resp. <varname>systemd.log_level=info</varname>
+ on <literal>SIGRTMIN+23</literal>) on
+ the kernel command
+ line.</para></listitem>
+ </varlistentry>
+
+ <varlistentry>
+ <term>SIGRTMIN+26</term>
+ <term>SIGRTMIN+27</term>
+ <term>SIGRTMIN+28</term>
+ <term>SIGRTMIN+29</term>
+
+ <listitem><para>Sets the log level to
+ <literal>journal-or-kmsg</literal>
+ (resp. <literal>console</literal> on
+ <literal>SIGRTMIN+27</literal>;
+ resp. <literal>kmsg</literal> on
+ <literal>SIGRTMIN+28</literal>;
+ resp. <literal>syslog-or-kmsg</literal>
+ on <literal>SIGRTMIN+29</literal>), as
+ controlled via
+ <varname>systemd.log_target=journal-or-kmsg</varname>
+ (resp. <varname>systemd.log_target=console</varname>
+ on <literal>SIGRTMIN+27</literal>;
+ resp. <varname>systemd.log_target=kmsg</varname>
+ on <literal>SIGRTMIN+28</literal>;
+ resp
+ <varname>systemd.log_target=syslog-or-kmsg</varname>
+ on <literal>SIGRTMIN+29</literal>) on
+ the kernel command
+ line.</para></listitem>
+ </varlistentry>
</variablelist>
</refsect1>
<term><varname>$XDG_DATA_HOME</varname></term>
<term><varname>$XDG_DATA_DIRS</varname></term>
- <listitem><para>The systemd session
+ <listitem><para>The systemd user
manager uses these variables in
accordance to the <ulink
url="http://standards.freedesktop.org/basedir-spec/basedir-spec-latest.html">XDG
<refsect1>
<title>Kernel Command Line</title>
- <para>When run as system instance systemd parses a few kernel command line arguments:</para>
+ <para>When run as system instance systemd parses a
+ number of kernel command line
+ arguments<footnote><para>If run inside a Linux
+ container these arguments may be passed as command
+ line arguments to systemd itself, next to any of the
+ command line options listed in the Options section
+ above. If run outside of Linux containers, these
+ arguments are parsed from
+ <filename>/proc/cmdline</filename>
+ instead.</para></footnote>:</para>
<variablelist>
<varlistentry>
units.</para></listitem>
</varlistentry>
- <varlistentry>
- <term><varname>systemd.log_target=</varname></term>
- <term><varname>systemd.log_level=</varname></term>
- <term><varname>systemd.log_color=</varname></term>
- <term><varname>systemd.log_location=</varname></term>
-
- <listitem><para>Controls log output,
- with the same effect as the
- <varname>$SYSTEMD_LOG_TARGET</varname>, <varname>$SYSTEMD_LOG_LEVEL</varname>, <varname>$SYSTEMD_LOG_COLOR</varname>, <varname>$SYSTEMD_LOG_LOCATION</varname>
- environment variables described above.</para></listitem>
- </varlistentry>
-
<varlistentry>
<term><varname>systemd.dump_core=</varname></term>
<listitem><para>Takes a boolean
argument. If <option>true</option>
systemd spawns a shell when it
- crashes. Otherwise no core dump is
- created. Defaults to
+ crashes. Otherwise no shell is
+ spawned. Defaults to
<option>false</option>, for security
reasons, as the shell is not protected
by any password
<literal>-1</literal>.</para></listitem>
</varlistentry>
+ <varlistentry>
+ <term><varname>systemd.confirm_spawn=</varname></term>
+
+ <listitem><para>Takes a boolean
+ argument. If <option>true</option>
+ asks for confirmation when spawning
+ processes. Defaults to
+ <option>false</option>.</para></listitem>
+ </varlistentry>
+
<varlistentry>
<term><varname>systemd.show_status=</varname></term>
<option>true</option>.</para></listitem>
</varlistentry>
+ <varlistentry>
+ <term><varname>systemd.sysv_console=</varname></term>
+
+ <listitem><para>Takes a boolean
+ argument. If <option>true</option>
+ output of SysV init scripts will be
+ directed to the console. Defaults to
+ <option>true</option>, unless
+ <option>quiet</option> is passed as
+ kernel command line option in which
+ case it defaults to
+ <option>false</option>.</para></listitem>
+ </varlistentry>
+
+ <varlistentry>
+ <term><varname>systemd.log_target=</varname></term>
+ <term><varname>systemd.log_level=</varname></term>
+ <term><varname>systemd.log_color=</varname></term>
+ <term><varname>systemd.log_location=</varname></term>
+
+ <listitem><para>Controls log output,
+ with the same effect as the
+ <varname>$SYSTEMD_LOG_TARGET</varname>, <varname>$SYSTEMD_LOG_LEVEL</varname>, <varname>$SYSTEMD_LOG_COLOR</varname>, <varname>$SYSTEMD_LOG_LOCATION</varname>
+ environment variables described above.</para></listitem>
+ </varlistentry>
+
+ <varlistentry>
+ <term><varname>systemd.default_standard_output=</varname></term>
+ <term><varname>systemd.default_standard_error=</varname></term>
+ <listitem><para>Controls default
+ standard output/error output for
+ services, with the same effect as the
+ <option>--default-standard-output=</option>
+ resp. <option>--default-standard-error=</option>
+ command line arguments described
+ above.</para></listitem>
+ </varlistentry>
+
+ <varlistentry>
+ <term><varname>systemd.setenv=</varname></term>
+
+ <listitem><para>Takes a string
+ argument in the form
+ VARIABLE=VALUE. May be used to set
+ environment variables for the init
+ process and all its children at boot
+ time. May be used more than once to
+ set multiple variables. If the equal
+ sign and variable are missing unsets
+ an environment variable which might be
+ passed in from the initial ram
+ disk.</para></listitem>
+ </varlistentry>
+
</variablelist>
</refsect1>
<variablelist>
<varlistentry>
- <term><filename>@/org/freedesktop/systemd1/notify</filename></term>
+ <term><filename>/run/systemd/notify</filename></term>
<listitem><para>Daemon status
- notification socket. This is an AF_UNIX
- datagram socket in the Linux abstract
- namespace, and is used to implement
- the daemon notification logic as
- implemented by
+ notification socket. This is an
+ AF_UNIX datagram socket and is used to
+ implement the daemon notification
+ logic as implemented by
<citerefentry><refentrytitle>sd_notify</refentrytitle><manvolnum>3</manvolnum></citerefentry>.</para></listitem>
</varlistentry>
<varlistentry>
- <term><filename>@/org/freedesktop/systemd1/logger</filename></term>
+ <term><filename>/run/systemd/shutdownd</filename></term>
<listitem><para>Used internally by the
- <filename>systemd-logger.service</filename>
- unit to connect STDOUT and/or STDERR
- of spawned processes to
- <citerefentry><refentrytitle>syslog</refentrytitle><manvolnum>3</manvolnum></citerefentry>
- or the kernel log buffer. This is an
- AF_UNIX stream socket in the Linux
- abstract namespace.</para></listitem>
+ <citerefentry><refentrytitle>shutdown</refentrytitle><manvolnum>8</manvolnum></citerefentry>
+ tool to implement delayed
+ shutdowns. This is an AF_UNIX datagram
+ socket.</para></listitem>
</varlistentry>
<varlistentry>
- <term><filename>@/org/freedesktop/systemd1/private</filename></term>
+ <term><filename>/run/systemd/private</filename></term>
<listitem><para>Used internally as
communication channel between
<citerefentry><refentrytitle>systemctl</refentrytitle><manvolnum>1</manvolnum></citerefentry>
and the systemd process. This is an
- AF_UNIX stream socket in the Linux
- abstract namespace. This interface is
- private to systemd and should not be
- used in external
+ AF_UNIX stream socket. This interface
+ is private to systemd and should not
+ be used in external
projects.</para></listitem>
</varlistentry>
<para>
<citerefentry><refentrytitle>systemctl</refentrytitle><manvolnum>1</manvolnum></citerefentry>,
<citerefentry><refentrytitle>systemadm</refentrytitle><manvolnum>1</manvolnum></citerefentry>,
- <citerefentry><refentrytitle>systemd-install</refentrytitle><manvolnum>1</manvolnum></citerefentry>,
<citerefentry><refentrytitle>systemd-notify</refentrytitle><manvolnum>1</manvolnum></citerefentry>,
<citerefentry><refentrytitle>daemon</refentrytitle><manvolnum>7</manvolnum></citerefentry>,
<citerefentry><refentrytitle>sd-daemon</refentrytitle><manvolnum>7</manvolnum></citerefentry>,