blob: be62b6b572307cf1ec9f96f3ce0e2a227896e067 [file] [log] [blame]
<?xml version='1.0'?>
<!DOCTYPE refentry PUBLIC "-//OASIS//DTD DocBook XML V4.5//EN"
"" [
<!ENTITY % entities SYSTEM "custom-entities.ent" >
<!-- SPDX-License-Identifier: LGPL-2.1-or-later -->
<refentry id="logind.conf" conditional='ENABLE_LOGIND'
<refpurpose>Login manager configuration files</refpurpose>
<para>These files configure various parameters of the systemd login manager,
<citerefentry><refentrytitle>systemd-logind.service</refentrytitle><manvolnum>8</manvolnum></citerefentry>. See
for a general description of the syntax.</para>
<xi:include href="standard-conf.xml" xpointer="main-conf" />
<para>All options are configured in the
[Login] section:</para>
<variablelist class='config-directives'>
<listitem><para>Takes a positive integer. Configures how many
virtual terminals (VTs) to allocate by default that, when
switched to and are previously unused,
<literal>autovt</literal> services are automatically spawned
on. These services are instantiated from the template unit
<filename>autovt@.service</filename> for the respective VT TTY
name, for example, <filename>autovt@tty4.service</filename>.
By default, <filename>autovt@.service</filename> is linked to
<filename>getty@.service</filename>. In other words, login
prompts are started dynamically as the user switches to unused
virtual terminals. Hence, this parameter controls how many
login <literal>gettys</literal> are available on the VTs. If a
VT is already used by some other subsystem (for example, a
graphical login), this kind of activation will not be
attempted. Note that the VT configured in
<varname>ReserveVT=</varname> is always subject to this kind
of activation, even if it is not one of the VTs configured
with the <varname>NAutoVTs=</varname> directive. Defaults to
6. When set to 0, automatic spawning of
<literal>autovt</literal> services is
<listitem><para>Takes a positive integer. Identifies one
virtual terminal that shall unconditionally be reserved for
<filename>autovt@.service</filename> activation (see above).
The VT selected with this option will be marked busy
unconditionally, so that no other subsystem will allocate it.
This functionality is useful to ensure that, regardless of how
many VTs are allocated by other subsystems, one login
<literal>getty</literal> is always available. Defaults to 6
(in other words, there will always be a
<literal>getty</literal> available on Alt-F6.). When set to 0,
VT reservation is disabled.</para></listitem>
<listitem><para>Takes a boolean argument. Configures whether the processes of a
user should be killed when the user logs out. If true, the scope unit
corresponding to the session and all processes inside that scope will be
terminated. If false, the scope is "abandoned", see
and processes are not killed. Defaults to <literal>&KILL_USER_PROCESSES;</literal>,
but see the options <varname>KillOnlyUsers=</varname> and
<varname>KillExcludeUsers=</varname> below.</para>
<para>In addition to session processes, user process may run under the user
manager unit <filename>user@.service</filename>. Depending on the linger
settings, this may allow users to run processes independent of their login
sessions. See the description of <command>enable-linger</command> in
<para>Note that setting <varname>KillUserProcesses=yes</varname>
will break tools like
<citerefentry project='die-net'><refentrytitle>screen</refentrytitle><manvolnum>1</manvolnum></citerefentry>
<citerefentry project='die-net'><refentrytitle>tmux</refentrytitle><manvolnum>1</manvolnum></citerefentry>,
unless they are moved out of the session scope. See example in
<listitem><para>These settings take space-separated lists of usernames that override the
<varname>KillUserProcesses=</varname> setting. A user name may be added to
<varname>KillExcludeUsers=</varname> to exclude the processes in the session scopes of that user from
being killed even if <varname>KillUserProcesses=yes</varname> is set. If
<varname>KillExcludeUsers=</varname> is not set, the <literal>root</literal> user is excluded by
default. <varname>KillExcludeUsers=</varname> may be set to an empty value to override this
default. If a user is not excluded, <varname>KillOnlyUsers=</varname> is checked next. If this
setting is specified, only the processes in the session scopes of those users will be
killed. Otherwise, users are subject to the <varname>KillUserProcesses=yes</varname> setting.
<listitem><para>Configures the action to take when the system
is idle. Takes one of
<literal>suspend-then-hibernate</literal>, and
Defaults to <literal>ignore</literal>.</para>
<para>Note that this requires that user sessions correctly
report the idle status to the system. The system will execute
the action after all sessions report that they are idle, no
idle inhibitor lock is active, and subsequently, the time
configured with <varname>IdleActionSec=</varname> (see below)
has expired.</para>
<listitem><para>Configures the delay after which the action
configured in <varname>IdleAction=</varname> (see above) is
taken after the system is idle.</para></listitem>
<listitem><para>Specifies the maximum time a system shutdown
or sleep request is delayed due to an inhibitor lock of type
<literal>delay</literal> being active before the inhibitor is
ignored and the operation executes anyway. Defaults to
<listitem><para>Specifies how long to keep the user record and per-user service
<filename>user@.service</filename> around for a user after they logged out fully. If set to zero, the per-user
service is terminated immediately when the last session of the user has ended. If this option is configured to
non-zero rapid logout/login cycles are sped up, as the user's service manager is not constantly restarted. If
set to <literal>infinity</literal> the per-user service for a user is never terminated again after first login,
and continues to run until system shutdown. Defaults to 10s.</para></listitem>
<listitem><para>Controls how logind shall handle the
system power, reboot and sleep keys and the lid switch to trigger
actions such as system power-off, reboot or suspend. Can be one of
<literal>suspend-then-hibernate</literal>, and
If <literal>ignore</literal>, logind will never handle these
keys. If <literal>lock</literal>, all running sessions will be
screen-locked; otherwise, the specified action will be taken
in the respective event. Only input devices with the
<literal>power-switch</literal> udev tag will be watched for
key/lid switch events. <varname>HandlePowerKey=</varname>
defaults to <literal>poweroff</literal>, <varname>HandleRebootKey=</varname>
defaults to <literal>reboot</literal>.
<varname>HandleSuspendKey=</varname> and
<varname>HandleLidSwitch=</varname> default to
<varname>HandleLidSwitchExternalPower=</varname> is completely
ignored by default (for backwards compatibility) — an explicit
value must be set before it will be used to determine
behaviour. <varname>HandleLidSwitchDocked=</varname> defaults
to <literal>ignore</literal>.
<varname>HandleHibernateKey=</varname> defaults to
<literal>hibernate</literal>. If the system is inserted in a
docking station, or if more than one display is connected, the
action specified by <varname>HandleLidSwitchDocked=</varname>
occurs; if the system is on external power the action (if any)
specified by <varname>HandleLidSwitchExternalPower=</varname>
occurs; otherwise the <varname>HandleLidSwitch=</varname>
action occurs.</para>
<para>A different application may disable logind's handling of system power and
sleep keys and the lid switch by taking a low-level inhibitor lock
(<literal>handle-power-key</literal>, <literal>handle-suspend-key</literal>,
<literal>handle-hibernate-key</literal>, <literal>handle-lid-switch</literal>,
This is most commonly used by graphical desktop environments
to take over suspend and hibernation handling, and to use their own configuration
mechanisms. If a low-level inhibitor lock is taken, logind will not take any
action when that key or switch is triggered and the <varname>Handle*=</varname>
settings are irrelevant.</para></listitem>
<listitem><para>Controls whether actions that <command>systemd-logind</command>
takes when the power, reboot and sleep keys and the lid switch are triggered are subject
to high-level inhibitor locks ("shutdown", "reboot", "sleep", "idle"). Low level inhibitor
locks (<literal>handle-power-key</literal>, <literal>handle-suspend-key</literal>,
<literal>handle-hibernate-key</literal>, <literal>handle-lid-switch</literal>,
are always honored, irrespective of this setting.</para>
<para>These settings take boolean arguments. If <literal>no</literal>, the
inhibitor locks taken by applications are respected. If <literal>yes</literal>,
"shutdown", "reboot" "sleep", and "idle" inhibitor locks are ignored.
<varname>HibernateKeyIgnoreInhibited=</varname> and
<varname>RebootKeyIgnoreInhibited=</varname> default to <literal>no</literal>.
<varname>LidSwitchIgnoreInhibited=</varname> defaults to <literal>yes</literal>.
This means that when <command>systemd-logind</command> is handling events by
itself (no low level inhibitor locks are taken by another application), the lid
switch does not respect suspend blockers by default, but the power and sleep keys
<listitem><para>Specifies a period of time after system startup or
system resume in which systemd will hold off on reacting to
lid events. This is required for the system to properly
detect any hotplugged devices so systemd can ignore lid events
if external monitors, or docks, are connected. If set to 0,
systemd will always react immediately, possibly before the
kernel fully probed all hotplugged devices. This is safe, as
long as you do not care for systemd to account for devices
that have been plugged or unplugged while the system was off.
Defaults to 30s.</para></listitem>
<listitem><para>Sets the size limit on the
<varname>$XDG_RUNTIME_DIR</varname> runtime directory for each
user who logs in. Takes a size in bytes, optionally suffixed
with the usual K, G, M, and T suffixes, to the base 1024
(IEC). Alternatively, a numerical percentage suffixed by
<literal>%</literal> may be specified, which sets the size
limit relative to the amount of physical RAM. Defaults to 10%.
Note that this size is a safety limit only. As each runtime
directory is a tmpfs file system, it will only consume as much
memory as is needed.</para></listitem>
<listitem><para>Sets the limit on number of inodes for the
<varname>$XDG_RUNTIME_DIR</varname> runtime directory for each
user who logs in. Takes a number, optionally suffixed with the
usual K, G, M, and T suffixes, to the base 1024 (IEC).
Defaults to <varname>RuntimeDirectorySize=</varname> divided
by 4096. Note that this size is a safety limit only.
As each runtime directory is a tmpfs file system, it will
only consume as much memory as is needed.</para></listitem>
<listitem><para>Controls the maximum number of concurrent inhibitors to permit. Defaults to 8192
<listitem><para>Controls the maximum number of concurrent user sessions to manage. Defaults to 8192
(8K). Depending on how the <filename></filename> module is included in the PAM stack
configuration, further login sessions will either be refused, or permitted but not tracked by
<listitem><para>Controls whether System V and POSIX IPC objects belonging to the user shall be removed when the
user fully logs out. Takes a boolean argument. If enabled, the user may not consume IPC resources after the
last of the user's sessions terminated. This covers System V semaphores, shared memory and message queues, as
well as POSIX shared memory and message queues. Note that IPC objects of the root user and other system users
are excluded from the effect of this setting. Defaults to <literal>yes</literal>.</para></listitem>
<title>See Also</title>