chiark / gitweb /
selinux: check PID 1 label instead of /selinux mount point to figure out if selinux...
[elogind.git] / man / systemd.exec.xml
index 5b0d2ce37b74de68d5d355912b904c9913a32833..b9a37da38e061c3874caf61f287e1ac17fbea2d0 100644 (file)
         <refsect1>
                 <title>Description</title>
 
-                <para>Unit configuration files for services, sockets
+                <para>Unit configuration files for services, sockets,
                 mount points and swap devices share a subset of
                 configuration options which define the execution
                 environment of spawned processes.</para>
 
                 <para>This man page lists the configuration options
-                shared by these three unit types. See
+                shared by these four unit types. See
                 <citerefentry><refentrytitle>systemd.unit</refentrytitle><manvolnum>5</manvolnum></citerefentry>
                 for the common options of all unit configuration
                 files, and
                                 TTY (see above). Defaults to
                                 <filename>/dev/console</filename>.</para></listitem>
                         </varlistentry>
+                        <varlistentry>
+                                <term><varname>TTYReset=</varname></term>
+                                <listitem><para>Reset the terminal
+                                device specified with
+                                <varname>TTYPath=</varname> before and
+                                after execution. Defaults to
+                                <literal>no</literal>.</para></listitem>
+                        </varlistentry>
+                        <varlistentry>
+                                <term><varname>TTYVHangup=</varname></term>
+                                <listitem><para>Disconnect all clients
+                                which have opened the terminal device
+                                specified with
+                                <varname>TTYPath=</varname>
+                                before and after execution. Defaults
+                                to
+                                <literal>no</literal>.</para></listitem>
+                        </varlistentry>
+                        <varlistentry>
+                                <term><varname>TTYVTDisallocate=</varname></term>
+                                <listitem><para>If the the terminal
+                                device specified with
+                                <varname>TTYPath=</varname> is a
+                                virtual console terminal try to
+                                deallocate the TTY before and after
+                                execution. This ensures that the
+                                screen and scrollback buffer is
+                                cleared. Defaults to
+                                <literal>no</literal>.</para></listitem>
+                        </varlistentry>
                         <varlistentry>
                                 <term><varname>SyslogIdentifier=</varname></term>
                                 <listitem><para>Sets the process name
                                 for details.</para></listitem>
                         </varlistentry>
 
+                        <varlistentry>
+                                <term><varname>ControlGroupModify=</varname></term>
+                                <listitem><para>Takes a boolean
+                                argument. If true, the control groups
+                                created for this unit will be owned by
+                                ther user specified with
+                                <varname>User=</varname> (and the
+                                configured group), and he can create
+                                subgroups as well as add processes to
+                                the group.</para></listitem>
+                        </varlistentry>
+
                         <varlistentry>
                                 <term><varname>CapabilityBoundingSet=</varname></term>
 
                                 executed process. See
                                 <citerefentry><refentrytitle>capabilities</refentrytitle><manvolnum>7</manvolnum></citerefentry>
                                 for details. Takes a whitespace
-                                seperated list of capability names as
+                                separated list of capability names as
                                 read by
                                 <citerefentry><refentrytitle>cap_from_name</refentrytitle><manvolnum>3</manvolnum></citerefentry>.
                                 Capabilities listed will be included
                                 removed. If the list of capabilities
                                 is prefixed with ~ all but the listed
                                 capabilities will be included, the
-                                effect of this assignment
+                                effect of the assignment
                                 inverted. Note that this option does
                                 not actually set or unset any
                                 capabilities in the effective,
                                 path for this unit is implied. This
                                 option may be used to place executed
                                 processes in arbitrary groups in
-                                arbitrary hierachies -- which can be
+                                arbitrary hierarchies -- which can be
                                 configured externally with additional execution limits. By default
                                 systemd will place all executed
                                 processes in separate per-unit control
                                 usual file access controls would
                                 permit this. Directories listed in
                                 <varname>InaccessibleDirectories=</varname>
-                                will be made inaccesible for processes
+                                will be made inaccessible for processes
                                 inside the namespace. Note that
                                 restricting access with these options
                                 does not extend to submounts of a