]>
Commit | Line | Data |
---|---|---|
023a4f67 | 1 | <?xml version='1.0'?> <!--*- Mode: nxml; nxml-child-indent: 2; indent-tabs-mode: nil -*--> |
dd1eb43b | 2 | <!DOCTYPE refentry PUBLIC "-//OASIS//DTD DocBook XML V4.2//EN" |
12b42c76 | 3 | "http://www.oasis-open.org/docbook/xml/4.2/docbookx.dtd"> |
dd1eb43b LP |
4 | |
5 | <!-- | |
6 | This file is part of systemd. | |
7 | ||
8 | Copyright 2010 Lennart Poettering | |
9 | ||
10 | systemd is free software; you can redistribute it and/or modify it | |
5430f7f2 LP |
11 | under the terms of the GNU Lesser General Public License as published by |
12 | the Free Software Foundation; either version 2.1 of the License, or | |
dd1eb43b LP |
13 | (at your option) any later version. |
14 | ||
15 | systemd is distributed in the hope that it will be useful, but | |
16 | WITHOUT ANY WARRANTY; without even the implied warranty of | |
17 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU | |
5430f7f2 | 18 | Lesser General Public License for more details. |
dd1eb43b | 19 | |
5430f7f2 | 20 | You should have received a copy of the GNU Lesser General Public License |
dd1eb43b LP |
21 | along with systemd; If not, see <http://www.gnu.org/licenses/>. |
22 | --> | |
23 | ||
24 | <refentry id="systemd.exec"> | |
798d3a52 ZJS |
25 | <refentryinfo> |
26 | <title>systemd.exec</title> | |
27 | <productname>systemd</productname> | |
28 | ||
29 | <authorgroup> | |
30 | <author> | |
31 | <contrib>Developer</contrib> | |
32 | <firstname>Lennart</firstname> | |
33 | <surname>Poettering</surname> | |
34 | <email>lennart@poettering.net</email> | |
35 | </author> | |
36 | </authorgroup> | |
37 | </refentryinfo> | |
38 | ||
39 | <refmeta> | |
40 | <refentrytitle>systemd.exec</refentrytitle> | |
41 | <manvolnum>5</manvolnum> | |
42 | </refmeta> | |
43 | ||
44 | <refnamediv> | |
45 | <refname>systemd.exec</refname> | |
46 | <refpurpose>Execution environment configuration</refpurpose> | |
47 | </refnamediv> | |
48 | ||
49 | <refsynopsisdiv> | |
50 | <para><filename><replaceable>service</replaceable>.service</filename>, | |
51 | <filename><replaceable>socket</replaceable>.socket</filename>, | |
52 | <filename><replaceable>mount</replaceable>.mount</filename>, | |
53 | <filename><replaceable>swap</replaceable>.swap</filename></para> | |
54 | </refsynopsisdiv> | |
55 | ||
56 | <refsect1> | |
57 | <title>Description</title> | |
58 | ||
59 | <para>Unit configuration files for services, sockets, mount | |
60 | points, and swap devices share a subset of configuration options | |
61 | which define the execution environment of spawned | |
62 | processes.</para> | |
63 | ||
64 | <para>This man page lists the configuration options shared by | |
65 | these four unit types. See | |
66 | <citerefentry><refentrytitle>systemd.unit</refentrytitle><manvolnum>5</manvolnum></citerefentry> | |
67 | for the common options of all unit configuration files, and | |
68 | <citerefentry><refentrytitle>systemd.service</refentrytitle><manvolnum>5</manvolnum></citerefentry>, | |
69 | <citerefentry><refentrytitle>systemd.socket</refentrytitle><manvolnum>5</manvolnum></citerefentry>, | |
70 | <citerefentry><refentrytitle>systemd.swap</refentrytitle><manvolnum>5</manvolnum></citerefentry>, | |
71 | and | |
72 | <citerefentry><refentrytitle>systemd.mount</refentrytitle><manvolnum>5</manvolnum></citerefentry> | |
73 | for more information on the specific unit configuration files. The | |
74 | execution specific configuration options are configured in the | |
75 | [Service], [Socket], [Mount], or [Swap] sections, depending on the | |
76 | unit type.</para> | |
74b47bbd | 77 | |
c7458f93 | 78 | <para>In addition, options which control resources through Linux Control Groups (cgroups) are listed in |
74b47bbd ZJS |
79 | <citerefentry><refentrytitle>systemd.resource-control</refentrytitle><manvolnum>5</manvolnum></citerefentry>. |
80 | Those options complement options listed here.</para> | |
798d3a52 ZJS |
81 | </refsect1> |
82 | ||
c129bd5d LP |
83 | <refsect1> |
84 | <title>Automatic Dependencies</title> | |
85 | ||
86 | <para>A few execution parameters result in additional, automatic | |
87 | dependencies to be added.</para> | |
88 | ||
915e6d16 LP |
89 | <para>Units with <varname>WorkingDirectory=</varname>, <varname>RootDirectory=</varname> or |
90 | <varname>RootImage=</varname> set automatically gain dependencies of type <varname>Requires=</varname> and | |
91 | <varname>After=</varname> on all mount units required to access the specified paths. This is equivalent to having | |
92 | them listed explicitly in <varname>RequiresMountsFor=</varname>.</para> | |
c129bd5d | 93 | |
d71f0505 LP |
94 | <para>Similar, units with <varname>PrivateTmp=</varname> enabled automatically get mount unit dependencies for all |
95 | mounts required to access <filename>/tmp</filename> and <filename>/var/tmp</filename>. They will also gain an | |
96 | automatic <varname>After=</varname> dependency on | |
97 | <citerefentry><refentrytitle>systemd-tmpfiles-setup.service</refentrytitle><manvolnum>8</manvolnum></citerefentry>.</para> | |
c129bd5d | 98 | |
dfe85b38 LP |
99 | <para>Units whose standard output or error output is connected to <option>journal</option>, <option>syslog</option> |
100 | or <option>kmsg</option> (or their combinations with console output, see below) automatically acquire dependencies | |
101 | of type <varname>After=</varname> on <filename>systemd-journald.socket</filename>.</para> | |
c129bd5d LP |
102 | </refsect1> |
103 | ||
798d3a52 ZJS |
104 | <refsect1> |
105 | <title>Options</title> | |
106 | ||
107 | <variablelist class='unit-directives'> | |
108 | ||
109 | <varlistentry> | |
110 | <term><varname>WorkingDirectory=</varname></term> | |
111 | ||
d251207d LP |
112 | <listitem><para>Takes a directory path relative to the service's root directory specified by |
113 | <varname>RootDirectory=</varname>, or the special value <literal>~</literal>. Sets the working directory for | |
114 | executed processes. If set to <literal>~</literal>, the home directory of the user specified in | |
115 | <varname>User=</varname> is used. If not set, defaults to the root directory when systemd is running as a | |
116 | system instance and the respective user's home directory if run as user. If the setting is prefixed with the | |
117 | <literal>-</literal> character, a missing working directory is not considered fatal. If | |
915e6d16 LP |
118 | <varname>RootDirectory=</varname>/<varname>RootImage=</varname> is not set, then |
119 | <varname>WorkingDirectory=</varname> is relative to the root of the system running the service manager. Note | |
120 | that setting this parameter might result in additional dependencies to be added to the unit (see | |
121 | above).</para></listitem> | |
798d3a52 ZJS |
122 | </varlistentry> |
123 | ||
124 | <varlistentry> | |
125 | <term><varname>RootDirectory=</varname></term> | |
126 | ||
d251207d LP |
127 | <listitem><para>Takes a directory path relative to the host's root directory (i.e. the root of the system |
128 | running the service manager). Sets the root directory for executed processes, with the <citerefentry | |
129 | project='man-pages'><refentrytitle>chroot</refentrytitle><manvolnum>2</manvolnum></citerefentry> system | |
130 | call. If this is used, it must be ensured that the process binary and all its auxiliary files are available in | |
131 | the <function>chroot()</function> jail. Note that setting this parameter might result in additional | |
132 | dependencies to be added to the unit (see above).</para> | |
133 | ||
5d997827 LP |
134 | <para>The <varname>MountAPIVFS=</varname> and <varname>PrivateUsers=</varname> settings are particularly useful |
135 | in conjunction with <varname>RootDirectory=</varname>. For details, see below.</para></listitem> | |
136 | </varlistentry> | |
137 | ||
915e6d16 LP |
138 | <varlistentry> |
139 | <term><varname>RootImage=</varname></term> | |
140 | <listitem><para>Takes a path to a block device node or regular file as argument. This call is similar to | |
6cf5a964 | 141 | <varname>RootDirectory=</varname> however mounts a file system hierarchy from a block device node or loopback |
915e6d16 LP |
142 | file instead of a directory. The device node or file system image file needs to contain a file system without a |
143 | partition table, or a file system within an MBR/MS-DOS or GPT partition table with only a single | |
144 | Linux-compatible partition, or a set of file systems within a GPT partition table that follows the <ulink | |
28a0ad81 | 145 | url="https://www.freedesktop.org/wiki/Specifications/DiscoverablePartitionsSpec/">Discoverable Partitions |
915e6d16 LP |
146 | Specification</ulink>.</para></listitem> |
147 | </varlistentry> | |
148 | ||
5d997827 LP |
149 | <varlistentry> |
150 | <term><varname>MountAPIVFS=</varname></term> | |
151 | ||
152 | <listitem><para>Takes a boolean argument. If on, a private mount namespace for the unit's processes is created | |
ef3116b5 ZJS |
153 | and the API file systems <filename>/proc</filename>, <filename>/sys</filename>, and <filename>/dev</filename> |
154 | are mounted inside of it, unless they are already mounted. Note that this option has no effect unless used in | |
155 | conjunction with <varname>RootDirectory=</varname>/<varname>RootImage=</varname> as these three mounts are | |
156 | generally mounted in the host anyway, and unless the root directory is changed, the private mount namespace | |
157 | will be a 1:1 copy of the host's, and include these three mounts. Note that the <filename>/dev</filename> file | |
158 | system of the host is bind mounted if this option is used without <varname>PrivateDevices=</varname>. To run | |
159 | the service with a private, minimal version of <filename>/dev/</filename>, combine this option with | |
5d997827 | 160 | <varname>PrivateDevices=</varname>.</para></listitem> |
798d3a52 ZJS |
161 | </varlistentry> |
162 | ||
163 | <varlistentry> | |
164 | <term><varname>User=</varname></term> | |
165 | <term><varname>Group=</varname></term> | |
166 | ||
29206d46 | 167 | <listitem><para>Set the UNIX user or group that the processes are executed as, respectively. Takes a single |
565dab8e | 168 | user or group name, or a numeric ID as argument. For system services (services run by the system service manager, |
47da760e LP |
169 | i.e. managed by PID 1) and for user services of the root user (services managed by root's instance of |
170 | <command>systemd --user</command>), the default is <literal>root</literal>, but <varname>User=</varname> may be | |
171 | used to specify a different user. For user services of any other user, switching user identity is not | |
172 | permitted, hence the only valid setting is the same user the user's service manager is running as. If no group | |
173 | is set, the default group of the user is used. This setting does not affect commands whose command line is | |
565dab8e LP |
174 | prefixed with <literal>+</literal>.</para> |
175 | ||
176 | <para>Note that restrictions on the user/group name syntax are enforced: the specified name must consist only | |
177 | of the characters a-z, A-Z, 0-9, <literal>_</literal> and <literal>-</literal>, except for the first character | |
178 | which must be one of a-z, A-Z or <literal>_</literal> (i.e. numbers and <literal>-</literal> are not permitted | |
179 | as first character). The user/group name must have at least one character, and at most 31. These restrictions | |
180 | are enforced in order to avoid ambiguities and to ensure user/group names and unit files remain portable among | |
181 | Linux systems.</para> | |
182 | ||
183 | <para>When used in conjunction with <varname>DynamicUser=</varname> the user/group name specified is | |
184 | dynamically allocated at the time the service is started, and released at the time the service is stopped — | |
185 | unless it is already allocated statically (see below). If <varname>DynamicUser=</varname> is not used the | |
186 | specified user and group must have been created statically in the user database no later than the moment the | |
187 | service is started, for example using the | |
188 | <citerefentry><refentrytitle>sysusers.d</refentrytitle><manvolnum>5</manvolnum></citerefentry> facility, which | |
189 | is applied at boot or package install time.</para></listitem> | |
29206d46 LP |
190 | </varlistentry> |
191 | ||
192 | <varlistentry> | |
193 | <term><varname>DynamicUser=</varname></term> | |
194 | ||
195 | <listitem><para>Takes a boolean parameter. If set, a UNIX user and group pair is allocated dynamically when the | |
196 | unit is started, and released as soon as it is stopped. The user and group will not be added to | |
197 | <filename>/etc/passwd</filename> or <filename>/etc/group</filename>, but are managed transiently during | |
198 | runtime. The <citerefentry><refentrytitle>nss-systemd</refentrytitle><manvolnum>8</manvolnum></citerefentry> | |
199 | glibc NSS module provides integration of these dynamic users/groups into the system's user and group | |
200 | databases. The user and group name to use may be configured via <varname>User=</varname> and | |
201 | <varname>Group=</varname> (see above). If these options are not used and dynamic user/group allocation is | |
202 | enabled for a unit, the name of the dynamic user/group is implicitly derived from the unit name. If the unit | |
203 | name without the type suffix qualifies as valid user name it is used directly, otherwise a name incorporating a | |
204 | hash of it is used. If a statically allocated user or group of the configured name already exists, it is used | |
205 | and no dynamic user/group is allocated. Dynamic users/groups are allocated from the UID/GID range | |
206 | 61184…65519. It is recommended to avoid this range for regular system or login users. At any point in time | |
207 | each UID/GID from this range is only assigned to zero or one dynamically allocated users/groups in | |
208 | use. However, UID/GIDs are recycled after a unit is terminated. Care should be taken that any processes running | |
209 | as part of a unit for which dynamic users/groups are enabled do not leave files or directories owned by these | |
210 | users/groups around, as a different unit might get the same UID/GID assigned later on, and thus gain access to | |
63bb64a0 | 211 | these files or directories. If <varname>DynamicUser=</varname> is enabled, <varname>RemoveIPC=</varname>, |
00d9ef85 LP |
212 | <varname>PrivateTmp=</varname> are implied. This ensures that the lifetime of IPC objects and temporary files |
213 | created by the executed processes is bound to the runtime of the service, and hence the lifetime of the dynamic | |
214 | user/group. Since <filename>/tmp</filename> and <filename>/var/tmp</filename> are usually the only | |
215 | world-writable directories on a system this ensures that a unit making use of dynamic user/group allocation | |
63bb64a0 LP |
216 | cannot leave files around after unit termination. Moreover <varname>ProtectSystem=strict</varname> and |
217 | <varname>ProtectHome=read-only</varname> are implied, thus prohibiting the service to write to arbitrary file | |
218 | system locations. In order to allow the service to write to certain directories, they have to be whitelisted | |
cfaf4b75 | 219 | using <varname>ReadWritePaths=</varname>, but care must be taken so that UID/GID recycling doesn't |
63bb64a0 LP |
220 | create security issues involving files created by the service. Use <varname>RuntimeDirectory=</varname> (see |
221 | below) in order to assign a writable runtime directory to a service, owned by the dynamic user/group and | |
222 | removed automatically when the unit is terminated. Defaults to off.</para></listitem> | |
798d3a52 ZJS |
223 | </varlistentry> |
224 | ||
225 | <varlistentry> | |
226 | <term><varname>SupplementaryGroups=</varname></term> | |
227 | ||
228 | <listitem><para>Sets the supplementary Unix groups the | |
229 | processes are executed as. This takes a space-separated list | |
230 | of group names or IDs. This option may be specified more than | |
b938cb90 JE |
231 | once, in which case all listed groups are set as supplementary |
232 | groups. When the empty string is assigned, the list of | |
798d3a52 ZJS |
233 | supplementary groups is reset, and all assignments prior to |
234 | this one will have no effect. In any way, this option does not | |
235 | override, but extends the list of supplementary groups | |
236 | configured in the system group database for the | |
43eb109a | 237 | user. This does not affect commands prefixed with <literal>+</literal>.</para></listitem> |
798d3a52 ZJS |
238 | </varlistentry> |
239 | ||
00d9ef85 LP |
240 | <varlistentry> |
241 | <term><varname>RemoveIPC=</varname></term> | |
242 | ||
243 | <listitem><para>Takes a boolean parameter. If set, all System V and POSIX IPC objects owned by the user and | |
244 | group the processes of this unit are run as are removed when the unit is stopped. This setting only has an | |
245 | effect if at least one of <varname>User=</varname>, <varname>Group=</varname> and | |
246 | <varname>DynamicUser=</varname> are used. It has no effect on IPC objects owned by the root user. Specifically, | |
247 | this removes System V semaphores, as well as System V and POSIX shared memory segments and message queues. If | |
248 | multiple units use the same user or group the IPC objects are removed when the last of these units is | |
249 | stopped. This setting is implied if <varname>DynamicUser=</varname> is set.</para></listitem> | |
250 | </varlistentry> | |
251 | ||
798d3a52 ZJS |
252 | <varlistentry> |
253 | <term><varname>Nice=</varname></term> | |
254 | ||
255 | <listitem><para>Sets the default nice level (scheduling | |
256 | priority) for executed processes. Takes an integer between -20 | |
257 | (highest priority) and 19 (lowest priority). See | |
258 | <citerefentry><refentrytitle>setpriority</refentrytitle><manvolnum>2</manvolnum></citerefentry> | |
259 | for details.</para></listitem> | |
260 | </varlistentry> | |
261 | ||
262 | <varlistentry> | |
263 | <term><varname>OOMScoreAdjust=</varname></term> | |
264 | ||
265 | <listitem><para>Sets the adjustment level for the | |
266 | Out-Of-Memory killer for executed processes. Takes an integer | |
267 | between -1000 (to disable OOM killing for this process) and | |
268 | 1000 (to make killing of this process under memory pressure | |
269 | very likely). See <ulink | |
270 | url="https://www.kernel.org/doc/Documentation/filesystems/proc.txt">proc.txt</ulink> | |
271 | for details.</para></listitem> | |
272 | </varlistentry> | |
273 | ||
274 | <varlistentry> | |
275 | <term><varname>IOSchedulingClass=</varname></term> | |
276 | ||
b938cb90 | 277 | <listitem><para>Sets the I/O scheduling class for executed |
798d3a52 ZJS |
278 | processes. Takes an integer between 0 and 3 or one of the |
279 | strings <option>none</option>, <option>realtime</option>, | |
280 | <option>best-effort</option> or <option>idle</option>. See | |
281 | <citerefentry><refentrytitle>ioprio_set</refentrytitle><manvolnum>2</manvolnum></citerefentry> | |
282 | for details.</para></listitem> | |
283 | </varlistentry> | |
284 | ||
285 | <varlistentry> | |
286 | <term><varname>IOSchedulingPriority=</varname></term> | |
287 | ||
b938cb90 | 288 | <listitem><para>Sets the I/O scheduling priority for executed |
798d3a52 ZJS |
289 | processes. Takes an integer between 0 (highest priority) and 7 |
290 | (lowest priority). The available priorities depend on the | |
b938cb90 | 291 | selected I/O scheduling class (see above). See |
798d3a52 ZJS |
292 | <citerefentry><refentrytitle>ioprio_set</refentrytitle><manvolnum>2</manvolnum></citerefentry> |
293 | for details.</para></listitem> | |
294 | </varlistentry> | |
295 | ||
296 | <varlistentry> | |
297 | <term><varname>CPUSchedulingPolicy=</varname></term> | |
298 | ||
299 | <listitem><para>Sets the CPU scheduling policy for executed | |
300 | processes. Takes one of | |
301 | <option>other</option>, | |
302 | <option>batch</option>, | |
303 | <option>idle</option>, | |
304 | <option>fifo</option> or | |
305 | <option>rr</option>. See | |
306 | <citerefentry><refentrytitle>sched_setscheduler</refentrytitle><manvolnum>2</manvolnum></citerefentry> | |
307 | for details.</para></listitem> | |
308 | </varlistentry> | |
309 | ||
310 | <varlistentry> | |
311 | <term><varname>CPUSchedulingPriority=</varname></term> | |
312 | ||
313 | <listitem><para>Sets the CPU scheduling priority for executed | |
314 | processes. The available priority range depends on the | |
315 | selected CPU scheduling policy (see above). For real-time | |
316 | scheduling policies an integer between 1 (lowest priority) and | |
317 | 99 (highest priority) can be used. See | |
318 | <citerefentry><refentrytitle>sched_setscheduler</refentrytitle><manvolnum>2</manvolnum></citerefentry> | |
319 | for details. </para></listitem> | |
320 | </varlistentry> | |
321 | ||
322 | <varlistentry> | |
323 | <term><varname>CPUSchedulingResetOnFork=</varname></term> | |
324 | ||
325 | <listitem><para>Takes a boolean argument. If true, elevated | |
326 | CPU scheduling priorities and policies will be reset when the | |
327 | executed processes fork, and can hence not leak into child | |
328 | processes. See | |
329 | <citerefentry><refentrytitle>sched_setscheduler</refentrytitle><manvolnum>2</manvolnum></citerefentry> | |
330 | for details. Defaults to false.</para></listitem> | |
331 | </varlistentry> | |
332 | ||
333 | <varlistentry> | |
334 | <term><varname>CPUAffinity=</varname></term> | |
335 | ||
336 | <listitem><para>Controls the CPU affinity of the executed | |
71b1c27a FB |
337 | processes. Takes a list of CPU indices or ranges separated by |
338 | either whitespace or commas. CPU ranges are specified by the | |
339 | lower and upper CPU indices separated by a dash. | |
b938cb90 | 340 | This option may be specified more than once, in which case the |
798d3a52 ZJS |
341 | specified CPU affinity masks are merged. If the empty string |
342 | is assigned, the mask is reset, all assignments prior to this | |
343 | will have no effect. See | |
344 | <citerefentry><refentrytitle>sched_setaffinity</refentrytitle><manvolnum>2</manvolnum></citerefentry> | |
345 | for details.</para></listitem> | |
346 | </varlistentry> | |
347 | ||
348 | <varlistentry> | |
349 | <term><varname>UMask=</varname></term> | |
350 | ||
351 | <listitem><para>Controls the file mode creation mask. Takes an | |
352 | access mode in octal notation. See | |
353 | <citerefentry><refentrytitle>umask</refentrytitle><manvolnum>2</manvolnum></citerefentry> | |
354 | for details. Defaults to 0022.</para></listitem> | |
355 | </varlistentry> | |
356 | ||
357 | <varlistentry> | |
358 | <term><varname>Environment=</varname></term> | |
359 | ||
360 | <listitem><para>Sets environment variables for executed | |
361 | processes. Takes a space-separated list of variable | |
b938cb90 | 362 | assignments. This option may be specified more than once, in |
798d3a52 ZJS |
363 | which case all listed variables will be set. If the same |
364 | variable is set twice, the later setting will override the | |
365 | earlier setting. If the empty string is assigned to this | |
366 | option, the list of environment variables is reset, all prior | |
367 | assignments have no effect. Variable expansion is not | |
368 | performed inside the strings, however, specifier expansion is | |
369 | possible. The $ character has no special meaning. If you need | |
b8e485fa | 370 | to assign a value containing spaces or the equals sign to a variable, use double |
798d3a52 ZJS |
371 | quotes (") for the assignment.</para> |
372 | ||
373 | <para>Example: | |
374 | <programlisting>Environment="VAR1=word1 word2" VAR2=word3 "VAR3=$word 5 6"</programlisting> | |
375 | gives three variables <literal>VAR1</literal>, | |
376 | <literal>VAR2</literal>, <literal>VAR3</literal> | |
377 | with the values <literal>word1 word2</literal>, | |
378 | <literal>word3</literal>, <literal>$word 5 6</literal>. | |
379 | </para> | |
380 | ||
381 | <para> | |
382 | See | |
383 | <citerefentry project='man-pages'><refentrytitle>environ</refentrytitle><manvolnum>7</manvolnum></citerefentry> | |
384 | for details about environment variables.</para></listitem> | |
385 | </varlistentry> | |
386 | <varlistentry> | |
387 | <term><varname>EnvironmentFile=</varname></term> | |
388 | <listitem><para>Similar to <varname>Environment=</varname> but | |
389 | reads the environment variables from a text file. The text | |
390 | file should contain new-line-separated variable assignments. | |
8f0d2981 RM |
391 | Empty lines, lines without an <literal>=</literal> separator, |
392 | or lines starting with ; or # will be ignored, | |
798d3a52 ZJS |
393 | which may be used for commenting. A line ending with a |
394 | backslash will be concatenated with the following one, | |
395 | allowing multiline variable definitions. The parser strips | |
396 | leading and trailing whitespace from the values of | |
397 | assignments, unless you use double quotes (").</para> | |
398 | ||
399 | <para>The argument passed should be an absolute filename or | |
400 | wildcard expression, optionally prefixed with | |
401 | <literal>-</literal>, which indicates that if the file does | |
402 | not exist, it will not be read and no error or warning message | |
403 | is logged. This option may be specified more than once in | |
404 | which case all specified files are read. If the empty string | |
405 | is assigned to this option, the list of file to read is reset, | |
406 | all prior assignments have no effect.</para> | |
407 | ||
408 | <para>The files listed with this directive will be read | |
409 | shortly before the process is executed (more specifically, | |
410 | after all processes from a previous unit state terminated. | |
411 | This means you can generate these files in one unit state, and | |
f407824d DH |
412 | read it with this option in the next).</para> |
413 | ||
414 | <para>Settings from these | |
798d3a52 ZJS |
415 | files override settings made with |
416 | <varname>Environment=</varname>. If the same variable is set | |
417 | twice from these files, the files will be read in the order | |
418 | they are specified and the later setting will override the | |
419 | earlier setting.</para></listitem> | |
420 | </varlistentry> | |
421 | ||
b4c14404 FB |
422 | <varlistentry> |
423 | <term><varname>PassEnvironment=</varname></term> | |
424 | ||
425 | <listitem><para>Pass environment variables from the systemd system | |
426 | manager to executed processes. Takes a space-separated list of variable | |
427 | names. This option may be specified more than once, in which case all | |
428 | listed variables will be set. If the empty string is assigned to this | |
429 | option, the list of environment variables is reset, all prior | |
430 | assignments have no effect. Variables that are not set in the system | |
431 | manager will not be passed and will be silently ignored.</para> | |
432 | ||
433 | <para>Variables passed from this setting are overridden by those passed | |
434 | from <varname>Environment=</varname> or | |
435 | <varname>EnvironmentFile=</varname>.</para> | |
436 | ||
437 | <para>Example: | |
438 | <programlisting>PassEnvironment=VAR1 VAR2 VAR3</programlisting> | |
439 | passes three variables <literal>VAR1</literal>, | |
440 | <literal>VAR2</literal>, <literal>VAR3</literal> | |
441 | with the values set for those variables in PID1.</para> | |
442 | ||
443 | <para> | |
444 | See | |
445 | <citerefentry project='man-pages'><refentrytitle>environ</refentrytitle><manvolnum>7</manvolnum></citerefentry> | |
446 | for details about environment variables.</para></listitem> | |
447 | </varlistentry> | |
448 | ||
798d3a52 ZJS |
449 | <varlistentry> |
450 | <term><varname>StandardInput=</varname></term> | |
451 | <listitem><para>Controls where file descriptor 0 (STDIN) of | |
452 | the executed processes is connected to. Takes one of | |
453 | <option>null</option>, | |
454 | <option>tty</option>, | |
455 | <option>tty-force</option>, | |
52c239d7 LB |
456 | <option>tty-fail</option>, |
457 | <option>socket</option> or | |
458 | <option>fd</option>.</para> | |
798d3a52 ZJS |
459 | |
460 | <para>If <option>null</option> is selected, standard input | |
461 | will be connected to <filename>/dev/null</filename>, i.e. all | |
462 | read attempts by the process will result in immediate | |
463 | EOF.</para> | |
464 | ||
465 | <para>If <option>tty</option> is selected, standard input is | |
466 | connected to a TTY (as configured by | |
467 | <varname>TTYPath=</varname>, see below) and the executed | |
468 | process becomes the controlling process of the terminal. If | |
469 | the terminal is already being controlled by another process, | |
470 | the executed process waits until the current controlling | |
471 | process releases the terminal.</para> | |
472 | ||
473 | <para><option>tty-force</option> is similar to | |
474 | <option>tty</option>, but the executed process is forcefully | |
475 | and immediately made the controlling process of the terminal, | |
476 | potentially removing previous controlling processes from the | |
477 | terminal.</para> | |
478 | ||
479 | <para><option>tty-fail</option> is similar to | |
480 | <option>tty</option> but if the terminal already has a | |
481 | controlling process start-up of the executed process | |
482 | fails.</para> | |
483 | ||
484 | <para>The <option>socket</option> option is only valid in | |
485 | socket-activated services, and only when the socket | |
486 | configuration file (see | |
487 | <citerefentry><refentrytitle>systemd.socket</refentrytitle><manvolnum>5</manvolnum></citerefentry> | |
488 | for details) specifies a single socket only. If this option is | |
489 | set, standard input will be connected to the socket the | |
490 | service was activated from, which is primarily useful for | |
491 | compatibility with daemons designed for use with the | |
492 | traditional | |
b5c7d097 | 493 | <citerefentry project='freebsd'><refentrytitle>inetd</refentrytitle><manvolnum>8</manvolnum></citerefentry> |
798d3a52 ZJS |
494 | daemon.</para> |
495 | ||
52c239d7 LB |
496 | <para>The <option>fd</option> option connects |
497 | the input stream to a single file descriptor provided by a socket unit. | |
498 | A custom named file descriptor can be specified as part of this option, | |
499 | after a <literal>:</literal> (e.g. <literal>fd:<replaceable>foobar</replaceable></literal>). | |
500 | If no name is specified, <literal>stdin</literal> is assumed | |
501 | (i.e. <literal>fd</literal> is equivalent to <literal>fd:stdin</literal>). | |
502 | At least one socket unit defining such name must be explicitly provided via the | |
503 | <varname>Sockets=</varname> option, and file descriptor name may differ | |
504 | from the name of its containing socket unit. | |
505 | If multiple matches are found, the first one will be used. | |
506 | See <varname>FileDescriptorName=</varname> in | |
507 | <citerefentry><refentrytitle>systemd.socket</refentrytitle><manvolnum>5</manvolnum></citerefentry> | |
508 | for more details about named descriptors and ordering.</para> | |
509 | ||
798d3a52 ZJS |
510 | <para>This setting defaults to |
511 | <option>null</option>.</para></listitem> | |
512 | </varlistentry> | |
c129bd5d | 513 | |
798d3a52 ZJS |
514 | <varlistentry> |
515 | <term><varname>StandardOutput=</varname></term> | |
516 | <listitem><para>Controls where file descriptor 1 (STDOUT) of | |
517 | the executed processes is connected to. Takes one of | |
518 | <option>inherit</option>, | |
519 | <option>null</option>, | |
520 | <option>tty</option>, | |
521 | <option>journal</option>, | |
522 | <option>syslog</option>, | |
523 | <option>kmsg</option>, | |
524 | <option>journal+console</option>, | |
525 | <option>syslog+console</option>, | |
52c239d7 LB |
526 | <option>kmsg+console</option>, |
527 | <option>socket</option> or | |
528 | <option>fd</option>.</para> | |
798d3a52 ZJS |
529 | |
530 | <para><option>inherit</option> duplicates the file descriptor | |
531 | of standard input for standard output.</para> | |
532 | ||
533 | <para><option>null</option> connects standard output to | |
534 | <filename>/dev/null</filename>, i.e. everything written to it | |
535 | will be lost.</para> | |
536 | ||
537 | <para><option>tty</option> connects standard output to a tty | |
538 | (as configured via <varname>TTYPath=</varname>, see below). If | |
539 | the TTY is used for output only, the executed process will not | |
540 | become the controlling process of the terminal, and will not | |
541 | fail or wait for other processes to release the | |
542 | terminal.</para> | |
543 | ||
544 | <para><option>journal</option> connects standard output with | |
545 | the journal which is accessible via | |
546 | <citerefentry><refentrytitle>journalctl</refentrytitle><manvolnum>1</manvolnum></citerefentry>. | |
547 | Note that everything that is written to syslog or kmsg (see | |
548 | below) is implicitly stored in the journal as well, the | |
549 | specific two options listed below are hence supersets of this | |
550 | one.</para> | |
551 | ||
552 | <para><option>syslog</option> connects standard output to the | |
553 | <citerefentry project='man-pages'><refentrytitle>syslog</refentrytitle><manvolnum>3</manvolnum></citerefentry> | |
554 | system syslog service, in addition to the journal. Note that | |
555 | the journal daemon is usually configured to forward everything | |
556 | it receives to syslog anyway, in which case this option is no | |
557 | different from <option>journal</option>.</para> | |
558 | ||
559 | <para><option>kmsg</option> connects standard output with the | |
560 | kernel log buffer which is accessible via | |
561 | <citerefentry project='man-pages'><refentrytitle>dmesg</refentrytitle><manvolnum>1</manvolnum></citerefentry>, | |
562 | in addition to the journal. The journal daemon might be | |
563 | configured to send all logs to kmsg anyway, in which case this | |
564 | option is no different from <option>journal</option>.</para> | |
565 | ||
566 | <para><option>journal+console</option>, | |
567 | <option>syslog+console</option> and | |
568 | <option>kmsg+console</option> work in a similar way as the | |
569 | three options above but copy the output to the system console | |
570 | as well.</para> | |
571 | ||
572 | <para><option>socket</option> connects standard output to a | |
573 | socket acquired via socket activation. The semantics are | |
574 | similar to the same option of | |
575 | <varname>StandardInput=</varname>.</para> | |
576 | ||
52c239d7 LB |
577 | <para>The <option>fd</option> option connects |
578 | the output stream to a single file descriptor provided by a socket unit. | |
579 | A custom named file descriptor can be specified as part of this option, | |
580 | after a <literal>:</literal> (e.g. <literal>fd:<replaceable>foobar</replaceable></literal>). | |
581 | If no name is specified, <literal>stdout</literal> is assumed | |
582 | (i.e. <literal>fd</literal> is equivalent to <literal>fd:stdout</literal>). | |
583 | At least one socket unit defining such name must be explicitly provided via the | |
584 | <varname>Sockets=</varname> option, and file descriptor name may differ | |
585 | from the name of its containing socket unit. | |
586 | If multiple matches are found, the first one will be used. | |
587 | See <varname>FileDescriptorName=</varname> in | |
588 | <citerefentry><refentrytitle>systemd.socket</refentrytitle><manvolnum>5</manvolnum></citerefentry> | |
589 | for more details about named descriptors and ordering.</para> | |
590 | ||
dfe85b38 LP |
591 | <para>If the standard output (or error output, see below) of a unit is connected to the journal, syslog or the |
592 | kernel log buffer, the unit will implicitly gain a dependency of type <varname>After=</varname> on | |
28c75e25 LP |
593 | <filename>systemd-journald.socket</filename> (also see the automatic dependencies section above).</para> |
594 | ||
798d3a52 ZJS |
595 | <para>This setting defaults to the value set with |
596 | <option>DefaultStandardOutput=</option> in | |
597 | <citerefentry><refentrytitle>systemd-system.conf</refentrytitle><manvolnum>5</manvolnum></citerefentry>, | |
c129bd5d LP |
598 | which defaults to <option>journal</option>. Note that setting |
599 | this parameter might result in additional dependencies to be | |
600 | added to the unit (see above).</para></listitem> | |
798d3a52 | 601 | </varlistentry> |
c129bd5d | 602 | |
798d3a52 ZJS |
603 | <varlistentry> |
604 | <term><varname>StandardError=</varname></term> | |
605 | <listitem><para>Controls where file descriptor 2 (STDERR) of | |
606 | the executed processes is connected to. The available options | |
607 | are identical to those of <varname>StandardOutput=</varname>, | |
52c239d7 | 608 | with some exceptions: if set to <option>inherit</option> the |
798d3a52 | 609 | file descriptor used for standard output is duplicated for |
52c239d7 LB |
610 | standard error, while <option>fd</option> operates on the error |
611 | stream and will look by default for a descriptor named | |
612 | <literal>stderr</literal>.</para> | |
613 | ||
614 | <para>This setting defaults to the value set with | |
798d3a52 ZJS |
615 | <option>DefaultStandardError=</option> in |
616 | <citerefentry><refentrytitle>systemd-system.conf</refentrytitle><manvolnum>5</manvolnum></citerefentry>, | |
c129bd5d LP |
617 | which defaults to <option>inherit</option>. Note that setting |
618 | this parameter might result in additional dependencies to be | |
619 | added to the unit (see above).</para></listitem> | |
798d3a52 | 620 | </varlistentry> |
c129bd5d | 621 | |
798d3a52 ZJS |
622 | <varlistentry> |
623 | <term><varname>TTYPath=</varname></term> | |
624 | <listitem><para>Sets the terminal device node to use if | |
625 | standard input, output, or error are connected to a TTY (see | |
626 | above). Defaults to | |
627 | <filename>/dev/console</filename>.</para></listitem> | |
628 | </varlistentry> | |
629 | <varlistentry> | |
630 | <term><varname>TTYReset=</varname></term> | |
631 | <listitem><para>Reset the terminal device specified with | |
632 | <varname>TTYPath=</varname> before and after execution. | |
633 | Defaults to <literal>no</literal>.</para></listitem> | |
634 | </varlistentry> | |
635 | <varlistentry> | |
636 | <term><varname>TTYVHangup=</varname></term> | |
637 | <listitem><para>Disconnect all clients which have opened the | |
638 | terminal device specified with <varname>TTYPath=</varname> | |
639 | before and after execution. Defaults to | |
640 | <literal>no</literal>.</para></listitem> | |
641 | </varlistentry> | |
642 | <varlistentry> | |
643 | <term><varname>TTYVTDisallocate=</varname></term> | |
644 | <listitem><para>If the terminal device specified with | |
645 | <varname>TTYPath=</varname> is a virtual console terminal, try | |
646 | to deallocate the TTY before and after execution. This ensures | |
647 | that the screen and scrollback buffer is cleared. Defaults to | |
648 | <literal>no</literal>.</para></listitem> | |
649 | </varlistentry> | |
650 | <varlistentry> | |
651 | <term><varname>SyslogIdentifier=</varname></term> | |
652 | <listitem><para>Sets the process name to prefix log lines sent | |
653 | to the logging system or the kernel log buffer with. If not | |
654 | set, defaults to the process name of the executed process. | |
655 | This option is only useful when | |
656 | <varname>StandardOutput=</varname> or | |
657 | <varname>StandardError=</varname> are set to | |
658 | <option>syslog</option>, <option>journal</option> or | |
659 | <option>kmsg</option> (or to the same settings in combination | |
660 | with <option>+console</option>).</para></listitem> | |
661 | </varlistentry> | |
662 | <varlistentry> | |
663 | <term><varname>SyslogFacility=</varname></term> | |
664 | <listitem><para>Sets the syslog facility to use when logging | |
665 | to syslog. One of <option>kern</option>, | |
666 | <option>user</option>, <option>mail</option>, | |
667 | <option>daemon</option>, <option>auth</option>, | |
668 | <option>syslog</option>, <option>lpr</option>, | |
669 | <option>news</option>, <option>uucp</option>, | |
670 | <option>cron</option>, <option>authpriv</option>, | |
671 | <option>ftp</option>, <option>local0</option>, | |
672 | <option>local1</option>, <option>local2</option>, | |
673 | <option>local3</option>, <option>local4</option>, | |
674 | <option>local5</option>, <option>local6</option> or | |
675 | <option>local7</option>. See | |
676 | <citerefentry project='man-pages'><refentrytitle>syslog</refentrytitle><manvolnum>3</manvolnum></citerefentry> | |
677 | for details. This option is only useful when | |
678 | <varname>StandardOutput=</varname> or | |
679 | <varname>StandardError=</varname> are set to | |
680 | <option>syslog</option>. Defaults to | |
681 | <option>daemon</option>.</para></listitem> | |
682 | </varlistentry> | |
683 | <varlistentry> | |
684 | <term><varname>SyslogLevel=</varname></term> | |
a8eaaee7 | 685 | <listitem><para>The default syslog level to use when logging to |
798d3a52 ZJS |
686 | syslog or the kernel log buffer. One of |
687 | <option>emerg</option>, | |
688 | <option>alert</option>, | |
689 | <option>crit</option>, | |
690 | <option>err</option>, | |
691 | <option>warning</option>, | |
692 | <option>notice</option>, | |
693 | <option>info</option>, | |
694 | <option>debug</option>. See | |
695 | <citerefentry project='man-pages'><refentrytitle>syslog</refentrytitle><manvolnum>3</manvolnum></citerefentry> | |
696 | for details. This option is only useful when | |
697 | <varname>StandardOutput=</varname> or | |
698 | <varname>StandardError=</varname> are set to | |
699 | <option>syslog</option> or <option>kmsg</option>. Note that | |
700 | individual lines output by the daemon might be prefixed with a | |
701 | different log level which can be used to override the default | |
702 | log level specified here. The interpretation of these prefixes | |
703 | may be disabled with <varname>SyslogLevelPrefix=</varname>, | |
b938cb90 | 704 | see below. For details, see |
798d3a52 ZJS |
705 | <citerefentry><refentrytitle>sd-daemon</refentrytitle><manvolnum>3</manvolnum></citerefentry>. |
706 | ||
707 | Defaults to | |
708 | <option>info</option>.</para></listitem> | |
709 | </varlistentry> | |
710 | ||
711 | <varlistentry> | |
712 | <term><varname>SyslogLevelPrefix=</varname></term> | |
713 | <listitem><para>Takes a boolean argument. If true and | |
714 | <varname>StandardOutput=</varname> or | |
715 | <varname>StandardError=</varname> are set to | |
716 | <option>syslog</option>, <option>kmsg</option> or | |
717 | <option>journal</option>, log lines written by the executed | |
718 | process that are prefixed with a log level will be passed on | |
719 | to syslog with this log level set but the prefix removed. If | |
720 | set to false, the interpretation of these prefixes is disabled | |
721 | and the logged lines are passed on as-is. For details about | |
722 | this prefixing see | |
723 | <citerefentry><refentrytitle>sd-daemon</refentrytitle><manvolnum>3</manvolnum></citerefentry>. | |
724 | Defaults to true.</para></listitem> | |
725 | </varlistentry> | |
726 | ||
727 | <varlistentry> | |
728 | <term><varname>TimerSlackNSec=</varname></term> | |
729 | <listitem><para>Sets the timer slack in nanoseconds for the | |
730 | executed processes. The timer slack controls the accuracy of | |
731 | wake-ups triggered by timers. See | |
732 | <citerefentry><refentrytitle>prctl</refentrytitle><manvolnum>2</manvolnum></citerefentry> | |
733 | for more information. Note that in contrast to most other time | |
734 | span definitions this parameter takes an integer value in | |
735 | nano-seconds if no unit is specified. The usual time units are | |
736 | understood too.</para></listitem> | |
737 | </varlistentry> | |
738 | ||
739 | <varlistentry> | |
740 | <term><varname>LimitCPU=</varname></term> | |
741 | <term><varname>LimitFSIZE=</varname></term> | |
742 | <term><varname>LimitDATA=</varname></term> | |
743 | <term><varname>LimitSTACK=</varname></term> | |
744 | <term><varname>LimitCORE=</varname></term> | |
745 | <term><varname>LimitRSS=</varname></term> | |
746 | <term><varname>LimitNOFILE=</varname></term> | |
747 | <term><varname>LimitAS=</varname></term> | |
748 | <term><varname>LimitNPROC=</varname></term> | |
749 | <term><varname>LimitMEMLOCK=</varname></term> | |
750 | <term><varname>LimitLOCKS=</varname></term> | |
751 | <term><varname>LimitSIGPENDING=</varname></term> | |
752 | <term><varname>LimitMSGQUEUE=</varname></term> | |
753 | <term><varname>LimitNICE=</varname></term> | |
754 | <term><varname>LimitRTPRIO=</varname></term> | |
755 | <term><varname>LimitRTTIME=</varname></term> | |
29857001 LP |
756 | <listitem><para>Set soft and hard limits on various resources for executed processes. See |
757 | <citerefentry><refentrytitle>setrlimit</refentrytitle><manvolnum>2</manvolnum></citerefentry> for details on | |
758 | the resource limit concept. Resource limits may be specified in two formats: either as single value to set a | |
759 | specific soft and hard limit to the same value, or as colon-separated pair <option>soft:hard</option> to set | |
760 | both limits individually (e.g. <literal>LimitAS=4G:16G</literal>). Use the string <varname>infinity</varname> | |
761 | to configure no limit on a specific resource. The multiplicative suffixes K, M, G, T, P and E (to the base | |
762 | 1024) may be used for resource limits measured in bytes (e.g. LimitAS=16G). For the limits referring to time | |
763 | values, the usual time units ms, s, min, h and so on may be used (see | |
764 | <citerefentry><refentrytitle>systemd.time</refentrytitle><manvolnum>7</manvolnum></citerefentry> for | |
765 | details). Note that if no time unit is specified for <varname>LimitCPU=</varname> the default unit of seconds | |
766 | is implied, while for <varname>LimitRTTIME=</varname> the default unit of microseconds is implied. Also, note | |
767 | that the effective granularity of the limits might influence their enforcement. For example, time limits | |
768 | specified for <varname>LimitCPU=</varname> will be rounded up implicitly to multiples of 1s. For | |
769 | <varname>LimitNICE=</varname> the value may be specified in two syntaxes: if prefixed with <literal>+</literal> | |
770 | or <literal>-</literal>, the value is understood as regular Linux nice value in the range -20..19. If not | |
771 | prefixed like this the value is understood as raw resource limit parameter in the range 0..40 (with 0 being | |
772 | equivalent to 1).</para> | |
a4c18002 LP |
773 | |
774 | <para>Note that most process resource limits configured with | |
775 | these options are per-process, and processes may fork in order | |
776 | to acquire a new set of resources that are accounted | |
777 | independently of the original process, and may thus escape | |
778 | limits set. Also note that <varname>LimitRSS=</varname> is not | |
779 | implemented on Linux, and setting it has no effect. Often it | |
780 | is advisable to prefer the resource controls listed in | |
781 | <citerefentry><refentrytitle>systemd.resource-control</refentrytitle><manvolnum>5</manvolnum></citerefentry> | |
782 | over these per-process limits, as they apply to services as a | |
783 | whole, may be altered dynamically at runtime, and are | |
784 | generally more expressive. For example, | |
785 | <varname>MemoryLimit=</varname> is a more powerful (and | |
786 | working) replacement for <varname>LimitRSS=</varname>.</para> | |
798d3a52 | 787 | |
f4c9356d LP |
788 | <para>For system units these resource limits may be chosen freely. For user units however (i.e. units run by a |
789 | per-user instance of | |
790 | <citerefentry><refentrytitle>systemd</refentrytitle><manvolnum>1</manvolnum></citerefentry>), these limits are | |
791 | bound by (possibly more restrictive) per-user limits enforced by the OS.</para> | |
792 | ||
793 | <para>Resource limits not configured explicitly for a unit default to the value configured in the various | |
794 | <varname>DefaultLimitCPU=</varname>, <varname>DefaultLimitFSIZE=</varname>, … options available in | |
795 | <citerefentry><refentrytitle>systemd-system.conf</refentrytitle><manvolnum>5</manvolnum></citerefentry>, and – | |
796 | if not configured there – the kernel or per-user defaults, as defined by the OS (the latter only for user | |
797 | services, see above).</para> | |
798 | ||
798d3a52 | 799 | <table> |
f4c9356d | 800 | <title>Resource limit directives, their equivalent <command>ulimit</command> shell commands and the unit used</title> |
798d3a52 | 801 | |
a4c18002 | 802 | <tgroup cols='3'> |
798d3a52 ZJS |
803 | <colspec colname='directive' /> |
804 | <colspec colname='equivalent' /> | |
a4c18002 | 805 | <colspec colname='unit' /> |
798d3a52 ZJS |
806 | <thead> |
807 | <row> | |
808 | <entry>Directive</entry> | |
f4c9356d | 809 | <entry><command>ulimit</command> equivalent</entry> |
a4c18002 | 810 | <entry>Unit</entry> |
798d3a52 ZJS |
811 | </row> |
812 | </thead> | |
813 | <tbody> | |
814 | <row> | |
a4c18002 | 815 | <entry>LimitCPU=</entry> |
798d3a52 | 816 | <entry>ulimit -t</entry> |
a4c18002 | 817 | <entry>Seconds</entry> |
798d3a52 ZJS |
818 | </row> |
819 | <row> | |
a4c18002 | 820 | <entry>LimitFSIZE=</entry> |
798d3a52 | 821 | <entry>ulimit -f</entry> |
a4c18002 | 822 | <entry>Bytes</entry> |
798d3a52 ZJS |
823 | </row> |
824 | <row> | |
a4c18002 | 825 | <entry>LimitDATA=</entry> |
798d3a52 | 826 | <entry>ulimit -d</entry> |
a4c18002 | 827 | <entry>Bytes</entry> |
798d3a52 ZJS |
828 | </row> |
829 | <row> | |
a4c18002 | 830 | <entry>LimitSTACK=</entry> |
798d3a52 | 831 | <entry>ulimit -s</entry> |
a4c18002 | 832 | <entry>Bytes</entry> |
798d3a52 ZJS |
833 | </row> |
834 | <row> | |
a4c18002 | 835 | <entry>LimitCORE=</entry> |
798d3a52 | 836 | <entry>ulimit -c</entry> |
a4c18002 | 837 | <entry>Bytes</entry> |
798d3a52 ZJS |
838 | </row> |
839 | <row> | |
a4c18002 | 840 | <entry>LimitRSS=</entry> |
798d3a52 | 841 | <entry>ulimit -m</entry> |
a4c18002 | 842 | <entry>Bytes</entry> |
798d3a52 ZJS |
843 | </row> |
844 | <row> | |
a4c18002 | 845 | <entry>LimitNOFILE=</entry> |
798d3a52 | 846 | <entry>ulimit -n</entry> |
a4c18002 | 847 | <entry>Number of File Descriptors</entry> |
798d3a52 ZJS |
848 | </row> |
849 | <row> | |
a4c18002 | 850 | <entry>LimitAS=</entry> |
798d3a52 | 851 | <entry>ulimit -v</entry> |
a4c18002 | 852 | <entry>Bytes</entry> |
798d3a52 ZJS |
853 | </row> |
854 | <row> | |
a4c18002 | 855 | <entry>LimitNPROC=</entry> |
798d3a52 | 856 | <entry>ulimit -u</entry> |
a4c18002 | 857 | <entry>Number of Processes</entry> |
798d3a52 ZJS |
858 | </row> |
859 | <row> | |
a4c18002 | 860 | <entry>LimitMEMLOCK=</entry> |
798d3a52 | 861 | <entry>ulimit -l</entry> |
a4c18002 | 862 | <entry>Bytes</entry> |
798d3a52 ZJS |
863 | </row> |
864 | <row> | |
a4c18002 | 865 | <entry>LimitLOCKS=</entry> |
798d3a52 | 866 | <entry>ulimit -x</entry> |
a4c18002 | 867 | <entry>Number of Locks</entry> |
798d3a52 ZJS |
868 | </row> |
869 | <row> | |
a4c18002 | 870 | <entry>LimitSIGPENDING=</entry> |
798d3a52 | 871 | <entry>ulimit -i</entry> |
a4c18002 | 872 | <entry>Number of Queued Signals</entry> |
798d3a52 ZJS |
873 | </row> |
874 | <row> | |
a4c18002 | 875 | <entry>LimitMSGQUEUE=</entry> |
798d3a52 | 876 | <entry>ulimit -q</entry> |
a4c18002 | 877 | <entry>Bytes</entry> |
798d3a52 ZJS |
878 | </row> |
879 | <row> | |
a4c18002 | 880 | <entry>LimitNICE=</entry> |
798d3a52 | 881 | <entry>ulimit -e</entry> |
a4c18002 | 882 | <entry>Nice Level</entry> |
798d3a52 ZJS |
883 | </row> |
884 | <row> | |
a4c18002 | 885 | <entry>LimitRTPRIO=</entry> |
798d3a52 | 886 | <entry>ulimit -r</entry> |
a4c18002 | 887 | <entry>Realtime Priority</entry> |
798d3a52 ZJS |
888 | </row> |
889 | <row> | |
a4c18002 | 890 | <entry>LimitRTTIME=</entry> |
798d3a52 | 891 | <entry>No equivalent</entry> |
a4c18002 | 892 | <entry>Microseconds</entry> |
798d3a52 ZJS |
893 | </row> |
894 | </tbody> | |
895 | </tgroup> | |
a4c18002 | 896 | </table></listitem> |
798d3a52 ZJS |
897 | </varlistentry> |
898 | ||
899 | <varlistentry> | |
900 | <term><varname>PAMName=</varname></term> | |
9eb484fa LP |
901 | <listitem><para>Sets the PAM service name to set up a session as. If set, the executed process will be |
902 | registered as a PAM session under the specified service name. This is only useful in conjunction with the | |
903 | <varname>User=</varname> setting, and is otherwise ignored. If not set, no PAM session will be opened for the | |
904 | executed processes. See <citerefentry | |
905 | project='man-pages'><refentrytitle>pam</refentrytitle><manvolnum>8</manvolnum></citerefentry> for | |
906 | details.</para> | |
907 | ||
908 | <para>Note that for each unit making use of this option a PAM session handler process will be maintained as | |
909 | part of the unit and stays around as long as the unit is active, to ensure that appropriate actions can be | |
910 | taken when the unit and hence the PAM session terminates. This process is named <literal>(sd-pam)</literal> and | |
911 | is an immediate child process of the unit's main process.</para></listitem> | |
798d3a52 ZJS |
912 | </varlistentry> |
913 | ||
914 | <varlistentry> | |
915 | <term><varname>CapabilityBoundingSet=</varname></term> | |
916 | ||
479050b3 LP |
917 | <listitem><para>Controls which capabilities to include in the capability bounding set for the executed |
918 | process. See <citerefentry | |
919 | project='man-pages'><refentrytitle>capabilities</refentrytitle><manvolnum>7</manvolnum></citerefentry> for | |
b2656f1b LP |
920 | details. Takes a whitespace-separated list of capability names, e.g. <constant>CAP_SYS_ADMIN</constant>, |
921 | <constant>CAP_DAC_OVERRIDE</constant>, <constant>CAP_SYS_PTRACE</constant>. Capabilities listed will be | |
922 | included in the bounding set, all others are removed. If the list of capabilities is prefixed with | |
923 | <literal>~</literal>, all but the listed capabilities will be included, the effect of the assignment | |
924 | inverted. Note that this option also affects the respective capabilities in the effective, permitted and | |
925 | inheritable capability sets. If this option is not used, the capability bounding set is not modified on process | |
926 | execution, hence no limits on the capabilities of the process are enforced. This option may appear more than | |
927 | once, in which case the bounding sets are merged. If the empty string is assigned to this option, the bounding | |
928 | set is reset to the empty capability set, and all prior settings have no effect. If set to | |
929 | <literal>~</literal> (without any further argument), the bounding set is reset to the full set of available | |
930 | capabilities, also undoing any previous settings. This does not affect commands prefixed with | |
931 | <literal>+</literal>.</para></listitem> | |
798d3a52 ZJS |
932 | </varlistentry> |
933 | ||
ece87975 IP |
934 | <varlistentry> |
935 | <term><varname>AmbientCapabilities=</varname></term> | |
936 | ||
b2656f1b LP |
937 | <listitem><para>Controls which capabilities to include in the ambient capability set for the executed |
938 | process. Takes a whitespace-separated list of capability names, e.g. <constant>CAP_SYS_ADMIN</constant>, | |
939 | <constant>CAP_DAC_OVERRIDE</constant>, <constant>CAP_SYS_PTRACE</constant>. This option may appear more than | |
940 | once in which case the ambient capability sets are merged. If the list of capabilities is prefixed with | |
941 | <literal>~</literal>, all but the listed capabilities will be included, the effect of the assignment | |
942 | inverted. If the empty string is assigned to this option, the ambient capability set is reset to the empty | |
943 | capability set, and all prior settings have no effect. If set to <literal>~</literal> (without any further | |
944 | argument), the ambient capability set is reset to the full set of available capabilities, also undoing any | |
945 | previous settings. Note that adding capabilities to ambient capability set adds them to the process's inherited | |
946 | capability set. </para><para> Ambient capability sets are useful if you want to execute a process as a | |
947 | non-privileged user but still want to give it some capabilities. Note that in this case option | |
948 | <constant>keep-caps</constant> is automatically added to <varname>SecureBits=</varname> to retain the | |
949 | capabilities over the user change. <varname>AmbientCapabilities=</varname> does not affect commands prefixed | |
950 | with <literal>+</literal>.</para></listitem> | |
ece87975 IP |
951 | </varlistentry> |
952 | ||
798d3a52 ZJS |
953 | <varlistentry> |
954 | <term><varname>SecureBits=</varname></term> | |
955 | <listitem><para>Controls the secure bits set for the executed | |
956 | process. Takes a space-separated combination of options from | |
957 | the following list: | |
958 | <option>keep-caps</option>, | |
959 | <option>keep-caps-locked</option>, | |
960 | <option>no-setuid-fixup</option>, | |
961 | <option>no-setuid-fixup-locked</option>, | |
962 | <option>noroot</option>, and | |
963 | <option>noroot-locked</option>. | |
b938cb90 | 964 | This option may appear more than once, in which case the secure |
798d3a52 | 965 | bits are ORed. If the empty string is assigned to this option, |
43eb109a | 966 | the bits are reset to 0. This does not affect commands prefixed with <literal>+</literal>. |
cf677fe6 | 967 | See <citerefentry project='man-pages'><refentrytitle>capabilities</refentrytitle><manvolnum>7</manvolnum></citerefentry> |
798d3a52 ZJS |
968 | for details.</para></listitem> |
969 | </varlistentry> | |
970 | ||
798d3a52 | 971 | <varlistentry> |
2a624c36 AP |
972 | <term><varname>ReadWritePaths=</varname></term> |
973 | <term><varname>ReadOnlyPaths=</varname></term> | |
974 | <term><varname>InaccessiblePaths=</varname></term> | |
798d3a52 | 975 | |
effbd6d2 LP |
976 | <listitem><para>Sets up a new file system namespace for executed processes. These options may be used to limit |
977 | access a process might have to the file system hierarchy. Each setting takes a space-separated list of paths | |
978 | relative to the host's root directory (i.e. the system running the service manager). Note that if paths | |
979 | contain symlinks, they are resolved relative to the root directory set with | |
915e6d16 | 980 | <varname>RootDirectory=</varname>/<varname>RootImage=</varname>.</para> |
effbd6d2 LP |
981 | |
982 | <para>Paths listed in <varname>ReadWritePaths=</varname> are accessible from within the namespace with the same | |
983 | access modes as from outside of it. Paths listed in <varname>ReadOnlyPaths=</varname> are accessible for | |
984 | reading only, writing will be refused even if the usual file access controls would permit this. Nest | |
985 | <varname>ReadWritePaths=</varname> inside of <varname>ReadOnlyPaths=</varname> in order to provide writable | |
986 | subdirectories within read-only directories. Use <varname>ReadWritePaths=</varname> in order to whitelist | |
987 | specific paths for write access if <varname>ProtectSystem=strict</varname> is used. Paths listed in | |
988 | <varname>InaccessiblePaths=</varname> will be made inaccessible for processes inside the namespace (along with | |
989 | everything below them in the file system hierarchy).</para> | |
990 | ||
991 | <para>Note that restricting access with these options does not extend to submounts of a directory that are | |
992 | created later on. Non-directory paths may be specified as well. These options may be specified more than once, | |
993 | in which case all paths listed will have limited access from within the namespace. If the empty string is | |
994 | assigned to this option, the specific list is reset, and all prior assignments have no effect.</para> | |
995 | ||
e778185b | 996 | <para>Paths in <varname>ReadWritePaths=</varname>, <varname>ReadOnlyPaths=</varname> and |
5327c910 LP |
997 | <varname>InaccessiblePaths=</varname> may be prefixed with <literal>-</literal>, in which case they will be |
998 | ignored when they do not exist. If prefixed with <literal>+</literal> the paths are taken relative to the root | |
915e6d16 LP |
999 | directory of the unit, as configured with <varname>RootDirectory=</varname>/<varname>RootImage=</varname>, |
1000 | instead of relative to the root directory of the host (see above). When combining <literal>-</literal> and | |
1001 | <literal>+</literal> on the same path make sure to specify <literal>-</literal> first, and <literal>+</literal> | |
1002 | second.</para> | |
5327c910 LP |
1003 | |
1004 | <para>Note that using this setting will disconnect propagation of mounts from the service to the host | |
1005 | (propagation in the opposite direction continues to work). This means that this setting may not be used for | |
1006 | services which shall be able to install mount points in the main mount namespace. Note that the effect of these | |
1007 | settings may be undone by privileged processes. In order to set up an effective sandboxed environment for a | |
1008 | unit it is thus recommended to combine these settings with either | |
1009 | <varname>CapabilityBoundingSet=~CAP_SYS_ADMIN</varname> or | |
1010 | <varname>SystemCallFilter=~@mount</varname>.</para></listitem> | |
798d3a52 ZJS |
1011 | </varlistentry> |
1012 | ||
d2d6c096 LP |
1013 | <varlistentry> |
1014 | <term><varname>BindPaths=</varname></term> | |
1015 | <term><varname>BindReadOnlyPaths=</varname></term> | |
1016 | ||
1017 | <listitem><para>Configures unit-specific bind mounts. A bind mount makes a particular file or directory | |
1018 | available at an additional place in the unit's view of the file system. Any bind mounts created with this | |
1019 | option are specific to the unit, and are not visible in the host's mount table. This option expects a | |
1020 | whitespace separated list of bind mount definitions. Each definition consists of a colon-separated triple of | |
1021 | source path, destination path and option string, where the latter two are optional. If only a source path is | |
1022 | specified the source and destination is taken to be the same. The option string may be either | |
1023 | <literal>rbind</literal> or <literal>norbind</literal> for configuring a recursive or non-recursive bind | |
98063016 | 1024 | mount. If the destination path is omitted, the option string must be omitted too.</para> |
d2d6c096 LP |
1025 | |
1026 | <para><varname>BindPaths=</varname> creates regular writable bind mounts (unless the source file system mount | |
1027 | is already marked read-only), while <varname>BindReadOnlyPaths=</varname> creates read-only bind mounts. These | |
1028 | settings may be used more than once, each usage appends to the unit's list of bind mounts. If the empty string | |
1029 | is assigned to either of these two options the entire list of bind mounts defined prior to this is reset. Note | |
1030 | that in this case both read-only and regular bind mounts are reset, regardless which of the two settings is | |
1031 | used.</para> | |
1032 | ||
915e6d16 LP |
1033 | <para>This option is particularly useful when <varname>RootDirectory=</varname>/<varname>RootImage=</varname> |
1034 | is used. In this case the source path refers to a path on the host file system, while the destination path | |
1035 | refers to a path below the root directory of the unit.</para></listitem> | |
d2d6c096 LP |
1036 | </varlistentry> |
1037 | ||
798d3a52 ZJS |
1038 | <varlistentry> |
1039 | <term><varname>PrivateTmp=</varname></term> | |
1040 | ||
00d9ef85 LP |
1041 | <listitem><para>Takes a boolean argument. If true, sets up a new file system namespace for the executed |
1042 | processes and mounts private <filename>/tmp</filename> and <filename>/var/tmp</filename> directories inside it | |
1043 | that is not shared by processes outside of the namespace. This is useful to secure access to temporary files of | |
1044 | the process, but makes sharing between processes via <filename>/tmp</filename> or <filename>/var/tmp</filename> | |
1045 | impossible. If this is enabled, all temporary files created by a service in these directories will be removed | |
1046 | after the service is stopped. Defaults to false. It is possible to run two or more units within the same | |
1047 | private <filename>/tmp</filename> and <filename>/var/tmp</filename> namespace by using the | |
798d3a52 | 1048 | <varname>JoinsNamespaceOf=</varname> directive, see |
00d9ef85 | 1049 | <citerefentry><refentrytitle>systemd.unit</refentrytitle><manvolnum>5</manvolnum></citerefentry> for |
effbd6d2 LP |
1050 | details. This setting is implied if <varname>DynamicUser=</varname> is set. For this setting the same |
1051 | restrictions regarding mount propagation and privileges apply as for <varname>ReadOnlyPaths=</varname> and | |
d71f0505 LP |
1052 | related calls, see above. Enabling this setting has the side effect of adding <varname>Requires=</varname> and |
1053 | <varname>After=</varname> dependencies on all mount units necessary to access <filename>/tmp</filename> and | |
1054 | <filename>/var/tmp</filename>. Moreover an implicitly <varname>After=</varname> ordering on | |
1055 | <citerefentry><refentrytitle>systemd-tmpfiles-setup.service</refentrytitle><manvolnum>8</manvolnum></citerefentry> | |
b0238568 ZJS |
1056 | is added.</para> |
1057 | ||
1058 | <para>Note that the implementation of this setting might be impossible (for example if mount namespaces | |
1059 | are not available), and the unit should be written in a way that does not solely rely on this setting for | |
1060 | security.</para></listitem> | |
798d3a52 ZJS |
1061 | </varlistentry> |
1062 | ||
1063 | <varlistentry> | |
1064 | <term><varname>PrivateDevices=</varname></term> | |
1065 | ||
b0238568 ZJS |
1066 | <listitem><para>Takes a boolean argument. If true, sets up a new <filename>/dev</filename> mount for the |
1067 | executed processes and only adds API pseudo devices such as <filename>/dev/null</filename>, | |
1068 | <filename>/dev/zero</filename> or | |
effbd6d2 | 1069 | <filename>/dev/random</filename> (as well as the pseudo TTY subsystem) to it, but no physical devices such as |
9221aec8 DH |
1070 | <filename>/dev/sda</filename>, system memory <filename>/dev/mem</filename>, system ports |
1071 | <filename>/dev/port</filename> and others. This is useful to securely turn off physical device access by the | |
8f81a5f6 DH |
1072 | executed process. Defaults to false. Enabling this option will install a system call filter to block low-level |
1073 | I/O system calls that are grouped in the <varname>@raw-io</varname> set, will also remove | |
2cd0a735 DH |
1074 | <constant>CAP_MKNOD</constant> and <constant>CAP_SYS_RAWIO</constant> from the capability bounding set for |
1075 | the unit (see above), and set <varname>DevicePolicy=closed</varname> (see | |
798d3a52 | 1076 | <citerefentry><refentrytitle>systemd.resource-control</refentrytitle><manvolnum>5</manvolnum></citerefentry> |
effbd6d2 LP |
1077 | for details). Note that using this setting will disconnect propagation of mounts from the service to the host |
1078 | (propagation in the opposite direction continues to work). This means that this setting may not be used for | |
b0238568 ZJS |
1079 | services which shall be able to install mount points in the main mount namespace. The new <filename>/dev</filename> |
1080 | will be mounted read-only and 'noexec'. The latter may break old programs which try to set up executable memory by | |
effbd6d2 | 1081 | using <citerefentry><refentrytitle>mmap</refentrytitle><manvolnum>2</manvolnum></citerefentry> of |
2d35b79c YW |
1082 | <filename>/dev/zero</filename> instead of using <constant>MAP_ANON</constant>. For this setting the same restrictions |
1083 | regarding mount propagation and privileges apply as for <varname>ReadOnlyPaths=</varname> and related calls, see above. | |
a7db8614 | 1084 | If turned on and if running in user mode, or in system mode, but without the <constant>CAP_SYS_ADMIN</constant> |
2d35b79c | 1085 | capability (e.g. setting <varname>User=</varname>), <varname>NoNewPrivileges=yes</varname> is implied. |
b0238568 ZJS |
1086 | </para> |
1087 | ||
1088 | <para>Note that the implementation of this setting might be impossible (for example if mount namespaces | |
1089 | are not available), and the unit should be written in a way that does not solely rely on this setting for | |
1090 | security.</para></listitem> | |
798d3a52 ZJS |
1091 | </varlistentry> |
1092 | ||
1093 | <varlistentry> | |
1094 | <term><varname>PrivateNetwork=</varname></term> | |
1095 | ||
1096 | <listitem><para>Takes a boolean argument. If true, sets up a | |
1097 | new network namespace for the executed processes and | |
1098 | configures only the loopback network device | |
1099 | <literal>lo</literal> inside it. No other network devices will | |
1100 | be available to the executed process. This is useful to | |
b0238568 | 1101 | turn off network access by the executed process. |
798d3a52 ZJS |
1102 | Defaults to false. It is possible to run two or more units |
1103 | within the same private network namespace by using the | |
1104 | <varname>JoinsNamespaceOf=</varname> directive, see | |
1105 | <citerefentry><refentrytitle>systemd.unit</refentrytitle><manvolnum>5</manvolnum></citerefentry> | |
1106 | for details. Note that this option will disconnect all socket | |
1107 | families from the host, this includes AF_NETLINK and AF_UNIX. | |
1108 | The latter has the effect that AF_UNIX sockets in the abstract | |
1109 | socket namespace will become unavailable to the processes | |
1110 | (however, those located in the file system will continue to be | |
b0238568 ZJS |
1111 | accessible).</para> |
1112 | ||
1113 | <para>Note that the implementation of this setting might be impossible (for example if network namespaces | |
1114 | are not available), and the unit should be written in a way that does not solely rely on this setting for | |
1115 | security.</para></listitem> | |
798d3a52 ZJS |
1116 | </varlistentry> |
1117 | ||
1118 | <varlistentry> | |
d251207d LP |
1119 | <term><varname>PrivateUsers=</varname></term> |
1120 | ||
1121 | <listitem><para>Takes a boolean argument. If true, sets up a new user namespace for the executed processes and | |
1122 | configures a minimal user and group mapping, that maps the <literal>root</literal> user and group as well as | |
1123 | the unit's own user and group to themselves and everything else to the <literal>nobody</literal> user and | |
1124 | group. This is useful to securely detach the user and group databases used by the unit from the rest of the | |
1125 | system, and thus to create an effective sandbox environment. All files, directories, processes, IPC objects and | |
2dd67817 | 1126 | other resources owned by users/groups not equaling <literal>root</literal> or the unit's own will stay visible |
d251207d LP |
1127 | from within the unit but appear owned by the <literal>nobody</literal> user and group. If this mode is enabled, |
1128 | all unit processes are run without privileges in the host user namespace (regardless if the unit's own | |
1129 | user/group is <literal>root</literal> or not). Specifically this means that the process will have zero process | |
1130 | capabilities on the host's user namespace, but full capabilities within the service's user namespace. Settings | |
1131 | such as <varname>CapabilityBoundingSet=</varname> will affect only the latter, and there's no way to acquire | |
1132 | additional capabilities in the host's user namespace. Defaults to off.</para> | |
1133 | ||
915e6d16 LP |
1134 | <para>This setting is particularly useful in conjunction with |
1135 | <varname>RootDirectory=</varname>/<varname>RootImage=</varname>, as the need to synchronize the user and group | |
1136 | databases in the root directory and on the host is reduced, as the only users and groups who need to be matched | |
b0238568 ZJS |
1137 | are <literal>root</literal>, <literal>nobody</literal> and the unit's own user and group.</para> |
1138 | ||
1139 | <para>Note that the implementation of this setting might be impossible (for example if user namespaces | |
1140 | are not available), and the unit should be written in a way that does not solely rely on this setting for | |
1141 | security.</para></listitem> | |
d251207d LP |
1142 | </varlistentry> |
1143 | ||
798d3a52 ZJS |
1144 | <varlistentry> |
1145 | <term><varname>ProtectSystem=</varname></term> | |
1146 | ||
3f815163 LP |
1147 | <listitem><para>Takes a boolean argument or the special values <literal>full</literal> or |
1148 | <literal>strict</literal>. If true, mounts the <filename>/usr</filename> and <filename>/boot</filename> | |
1149 | directories read-only for processes invoked by this unit. If set to <literal>full</literal>, the | |
1150 | <filename>/etc</filename> directory is mounted read-only, too. If set to <literal>strict</literal> the entire | |
1151 | file system hierarchy is mounted read-only, except for the API file system subtrees <filename>/dev</filename>, | |
1152 | <filename>/proc</filename> and <filename>/sys</filename> (protect these directories using | |
1153 | <varname>PrivateDevices=</varname>, <varname>ProtectKernelTunables=</varname>, | |
1154 | <varname>ProtectControlGroups=</varname>). This setting ensures that any modification of the vendor-supplied | |
1155 | operating system (and optionally its configuration, and local mounts) is prohibited for the service. It is | |
1156 | recommended to enable this setting for all long-running services, unless they are involved with system updates | |
1157 | or need to modify the operating system in other ways. If this option is used, | |
effbd6d2 LP |
1158 | <varname>ReadWritePaths=</varname> may be used to exclude specific directories from being made read-only. This |
1159 | setting is implied if <varname>DynamicUser=</varname> is set. For this setting the same restrictions regarding | |
1160 | mount propagation and privileges apply as for <varname>ReadOnlyPaths=</varname> and related calls, see | |
1161 | above. Defaults to off.</para></listitem> | |
798d3a52 ZJS |
1162 | </varlistentry> |
1163 | ||
1164 | <varlistentry> | |
1165 | <term><varname>ProtectHome=</varname></term> | |
1166 | ||
effbd6d2 LP |
1167 | <listitem><para>Takes a boolean argument or <literal>read-only</literal>. If true, the directories |
1168 | <filename>/home</filename>, <filename>/root</filename> and <filename>/run/user</filename> are made inaccessible | |
1169 | and empty for processes invoked by this unit. If set to <literal>read-only</literal>, the three directories are | |
1170 | made read-only instead. It is recommended to enable this setting for all long-running services (in particular | |
1171 | network-facing ones), to ensure they cannot get access to private user data, unless the services actually | |
1172 | require access to the user's private data. This setting is implied if <varname>DynamicUser=</varname> is | |
1173 | set. For this setting the same restrictions regarding mount propagation and privileges apply as for | |
1174 | <varname>ReadOnlyPaths=</varname> and related calls, see above.</para></listitem> | |
59eeb84b LP |
1175 | </varlistentry> |
1176 | ||
1177 | <varlistentry> | |
1178 | <term><varname>ProtectKernelTunables=</varname></term> | |
1179 | ||
1180 | <listitem><para>Takes a boolean argument. If true, kernel variables accessible through | |
49accde7 DH |
1181 | <filename>/proc/sys</filename>, <filename>/sys</filename>, <filename>/proc/sysrq-trigger</filename>, |
1182 | <filename>/proc/latency_stats</filename>, <filename>/proc/acpi</filename>, | |
1183 | <filename>/proc/timer_stats</filename>, <filename>/proc/fs</filename> and <filename>/proc/irq</filename> will | |
525872bf LP |
1184 | be made read-only to all processes of the unit. Usually, tunable kernel variables should be initialized only at |
1185 | boot-time, for example with the | |
1186 | <citerefentry><refentrytitle>sysctl.d</refentrytitle><manvolnum>5</manvolnum></citerefentry> mechanism. Few | |
1187 | services need to write to these at runtime; it is hence recommended to turn this on for most services. For this | |
1188 | setting the same restrictions regarding mount propagation and privileges apply as for | |
1189 | <varname>ReadOnlyPaths=</varname> and related calls, see above. Defaults to off. If turned on and if running | |
1190 | in user mode, or in system mode, but without the <constant>CAP_SYS_ADMIN</constant> capability (e.g. services | |
1191 | for which <varname>User=</varname> is set), <varname>NoNewPrivileges=yes</varname> is implied. Note that this | |
1192 | option does not prevent indirect changes to kernel tunables effected by IPC calls to other processes. However, | |
1193 | <varname>InaccessiblePaths=</varname> may be used to make relevant IPC file system objects inaccessible. If | |
1194 | <varname>ProtectKernelTunables=</varname> is set, <varname>MountAPIVFS=yes</varname> is | |
1195 | implied.</para></listitem> | |
59eeb84b LP |
1196 | </varlistentry> |
1197 | ||
85265556 DH |
1198 | <varlistentry> |
1199 | <term><varname>ProtectKernelModules=</varname></term> | |
1200 | ||
1201 | <listitem><para>Takes a boolean argument. If true, explicit module loading will | |
1202 | be denied. This allows to turn off module load and unload operations on modular | |
1203 | kernels. It is recommended to turn this on for most services that do not need special | |
1204 | file systems or extra kernel modules to work. Default to off. Enabling this option | |
1205 | removes <constant>CAP_SYS_MODULE</constant> from the capability bounding set for | |
1206 | the unit, and installs a system call filter to block module system calls, | |
1207 | also <filename>/usr/lib/modules</filename> is made inaccessible. For this | |
1208 | setting the same restrictions regarding mount propagation and privileges | |
1209 | apply as for <varname>ReadOnlyPaths=</varname> and related calls, see above. | |
1210 | Note that limited automatic module loading due to user configuration or kernel | |
1211 | mapping tables might still happen as side effect of requested user operations, | |
1212 | both privileged and unprivileged. To disable module auto-load feature please see | |
1213 | <citerefentry><refentrytitle>sysctl.d</refentrytitle><manvolnum>5</manvolnum></citerefentry> | |
1214 | <constant>kernel.modules_disabled</constant> mechanism and | |
1215 | <filename>/proc/sys/kernel/modules_disabled</filename> documentation. | |
1216 | If turned on and if running in user mode, or in system mode, but without the <constant>CAP_SYS_ADMIN</constant> | |
1217 | capability (e.g. setting <varname>User=</varname>), <varname>NoNewPrivileges=yes</varname> | |
1218 | is implied. | |
1219 | </para></listitem> | |
1220 | </varlistentry> | |
1221 | ||
59eeb84b LP |
1222 | <varlistentry> |
1223 | <term><varname>ProtectControlGroups=</varname></term> | |
1224 | ||
effbd6d2 LP |
1225 | <listitem><para>Takes a boolean argument. If true, the Linux Control Groups (<citerefentry |
1226 | project='man-pages'><refentrytitle>cgroups</refentrytitle><manvolnum>7</manvolnum></citerefentry>) hierarchies | |
1227 | accessible through <filename>/sys/fs/cgroup</filename> will be made read-only to all processes of the | |
1228 | unit. Except for container managers no services should require write access to the control groups hierarchies; | |
1229 | it is hence recommended to turn this on for most services. For this setting the same restrictions regarding | |
1230 | mount propagation and privileges apply as for <varname>ReadOnlyPaths=</varname> and related calls, see | |
525872bf LP |
1231 | above. Defaults to off. If <varname>ProtectControlGroups=</varname> is set, <varname>MountAPIVFS=yes</varname> is |
1232 | implied.</para></listitem> | |
798d3a52 ZJS |
1233 | </varlistentry> |
1234 | ||
1235 | <varlistentry> | |
1236 | <term><varname>MountFlags=</varname></term> | |
1237 | ||
effbd6d2 LP |
1238 | <listitem><para>Takes a mount propagation flag: <option>shared</option>, <option>slave</option> or |
1239 | <option>private</option>, which control whether mounts in the file system namespace set up for this unit's | |
7141028d | 1240 | processes will receive or propagate mounts and unmounts. See <citerefentry |
effbd6d2 LP |
1241 | project='man-pages'><refentrytitle>mount</refentrytitle><manvolnum>2</manvolnum></citerefentry> for |
1242 | details. Defaults to <option>shared</option>. Use <option>shared</option> to ensure that mounts and unmounts | |
fa2a3966 IK |
1243 | are propagated from systemd's namespace to the service's namespace and vice versa. Use <option>slave</option> |
1244 | to run processes so that none of their mounts and unmounts will propagate to the host. Use <option>private</option> | |
374e6922 | 1245 | to also ensure that no mounts and unmounts from the host will propagate into the unit processes' namespace. |
4b957756 IK |
1246 | If this is set to <option>slave</option> or <option>private</option>, any mounts created by spawned processes |
1247 | will be unmounted after the completion of the current command line of <varname>ExecStartPre=</varname>, | |
1248 | <varname>ExecStartPost=</varname>, <varname>ExecStart=</varname>, | |
1249 | and <varname>ExecStopPost=</varname>. Note that | |
effbd6d2 LP |
1250 | <option>slave</option> means that file systems mounted on the host might stay mounted continuously in the |
1251 | unit's namespace, and thus keep the device busy. Note that the file system namespace related options | |
1252 | (<varname>PrivateTmp=</varname>, <varname>PrivateDevices=</varname>, <varname>ProtectSystem=</varname>, | |
1253 | <varname>ProtectHome=</varname>, <varname>ProtectKernelTunables=</varname>, | |
1254 | <varname>ProtectControlGroups=</varname>, <varname>ReadOnlyPaths=</varname>, | |
1255 | <varname>InaccessiblePaths=</varname>, <varname>ReadWritePaths=</varname>) require that mount and unmount | |
1256 | propagation from the unit's file system namespace is disabled, and hence downgrade <option>shared</option> to | |
798d3a52 ZJS |
1257 | <option>slave</option>. </para></listitem> |
1258 | </varlistentry> | |
1259 | ||
1260 | <varlistentry> | |
1261 | <term><varname>UtmpIdentifier=</varname></term> | |
1262 | ||
1263 | <listitem><para>Takes a four character identifier string for | |
023a4f67 LP |
1264 | an <citerefentry |
1265 | project='man-pages'><refentrytitle>utmp</refentrytitle><manvolnum>5</manvolnum></citerefentry> | |
1266 | and wtmp entry for this service. This should only be | |
1267 | set for services such as <command>getty</command> | |
1268 | implementations (such as <citerefentry | |
1269 | project='die-net'><refentrytitle>agetty</refentrytitle><manvolnum>8</manvolnum></citerefentry>) | |
798d3a52 | 1270 | where utmp/wtmp entries must be created and cleared before and |
023a4f67 LP |
1271 | after execution, or for services that shall be executed as if |
1272 | they were run by a <command>getty</command> process (see | |
1273 | below). If the configured string is longer than four | |
798d3a52 ZJS |
1274 | characters, it is truncated and the terminal four characters |
1275 | are used. This setting interprets %I style string | |
1276 | replacements. This setting is unset by default, i.e. no | |
1277 | utmp/wtmp entries are created or cleaned up for this | |
1278 | service.</para></listitem> | |
1279 | </varlistentry> | |
1280 | ||
023a4f67 LP |
1281 | <varlistentry> |
1282 | <term><varname>UtmpMode=</varname></term> | |
1283 | ||
1284 | <listitem><para>Takes one of <literal>init</literal>, | |
1285 | <literal>login</literal> or <literal>user</literal>. If | |
1286 | <varname>UtmpIdentifier=</varname> is set, controls which | |
1287 | type of <citerefentry | |
1288 | project='man-pages'><refentrytitle>utmp</refentrytitle><manvolnum>5</manvolnum></citerefentry>/wtmp | |
1289 | entries for this service are generated. This setting has no | |
1290 | effect unless <varname>UtmpIdentifier=</varname> is set | |
1291 | too. If <literal>init</literal> is set, only an | |
1292 | <constant>INIT_PROCESS</constant> entry is generated and the | |
6cd16034 LP |
1293 | invoked process must implement a |
1294 | <command>getty</command>-compatible utmp/wtmp logic. If | |
1295 | <literal>login</literal> is set, first an | |
a8eaaee7 | 1296 | <constant>INIT_PROCESS</constant> entry, followed by a |
6cd16034 | 1297 | <constant>LOGIN_PROCESS</constant> entry is generated. In |
b938cb90 | 1298 | this case, the invoked process must implement a <citerefentry |
023a4f67 LP |
1299 | project='die-net'><refentrytitle>login</refentrytitle><manvolnum>1</manvolnum></citerefentry>-compatible |
1300 | utmp/wtmp logic. If <literal>user</literal> is set, first an | |
1301 | <constant>INIT_PROCESS</constant> entry, then a | |
a8eaaee7 | 1302 | <constant>LOGIN_PROCESS</constant> entry and finally a |
023a4f67 | 1303 | <constant>USER_PROCESS</constant> entry is generated. In this |
b938cb90 | 1304 | case, the invoked process may be any process that is suitable |
023a4f67 LP |
1305 | to be run as session leader. Defaults to |
1306 | <literal>init</literal>.</para></listitem> | |
1307 | </varlistentry> | |
1308 | ||
798d3a52 ZJS |
1309 | <varlistentry> |
1310 | <term><varname>SELinuxContext=</varname></term> | |
1311 | ||
1312 | <listitem><para>Set the SELinux security context of the | |
1313 | executed process. If set, this will override the automated | |
1314 | domain transition. However, the policy still needs to | |
1315 | authorize the transition. This directive is ignored if SELinux | |
1316 | is disabled. If prefixed by <literal>-</literal>, all errors | |
43eb109a | 1317 | will be ignored. This does not affect commands prefixed with <literal>+</literal>. |
cf677fe6 | 1318 | See <citerefentry project='die-net'><refentrytitle>setexeccon</refentrytitle><manvolnum>3</manvolnum></citerefentry> |
798d3a52 ZJS |
1319 | for details.</para></listitem> |
1320 | </varlistentry> | |
1321 | ||
1322 | <varlistentry> | |
1323 | <term><varname>AppArmorProfile=</varname></term> | |
1324 | ||
1325 | <listitem><para>Takes a profile name as argument. The process | |
1326 | executed by the unit will switch to this profile when started. | |
1327 | Profiles must already be loaded in the kernel, or the unit | |
1328 | will fail. This result in a non operation if AppArmor is not | |
1329 | enabled. If prefixed by <literal>-</literal>, all errors will | |
43eb109a | 1330 | be ignored. This does not affect commands prefixed with <literal>+</literal>.</para></listitem> |
798d3a52 ZJS |
1331 | </varlistentry> |
1332 | ||
1333 | <varlistentry> | |
1334 | <term><varname>SmackProcessLabel=</varname></term> | |
1335 | ||
1336 | <listitem><para>Takes a <option>SMACK64</option> security | |
1337 | label as argument. The process executed by the unit will be | |
1338 | started under this label and SMACK will decide whether the | |
b938cb90 | 1339 | process is allowed to run or not, based on it. The process |
798d3a52 ZJS |
1340 | will continue to run under the label specified here unless the |
1341 | executable has its own <option>SMACK64EXEC</option> label, in | |
1342 | which case the process will transition to run under that | |
1343 | label. When not specified, the label that systemd is running | |
1344 | under is used. This directive is ignored if SMACK is | |
1345 | disabled.</para> | |
1346 | ||
1347 | <para>The value may be prefixed by <literal>-</literal>, in | |
1348 | which case all errors will be ignored. An empty value may be | |
cf677fe6 | 1349 | specified to unset previous assignments. This does not affect |
43eb109a | 1350 | commands prefixed with <literal>+</literal>.</para> |
798d3a52 ZJS |
1351 | </listitem> |
1352 | </varlistentry> | |
1353 | ||
1354 | <varlistentry> | |
1355 | <term><varname>IgnoreSIGPIPE=</varname></term> | |
1356 | ||
1357 | <listitem><para>Takes a boolean argument. If true, causes | |
1358 | <constant>SIGPIPE</constant> to be ignored in the executed | |
1359 | process. Defaults to true because <constant>SIGPIPE</constant> | |
1360 | generally is useful only in shell pipelines.</para></listitem> | |
1361 | </varlistentry> | |
1362 | ||
1363 | <varlistentry> | |
1364 | <term><varname>NoNewPrivileges=</varname></term> | |
1365 | ||
add00535 LP |
1366 | <listitem><para>Takes a boolean argument. If true, ensures that the service process and all its children can |
1367 | never gain new privileges through <function>execve()</function> (e.g. via setuid or setgid bits, or filesystem | |
1368 | capabilities). This is the simplest and most effective way to ensure that a process and its children can never | |
a7db8614 | 1369 | elevate privileges again. Defaults to false, but certain settings force |
add00535 LP |
1370 | <varname>NoNewPrivileges=yes</varname>, ignoring the value of this setting. This is the case when |
1371 | <varname>SystemCallFilter=</varname>, <varname>SystemCallArchitectures=</varname>, | |
1372 | <varname>RestrictAddressFamilies=</varname>, <varname>RestrictNamespaces=</varname>, | |
1373 | <varname>PrivateDevices=</varname>, <varname>ProtectKernelTunables=</varname>, | |
1374 | <varname>ProtectKernelModules=</varname>, <varname>MemoryDenyWriteExecute=</varname>, or | |
1375 | <varname>RestrictRealtime=</varname> are specified.</para></listitem> | |
798d3a52 ZJS |
1376 | </varlistentry> |
1377 | ||
1378 | <varlistentry> | |
1379 | <term><varname>SystemCallFilter=</varname></term> | |
1380 | ||
c79aff9a LP |
1381 | <listitem><para>Takes a space-separated list of system call names. If this setting is used, all system calls |
1382 | executed by the unit processes except for the listed ones will result in immediate process termination with the | |
1383 | <constant>SIGSYS</constant> signal (whitelisting). If the first character of the list is <literal>~</literal>, | |
1384 | the effect is inverted: only the listed system calls will result in immediate process termination | |
1385 | (blacklisting). If running in user mode, or in system mode, but without the <constant>CAP_SYS_ADMIN</constant> | |
1386 | capability (e.g. setting <varname>User=nobody</varname>), <varname>NoNewPrivileges=yes</varname> is | |
1387 | implied. This feature makes use of the Secure Computing Mode 2 interfaces of the kernel ('seccomp filtering') | |
1388 | and is useful for enforcing a minimal sandboxing environment. Note that the <function>execve</function>, | |
1389 | <function>exit</function>, <function>exit_group</function>, <function>getrlimit</function>, | |
1390 | <function>rt_sigreturn</function>, <function>sigreturn</function> system calls and the system calls for | |
1391 | querying time and sleeping are implicitly whitelisted and do not need to be listed explicitly. This option may | |
1392 | be specified more than once, in which case the filter masks are merged. If the empty string is assigned, the | |
1393 | filter is reset, all prior assignments will have no effect. This does not affect commands prefixed with | |
1394 | <literal>+</literal>.</para> | |
798d3a52 | 1395 | |
0b8fab97 LP |
1396 | <para>Note that on systems supporting multiple ABIs (such as x86/x86-64) it is recommended to turn off |
1397 | alternative ABIs for services, so that they cannot be used to circumvent the restrictions of this | |
1398 | option. Specifically, it is recommended to combine this option with | |
1399 | <varname>SystemCallArchitectures=native</varname> or similar.</para> | |
1400 | ||
2ca8dc15 LP |
1401 | <para>Note that strict system call filters may impact execution and error handling code paths of the service |
1402 | invocation. Specifically, access to the <function>execve</function> system call is required for the execution | |
1403 | of the service binary — if it is blocked service invocation will necessarily fail. Also, if execution of the | |
1404 | service binary fails for some reason (for example: missing service executable), the error handling logic might | |
1405 | require access to an additional set of system calls in order to process and log this failure correctly. It | |
1406 | might be necessary to temporarily disable system call filters in order to simplify debugging of such | |
1407 | failures.</para> | |
1408 | ||
798d3a52 ZJS |
1409 | <para>If you specify both types of this option (i.e. |
1410 | whitelisting and blacklisting), the first encountered will | |
1411 | take precedence and will dictate the default action | |
1412 | (termination or approval of a system call). Then the next | |
1413 | occurrences of this option will add or delete the listed | |
1414 | system calls from the set of the filtered system calls, | |
1415 | depending of its type and the default action. (For example, if | |
1416 | you have started with a whitelisting of | |
1417 | <function>read</function> and <function>write</function>, and | |
1418 | right after it add a blacklisting of | |
1419 | <function>write</function>, then <function>write</function> | |
201c1cc2 TM |
1420 | will be removed from the set.)</para> |
1421 | ||
1422 | <para>As the number of possible system | |
1423 | calls is large, predefined sets of system calls are provided. | |
1424 | A set starts with <literal>@</literal> character, followed by | |
1425 | name of the set. | |
1426 | ||
1427 | <table> | |
1428 | <title>Currently predefined system call sets</title> | |
1429 | ||
1430 | <tgroup cols='2'> | |
1431 | <colspec colname='set' /> | |
1432 | <colspec colname='description' /> | |
1433 | <thead> | |
1434 | <row> | |
1435 | <entry>Set</entry> | |
1436 | <entry>Description</entry> | |
1437 | </row> | |
1438 | </thead> | |
1439 | <tbody> | |
133ddbbe LP |
1440 | <row> |
1441 | <entry>@basic-io</entry> | |
1442 | <entry>System calls for basic I/O: reading, writing, seeking, file descriptor duplication and closing (<citerefentry project='man-pages'><refentrytitle>read</refentrytitle><manvolnum>2</manvolnum></citerefentry>, <citerefentry project='man-pages'><refentrytitle>write</refentrytitle><manvolnum>2</manvolnum></citerefentry>, and related calls)</entry> | |
1443 | </row> | |
201c1cc2 TM |
1444 | <row> |
1445 | <entry>@clock</entry> | |
1f9ac68b LP |
1446 | <entry>System calls for changing the system clock (<citerefentry project='man-pages'><refentrytitle>adjtimex</refentrytitle><manvolnum>2</manvolnum></citerefentry>, <citerefentry project='man-pages'><refentrytitle>settimeofday</refentrytitle><manvolnum>2</manvolnum></citerefentry>, and related calls)</entry> |
1447 | </row> | |
1448 | <row> | |
1449 | <entry>@cpu-emulation</entry> | |
1450 | <entry>System calls for CPU emulation functionality (<citerefentry project='man-pages'><refentrytitle>vm86</refentrytitle><manvolnum>2</manvolnum></citerefentry> and related calls)</entry> | |
1451 | </row> | |
1452 | <row> | |
1453 | <entry>@debug</entry> | |
1454 | <entry>Debugging, performance monitoring and tracing functionality (<citerefentry project='man-pages'><refentrytitle>ptrace</refentrytitle><manvolnum>2</manvolnum></citerefentry>, <citerefentry project='man-pages'><refentrytitle>perf_event_open</refentrytitle><manvolnum>2</manvolnum></citerefentry> and related calls)</entry> | |
201c1cc2 | 1455 | </row> |
1a1b13c9 LP |
1456 | <row> |
1457 | <entry>@file-system</entry> | |
1458 | <entry>File system operations: opening, creating files and directories for read and write, renaming and removing them, reading file properties, or creating hard and symbolic links.</entry> | |
1459 | </row> | |
201c1cc2 TM |
1460 | <row> |
1461 | <entry>@io-event</entry> | |
1f9ac68b | 1462 | <entry>Event loop system calls (<citerefentry project='man-pages'><refentrytitle>poll</refentrytitle><manvolnum>2</manvolnum></citerefentry>, <citerefentry project='man-pages'><refentrytitle>select</refentrytitle><manvolnum>2</manvolnum></citerefentry>, <citerefentry project='man-pages'><refentrytitle>epoll</refentrytitle><manvolnum>7</manvolnum></citerefentry>, <citerefentry project='man-pages'><refentrytitle>eventfd</refentrytitle><manvolnum>2</manvolnum></citerefentry> and related calls)</entry> |
201c1cc2 TM |
1463 | </row> |
1464 | <row> | |
1465 | <entry>@ipc</entry> | |
cd5bfd7e | 1466 | <entry>Pipes, SysV IPC, POSIX Message Queues and other IPC (<citerefentry project='man-pages'><refentrytitle>mq_overview</refentrytitle><manvolnum>7</manvolnum></citerefentry>, <citerefentry project='man-pages'><refentrytitle>svipc</refentrytitle><manvolnum>7</manvolnum></citerefentry>)</entry> |
1f9ac68b LP |
1467 | </row> |
1468 | <row> | |
1469 | <entry>@keyring</entry> | |
1470 | <entry>Kernel keyring access (<citerefentry project='man-pages'><refentrytitle>keyctl</refentrytitle><manvolnum>2</manvolnum></citerefentry> and related calls)</entry> | |
201c1cc2 TM |
1471 | </row> |
1472 | <row> | |
1473 | <entry>@module</entry> | |
d5efc18b | 1474 | <entry>Loading and unloading of kernel modules (<citerefentry project='man-pages'><refentrytitle>init_module</refentrytitle><manvolnum>2</manvolnum></citerefentry>, <citerefentry project='man-pages'><refentrytitle>delete_module</refentrytitle><manvolnum>2</manvolnum></citerefentry> and related calls)</entry> |
201c1cc2 TM |
1475 | </row> |
1476 | <row> | |
1477 | <entry>@mount</entry> | |
d5efc18b | 1478 | <entry>Mounting and unmounting of file systems (<citerefentry project='man-pages'><refentrytitle>mount</refentrytitle><manvolnum>2</manvolnum></citerefentry>, <citerefentry project='man-pages'><refentrytitle>chroot</refentrytitle><manvolnum>2</manvolnum></citerefentry>, and related calls)</entry> |
201c1cc2 TM |
1479 | </row> |
1480 | <row> | |
1481 | <entry>@network-io</entry> | |
1f9ac68b | 1482 | <entry>Socket I/O (including local AF_UNIX): <citerefentry project='man-pages'><refentrytitle>socket</refentrytitle><manvolnum>7</manvolnum></citerefentry>, <citerefentry project='man-pages'><refentrytitle>unix</refentrytitle><manvolnum>7</manvolnum></citerefentry></entry> |
201c1cc2 TM |
1483 | </row> |
1484 | <row> | |
1485 | <entry>@obsolete</entry> | |
1f9ac68b | 1486 | <entry>Unusual, obsolete or unimplemented (<citerefentry project='man-pages'><refentrytitle>create_module</refentrytitle><manvolnum>2</manvolnum></citerefentry>, <citerefentry project='man-pages'><refentrytitle>gtty</refentrytitle><manvolnum>2</manvolnum></citerefentry>, …)</entry> |
201c1cc2 TM |
1487 | </row> |
1488 | <row> | |
1489 | <entry>@privileged</entry> | |
1f9ac68b | 1490 | <entry>All system calls which need super-user capabilities (<citerefentry project='man-pages'><refentrytitle>capabilities</refentrytitle><manvolnum>7</manvolnum></citerefentry>)</entry> |
201c1cc2 TM |
1491 | </row> |
1492 | <row> | |
1493 | <entry>@process</entry> | |
d5efc18b | 1494 | <entry>Process control, execution, namespaceing operations (<citerefentry project='man-pages'><refentrytitle>clone</refentrytitle><manvolnum>2</manvolnum></citerefentry>, <citerefentry project='man-pages'><refentrytitle>kill</refentrytitle><manvolnum>2</manvolnum></citerefentry>, <citerefentry project='man-pages'><refentrytitle>namespaces</refentrytitle><manvolnum>7</manvolnum></citerefentry>, …</entry> |
201c1cc2 TM |
1495 | </row> |
1496 | <row> | |
1497 | <entry>@raw-io</entry> | |
aa6b9cec | 1498 | <entry>Raw I/O port access (<citerefentry project='man-pages'><refentrytitle>ioperm</refentrytitle><manvolnum>2</manvolnum></citerefentry>, <citerefentry project='man-pages'><refentrytitle>iopl</refentrytitle><manvolnum>2</manvolnum></citerefentry>, <function>pciconfig_read()</function>, …)</entry> |
201c1cc2 | 1499 | </row> |
bd2ab3f4 LP |
1500 | <row> |
1501 | <entry>@reboot</entry> | |
1502 | <entry>System calls for rebooting and reboot preparation (<citerefentry project='man-pages'><refentrytitle>reboot</refentrytitle><manvolnum>2</manvolnum></citerefentry>, <function>kexec()</function>, …)</entry> | |
1503 | </row> | |
133ddbbe LP |
1504 | <row> |
1505 | <entry>@resources</entry> | |
1506 | <entry>System calls for changing resource limits, memory and scheduling parameters (<citerefentry project='man-pages'><refentrytitle>setrlimit</refentrytitle><manvolnum>2</manvolnum></citerefentry>, <citerefentry project='man-pages'><refentrytitle>setpriority</refentrytitle><manvolnum>2</manvolnum></citerefentry>, …)</entry> | |
1507 | </row> | |
6eaaeee9 LP |
1508 | <row> |
1509 | <entry>@setuid</entry> | |
1510 | <entry>System calls for changing user ID and group ID credentials, (<citerefentry project='man-pages'><refentrytitle>setuid</refentrytitle><manvolnum>2</manvolnum></citerefentry>, <citerefentry project='man-pages'><refentrytitle>setgid</refentrytitle><manvolnum>2</manvolnum></citerefentry>, <citerefentry project='man-pages'><refentrytitle>setresuid</refentrytitle><manvolnum>2</manvolnum></citerefentry>, …)</entry> | |
1511 | </row> | |
bd2ab3f4 LP |
1512 | <row> |
1513 | <entry>@swap</entry> | |
1514 | <entry>System calls for enabling/disabling swap devices (<citerefentry project='man-pages'><refentrytitle>swapon</refentrytitle><manvolnum>2</manvolnum></citerefentry>, <citerefentry project='man-pages'><refentrytitle>swapoff</refentrytitle><manvolnum>2</manvolnum></citerefentry>)</entry> | |
1515 | </row> | |
201c1cc2 TM |
1516 | </tbody> |
1517 | </tgroup> | |
1518 | </table> | |
1519 | ||
869feb33 ZJS |
1520 | Note, that as new system calls are added to the kernel, additional system calls might be |
1521 | added to the groups above. Contents of the sets may also change between systemd | |
1522 | versions. In addition, the list of system calls depends on the kernel version and | |
1523 | architecture for which systemd was compiled. Use | |
1524 | <command>systemd-analyze syscall-filter</command> to list the actual list of system calls in | |
1525 | each filter. | |
1526 | </para> | |
effbd6d2 LP |
1527 | |
1528 | <para>It is recommended to combine the file system namespacing related options with | |
1529 | <varname>SystemCallFilter=~@mount</varname>, in order to prohibit the unit's processes to undo the | |
1530 | mappings. Specifically these are the options <varname>PrivateTmp=</varname>, | |
1531 | <varname>PrivateDevices=</varname>, <varname>ProtectSystem=</varname>, <varname>ProtectHome=</varname>, | |
1532 | <varname>ProtectKernelTunables=</varname>, <varname>ProtectControlGroups=</varname>, | |
1533 | <varname>ReadOnlyPaths=</varname>, <varname>InaccessiblePaths=</varname> and | |
1534 | <varname>ReadWritePaths=</varname>.</para></listitem> | |
798d3a52 ZJS |
1535 | </varlistentry> |
1536 | ||
1537 | <varlistentry> | |
1538 | <term><varname>SystemCallErrorNumber=</varname></term> | |
1539 | ||
1540 | <listitem><para>Takes an <literal>errno</literal> error number | |
1541 | name to return when the system call filter configured with | |
1542 | <varname>SystemCallFilter=</varname> is triggered, instead of | |
1543 | terminating the process immediately. Takes an error name such | |
1544 | as <constant>EPERM</constant>, <constant>EACCES</constant> or | |
1545 | <constant>EUCLEAN</constant>. When this setting is not used, | |
1546 | or when the empty string is assigned, the process will be | |
1547 | terminated immediately when the filter is | |
1548 | triggered.</para></listitem> | |
1549 | </varlistentry> | |
1550 | ||
1551 | <varlistentry> | |
1552 | <term><varname>SystemCallArchitectures=</varname></term> | |
1553 | ||
0b8fab97 LP |
1554 | <listitem><para>Takes a space-separated list of architecture identifiers to include in the system call |
1555 | filter. The known architecture identifiers are the same as for <varname>ConditionArchitecture=</varname> | |
1556 | described in <citerefentry><refentrytitle>systemd.unit</refentrytitle><manvolnum>5</manvolnum></citerefentry>, | |
1557 | as well as <constant>x32</constant>, <constant>mips64-n32</constant>, <constant>mips64-le-n32</constant>, and | |
1558 | the special identifier <constant>native</constant>. Only system calls of the specified architectures will be | |
1559 | permitted to processes of this unit. This is an effective way to disable compatibility with non-native | |
1560 | architectures for processes, for example to prohibit execution of 32-bit x86 binaries on 64-bit x86-64 | |
1561 | systems. The special <constant>native</constant> identifier implicitly maps to the native architecture of the | |
1562 | system (or more strictly: to the architecture the system manager is compiled for). If running in user mode, or | |
1563 | in system mode, but without the <constant>CAP_SYS_ADMIN</constant> capability (e.g. setting | |
1564 | <varname>User=nobody</varname>), <varname>NoNewPrivileges=yes</varname> is implied. Note that setting this | |
1565 | option to a non-empty list implies that <constant>native</constant> is included too. By default, this option is | |
1566 | set to the empty list, i.e. no system call architecture filtering is applied.</para> | |
1567 | ||
1568 | <para>Note that system call filtering is not equally effective on all architectures. For example, on x86 | |
1569 | filtering of network socket-related calls is not possible, due to ABI limitations — a limitation that x86-64 | |
1570 | does not have, however. On systems supporting multiple ABIs at the same time — such as x86/x86-64 — it is hence | |
1571 | recommended to limit the set of permitted system call architectures so that secondary ABIs may not be used to | |
1572 | circumvent the restrictions applied to the native ABI of the system. In particular, setting | |
c29ebc1a | 1573 | <varname>SystemCallArchitectures=native</varname> is a good choice for disabling non-native ABIs.</para> |
0b8fab97 LP |
1574 | |
1575 | <para>System call architectures may also be restricted system-wide via the | |
1576 | <varname>SystemCallArchitectures=</varname> option in the global configuration. See | |
1577 | <citerefentry><refentrytitle>systemd-system.conf</refentrytitle><manvolnum>5</manvolnum></citerefentry> for | |
1578 | details.</para></listitem> | |
798d3a52 ZJS |
1579 | </varlistentry> |
1580 | ||
1581 | <varlistentry> | |
1582 | <term><varname>RestrictAddressFamilies=</varname></term> | |
1583 | ||
142bd808 LP |
1584 | <listitem><para>Restricts the set of socket address families accessible to the processes of this unit. Takes a |
1585 | space-separated list of address family names to whitelist, such as <constant>AF_UNIX</constant>, | |
1586 | <constant>AF_INET</constant> or <constant>AF_INET6</constant>. When prefixed with <constant>~</constant> the | |
1587 | listed address families will be applied as blacklist, otherwise as whitelist. Note that this restricts access | |
1588 | to the <citerefentry | |
1589 | project='man-pages'><refentrytitle>socket</refentrytitle><manvolnum>2</manvolnum></citerefentry> system call | |
1590 | only. Sockets passed into the process by other means (for example, by using socket activation with socket | |
1591 | units, see <citerefentry><refentrytitle>systemd.socket</refentrytitle><manvolnum>5</manvolnum></citerefentry>) | |
1592 | are unaffected. Also, sockets created with <function>socketpair()</function> (which creates connected AF_UNIX | |
1593 | sockets only) are unaffected. Note that this option has no effect on 32-bit x86, s390, s390x, mips, mips-le, | |
0b8fab97 LP |
1594 | ppc, ppc-le, pcc64, ppc64-le and is ignored (but works correctly on other ABIs, including x86-64). Note that on |
1595 | systems supporting multiple ABIs (such as x86/x86-64) it is recommended to turn off alternative ABIs for | |
1596 | services, so that they cannot be used to circumvent the restrictions of this option. Specifically, it is | |
1597 | recommended to combine this option with <varname>SystemCallArchitectures=native</varname> or similar. If | |
142bd808 LP |
1598 | running in user mode, or in system mode, but without the <constant>CAP_SYS_ADMIN</constant> capability |
1599 | (e.g. setting <varname>User=nobody</varname>), <varname>NoNewPrivileges=yes</varname> is implied. By default, | |
1600 | no restrictions apply, all address families are accessible to processes. If assigned the empty string, any | |
1601 | previous address familiy restriction changes are undone. This setting does not affect commands prefixed with | |
1602 | <literal>+</literal>.</para> | |
1603 | ||
1604 | <para>Use this option to limit exposure of processes to remote access, in particular via exotic and sensitive | |
1605 | network protocols, such as <constant>AF_PACKET</constant>. Note that in most cases, the local | |
1606 | <constant>AF_UNIX</constant> address family should be included in the configured whitelist as it is frequently | |
1607 | used for local communication, including for | |
798d3a52 | 1608 | <citerefentry><refentrytitle>syslog</refentrytitle><manvolnum>2</manvolnum></citerefentry> |
142bd808 | 1609 | logging.</para></listitem> |
798d3a52 ZJS |
1610 | </varlistentry> |
1611 | ||
add00535 LP |
1612 | <varlistentry> |
1613 | <term><varname>RestrictNamespaces=</varname></term> | |
1614 | ||
1615 | <listitem><para>Restricts access to Linux namespace functionality for the processes of this unit. For details | |
1616 | about Linux namespaces, see | |
98e9d710 | 1617 | <citerefentry project='man-pages'><refentrytitle>namespaces</refentrytitle><manvolnum>7</manvolnum></citerefentry>. Either takes a |
add00535 LP |
1618 | boolean argument, or a space-separated list of namespace type identifiers. If false (the default), no |
1619 | restrictions on namespace creation and switching are made. If true, access to any kind of namespacing is | |
1620 | prohibited. Otherwise, a space-separated list of namespace type identifiers must be specified, consisting of | |
1621 | any combination of: <constant>cgroup</constant>, <constant>ipc</constant>, <constant>net</constant>, | |
1622 | <constant>mnt</constant>, <constant>pid</constant>, <constant>user</constant> and <constant>uts</constant>. Any | |
1623 | namespace type listed is made accessible to the unit's processes, access to namespace types not listed is | |
ceabfb88 | 1624 | prohibited (whitelisting). By prepending the list with a single tilde character (<literal>~</literal>) the |
add00535 LP |
1625 | effect may be inverted: only the listed namespace types will be made inaccessible, all unlisted ones are |
1626 | permitted (blacklisting). If the empty string is assigned, the default namespace restrictions are applied, | |
1627 | which is equivalent to false. Internally, this setting limits access to the | |
1628 | <citerefentry><refentrytitle>unshare</refentrytitle><manvolnum>2</manvolnum></citerefentry>, | |
1629 | <citerefentry><refentrytitle>clone</refentrytitle><manvolnum>2</manvolnum></citerefentry> and | |
1630 | <citerefentry><refentrytitle>setns</refentrytitle><manvolnum>2</manvolnum></citerefentry> system calls, taking | |
1631 | the specified flags parameters into account. Note that — if this option is used — in addition to restricting | |
1632 | creation and switching of the specified types of namespaces (or all of them, if true) access to the | |
ae9d60ce | 1633 | <function>setns()</function> system call with a zero flags parameter is prohibited. This setting is only |
a3645cc6 JC |
1634 | supported on x86, x86-64, mips, mips-le, mips64, mips64-le, mips64-n32, mips64-le-n32, ppc64, ppc64-le, |
1635 | s390 and s390x, and enforces no restrictions on other architectures. If running in user | |
ae9d60ce LP |
1636 | mode, or in system mode, but without the <constant>CAP_SYS_ADMIN</constant> capability (e.g. setting |
1637 | <varname>User=</varname>), <varname>NoNewPrivileges=yes</varname> is implied. </para></listitem> | |
add00535 LP |
1638 | </varlistentry> |
1639 | ||
798d3a52 ZJS |
1640 | <varlistentry> |
1641 | <term><varname>Personality=</varname></term> | |
1642 | ||
7882632d LP |
1643 | <listitem><para>Controls which kernel architecture <citerefentry |
1644 | project='man-pages'><refentrytitle>uname</refentrytitle><manvolnum>2</manvolnum></citerefentry> shall report, | |
1645 | when invoked by unit processes. Takes one of the architecture identifiers <constant>x86</constant>, | |
1646 | <constant>x86-64</constant>, <constant>ppc</constant>, <constant>ppc-le</constant>, <constant>ppc64</constant>, | |
1647 | <constant>ppc64-le</constant>, <constant>s390</constant> or <constant>s390x</constant>. Which personality | |
1648 | architectures are supported depends on the system architecture. Usually the 64bit versions of the various | |
1649 | system architectures support their immediate 32bit personality architecture counterpart, but no others. For | |
1650 | example, <constant>x86-64</constant> systems support the <constant>x86-64</constant> and | |
1651 | <constant>x86</constant> personalities but no others. The personality feature is useful when running 32-bit | |
1652 | services on a 64-bit host system. If not specified, the personality is left unmodified and thus reflects the | |
1653 | personality of the host system's kernel.</para></listitem> | |
798d3a52 ZJS |
1654 | </varlistentry> |
1655 | ||
1656 | <varlistentry> | |
1657 | <term><varname>RuntimeDirectory=</varname></term> | |
798d3a52 | 1658 | |
23a7448e YW |
1659 | <listitem><para>Takes a whitespace-separated list of directory names. The specified directory names must be |
1660 | relative, and may not include <literal>.</literal> or <literal>..</literal>. If set, one or more directories | |
1661 | including their parents by the specified names will be created below <filename>/run</filename> (for system | |
1662 | services) or below <varname>$XDG_RUNTIME_DIR</varname> (for user services) when the unit is started. The | |
1663 | lowest subdirectories are removed when the unit is stopped. It is possible to preserve the directories if | |
1664 | <varname>RuntimeDirectoryPreserve=</varname> is configured to <option>restart</option> or <option>yes</option>. | |
1665 | The lowest subdirectories will have the access mode specified in <varname>RuntimeDirectoryMode=</varname>, | |
1666 | and be owned by the user and group specified in <varname>User=</varname> and <varname>Group=</varname>. | |
1667 | This implies <varname>ReadWritePaths=</varname>, that is, the directories specified | |
1668 | in this option are accessible with the access mode specified in <varname>RuntimeDirectoryMode=</varname> | |
1669 | even if <varname>ProtectSystem=</varname> is set to <option>strict</option>. | |
1670 | Use this to manage one or more runtime directories of the unit and bind their | |
1671 | lifetime to the daemon runtime. This is particularly useful for unprivileged daemons that cannot create | |
1672 | runtime directories in <filename>/run</filename> due to lack of privileges, and to make sure the runtime | |
1673 | directory is cleaned up automatically after use. For runtime directories that require more complex or | |
1674 | different configuration or lifetime guarantees, please consider using | |
1675 | <citerefentry><refentrytitle>tmpfiles.d</refentrytitle><manvolnum>5</manvolnum></citerefentry>.</para> | |
1676 | ||
1677 | <para>Example: if a system service unit has the following, | |
1678 | <programlisting>RuntimeDirectory=foo/bar baz</programlisting> | |
1679 | the service manager creates <filename>/run/foo</filename> (if it does not exist), <filename>/run/foo/bar</filename>, | |
1680 | and <filename>/run/baz</filename>. The directories <filename>/run/foo/bar</filename> and <filename>/run/baz</filename> | |
1681 | except <filename>/run/foo</filename> are owned by the user and group specified in <varname>User=</varname> and | |
1682 | <varname>Group=</varname>, and removed when the service is stopped. | |
1683 | </para></listitem> | |
798d3a52 ZJS |
1684 | </varlistentry> |
1685 | ||
3536f49e YW |
1686 | <varlistentry> |
1687 | <term><varname>StateDirectory=</varname></term> | |
1688 | <term><varname>CacheDirectory=</varname></term> | |
1689 | <term><varname>LogsDirectory=</varname></term> | |
1690 | <term><varname>ConfigurationDirectory=</varname></term> | |
1691 | ||
1692 | <listitem><para>Takes a whitespace-separated list of directory names. If set, as similar to | |
1693 | <varname>RuntimeDirectory=</varname>, one or more directories including their parents by the specified names | |
1694 | will be created below <filename>/var/lib</filename>, <filename>/var/cache</filename>, <filename>/var/log</filename>, | |
1695 | or <filename>/etc</filename>, respectively, when the unit is started. | |
1696 | Unlike <varname>RuntimeDirectory=</varname>, the directories are not removed when the unit is stopped. | |
1697 | The lowest subdirectories will be owned by the user and group specified in <varname>User=</varname> | |
1698 | and <varname>Group=</varname>. The options imply <varname>ReadWritePaths=</varname>. | |
1699 | </para></listitem> | |
1700 | </varlistentry> | |
1701 | ||
189cd8c2 ZJS |
1702 | <varlistentry> |
1703 | <term><varname>RuntimeDirectoryMode=</varname></term> | |
3536f49e YW |
1704 | <term><varname>StateDirectoryMode=</varname></term> |
1705 | <term><varname>CacheDirectoryMode=</varname></term> | |
1706 | <term><varname>LogsDirectoryMode=</varname></term> | |
1707 | <term><varname>ConfigurationDirectoryMode=</varname></term> | |
189cd8c2 ZJS |
1708 | |
1709 | <listitem><para>Specifies the access mode of the directories specified in | |
3536f49e YW |
1710 | <varname>RuntimeDirectory=</varname>, <varname>StateDirectory=</varname>, <varname>CacheDirectory=</varname>, |
1711 | <varname>LogsDirectory=</varname>, or <varname>ConfigurationDirectory=</varname>, respectively, as an octal number. | |
1712 | Defaults to <constant>0755</constant>. See "Permissions" in | |
23a7448e YW |
1713 | <citerefentry project='man-pages'><refentrytitle>path_resolution</refentrytitle><manvolnum>7</manvolnum></citerefentry> |
1714 | for a discussion of the meaning of permission bits. | |
189cd8c2 ZJS |
1715 | </para></listitem> |
1716 | </varlistentry> | |
1717 | ||
53f47dfc YW |
1718 | <varlistentry> |
1719 | <term><varname>RuntimeDirectoryPreserve=</varname></term> | |
1720 | ||
1721 | <listitem><para>Takes a boolean argument or <option>restart</option>. | |
1722 | If set to <option>no</option> (the default), the directories specified in <varname>RuntimeDirectory=</varname> | |
1723 | are always removed when the service stops. If set to <option>restart</option> the directories are preserved | |
1724 | when the service is both automatically and manually restarted. Here, the automatic restart means the operation | |
1725 | specified in <varname>Restart=</varname>, and manual restart means the one triggered by | |
1726 | <command>systemctl restart foo.service</command>. If set to <option>yes</option>, then the directories are not | |
1727 | removed when the service is stopped. Note that since the runtime directory <filename>/run</filename> is a mount | |
1728 | point of <literal>tmpfs</literal>, then for system services the directories specified in | |
1729 | <varname>RuntimeDirectory=</varname> are removed when the system is rebooted. | |
189cd8c2 ZJS |
1730 | </para></listitem> |
1731 | </varlistentry> | |
1732 | ||
f3e43635 TM |
1733 | <varlistentry> |
1734 | <term><varname>MemoryDenyWriteExecute=</varname></term> | |
1735 | ||
1736 | <listitem><para>Takes a boolean argument. If set, attempts to create memory mappings that are writable and | |
8a50cf69 LP |
1737 | executable at the same time, or to change existing memory mappings to become executable, or mapping shared |
1738 | memory segments as executable are prohibited. Specifically, a system call filter is added that rejects | |
1739 | <citerefentry><refentrytitle>mmap</refentrytitle><manvolnum>2</manvolnum></citerefentry> system calls with both | |
1740 | <constant>PROT_EXEC</constant> and <constant>PROT_WRITE</constant> set, | |
1741 | <citerefentry><refentrytitle>mprotect</refentrytitle><manvolnum>2</manvolnum></citerefentry> system calls with | |
1742 | <constant>PROT_EXEC</constant> set and | |
1743 | <citerefentry><refentrytitle>shmat</refentrytitle><manvolnum>2</manvolnum></citerefentry> system calls with | |
03c3c520 ZJS |
1744 | <constant>SHM_EXEC</constant> set. Note that this option is incompatible with programs and libraries that |
1745 | generate program code dynamically at runtime, including JIT execution engines, executable stacks, and code | |
8a50cf69 LP |
1746 | "trampoline" feature of various C compilers. This option improves service security, as it makes harder for |
1747 | software exploits to change running code dynamically. Note that this feature is fully available on x86-64, and | |
0b8fab97 LP |
1748 | partially on x86. Specifically, the <function>shmat()</function> protection is not available on x86. Note that |
1749 | on systems supporting multiple ABIs (such as x86/x86-64) it is recommended to turn off alternative ABIs for | |
1750 | services, so that they cannot be used to circumvent the restrictions of this option. Specifically, it is | |
1751 | recommended to combine this option with <varname>SystemCallArchitectures=native</varname> or similar. If | |
1752 | running in user mode, or in system mode, but without the <constant>CAP_SYS_ADMIN</constant> capability | |
3536f49e | 1753 | (e.g. setting <varname>User=</varname>), <varname>NoNewPrivileges=yes</varname> is implied.</para></listitem> |
f3e43635 TM |
1754 | </varlistentry> |
1755 | ||
f4170c67 LP |
1756 | <varlistentry> |
1757 | <term><varname>RestrictRealtime=</varname></term> | |
1758 | ||
1759 | <listitem><para>Takes a boolean argument. If set, any attempts to enable realtime scheduling in a process of | |
1760 | the unit are refused. This restricts access to realtime task scheduling policies such as | |
1761 | <constant>SCHED_FIFO</constant>, <constant>SCHED_RR</constant> or <constant>SCHED_DEADLINE</constant>. See | |
0a07667d | 1762 | <citerefentry project='man-pages'><refentrytitle>sched</refentrytitle><manvolnum>7</manvolnum></citerefentry> for details about |
a7db8614 DH |
1763 | these scheduling policies. If running in user mode, or in system mode, but |
1764 | without the <constant>CAP_SYS_ADMIN</constant> capability | |
1765 | (e.g. setting <varname>User=</varname>), <varname>NoNewPrivileges=yes</varname> | |
1766 | is implied. Realtime scheduling policies may be used to monopolize CPU time for longer periods | |
f4170c67 LP |
1767 | of time, and may hence be used to lock up or otherwise trigger Denial-of-Service situations on the system. It |
1768 | is hence recommended to restrict access to realtime scheduling to the few programs that actually require | |
1769 | them. Defaults to off.</para></listitem> | |
1770 | </varlistentry> | |
1771 | ||
798d3a52 ZJS |
1772 | </variablelist> |
1773 | </refsect1> | |
1774 | ||
1775 | <refsect1> | |
1776 | <title>Environment variables in spawned processes</title> | |
1777 | ||
1778 | <para>Processes started by the system are executed in a clean | |
1779 | environment in which select variables listed below are set. System | |
1780 | processes started by systemd do not inherit variables from PID 1, | |
1781 | but processes started by user systemd instances inherit all | |
1782 | environment variables from the user systemd instance. | |
1783 | </para> | |
1784 | ||
1785 | <variablelist class='environment-variables'> | |
1786 | <varlistentry> | |
1787 | <term><varname>$PATH</varname></term> | |
1788 | ||
1789 | <listitem><para>Colon-separated list of directories to use | |
1790 | when launching executables. Systemd uses a fixed value of | |
1791 | <filename>/usr/local/sbin</filename>:<filename>/usr/local/bin</filename>:<filename>/usr/sbin</filename>:<filename>/usr/bin</filename>:<filename>/sbin</filename>:<filename>/bin</filename>. | |
1792 | </para></listitem> | |
1793 | </varlistentry> | |
1794 | ||
1795 | <varlistentry> | |
1796 | <term><varname>$LANG</varname></term> | |
1797 | ||
1798 | <listitem><para>Locale. Can be set in | |
3ba3a79d | 1799 | <citerefentry project='man-pages'><refentrytitle>locale.conf</refentrytitle><manvolnum>5</manvolnum></citerefentry> |
798d3a52 ZJS |
1800 | or on the kernel command line (see |
1801 | <citerefentry><refentrytitle>systemd</refentrytitle><manvolnum>1</manvolnum></citerefentry> | |
1802 | and | |
1803 | <citerefentry><refentrytitle>kernel-command-line</refentrytitle><manvolnum>7</manvolnum></citerefentry>). | |
1804 | </para></listitem> | |
1805 | </varlistentry> | |
1806 | ||
1807 | <varlistentry> | |
1808 | <term><varname>$USER</varname></term> | |
1809 | <term><varname>$LOGNAME</varname></term> | |
1810 | <term><varname>$HOME</varname></term> | |
1811 | <term><varname>$SHELL</varname></term> | |
1812 | ||
1813 | <listitem><para>User name (twice), home directory, and the | |
23deef88 LP |
1814 | login shell. The variables are set for the units that have |
1815 | <varname>User=</varname> set, which includes user | |
1816 | <command>systemd</command> instances. See | |
3ba3a79d | 1817 | <citerefentry project='die-net'><refentrytitle>passwd</refentrytitle><manvolnum>5</manvolnum></citerefentry>. |
798d3a52 ZJS |
1818 | </para></listitem> |
1819 | </varlistentry> | |
1820 | ||
4b58153d LP |
1821 | <varlistentry> |
1822 | <term><varname>$INVOCATION_ID</varname></term> | |
1823 | ||
1824 | <listitem><para>Contains a randomized, unique 128bit ID identifying each runtime cycle of the unit, formatted | |
1825 | as 32 character hexadecimal string. A new ID is assigned each time the unit changes from an inactive state into | |
1826 | an activating or active state, and may be used to identify this specific runtime cycle, in particular in data | |
1827 | stored offline, such as the journal. The same ID is passed to all processes run as part of the | |
1828 | unit.</para></listitem> | |
1829 | </varlistentry> | |
1830 | ||
798d3a52 ZJS |
1831 | <varlistentry> |
1832 | <term><varname>$XDG_RUNTIME_DIR</varname></term> | |
1833 | ||
1834 | <listitem><para>The directory for volatile state. Set for the | |
1835 | user <command>systemd</command> instance, and also in user | |
1836 | sessions. See | |
1837 | <citerefentry><refentrytitle>pam_systemd</refentrytitle><manvolnum>8</manvolnum></citerefentry>. | |
1838 | </para></listitem> | |
1839 | </varlistentry> | |
1840 | ||
1841 | <varlistentry> | |
1842 | <term><varname>$XDG_SESSION_ID</varname></term> | |
1843 | <term><varname>$XDG_SEAT</varname></term> | |
1844 | <term><varname>$XDG_VTNR</varname></term> | |
1845 | ||
1846 | <listitem><para>The identifier of the session, the seat name, | |
1847 | and virtual terminal of the session. Set by | |
1848 | <citerefentry><refentrytitle>pam_systemd</refentrytitle><manvolnum>8</manvolnum></citerefentry> | |
1849 | for login sessions. <varname>$XDG_SEAT</varname> and | |
1850 | <varname>$XDG_VTNR</varname> will only be set when attached to | |
1851 | a seat and a tty.</para></listitem> | |
1852 | </varlistentry> | |
1853 | ||
1854 | <varlistentry> | |
1855 | <term><varname>$MAINPID</varname></term> | |
1856 | ||
2dd67817 | 1857 | <listitem><para>The PID of the unit's main process if it is |
798d3a52 ZJS |
1858 | known. This is only set for control processes as invoked by |
1859 | <varname>ExecReload=</varname> and similar. </para></listitem> | |
1860 | </varlistentry> | |
1861 | ||
1862 | <varlistentry> | |
1863 | <term><varname>$MANAGERPID</varname></term> | |
1864 | ||
1865 | <listitem><para>The PID of the user <command>systemd</command> | |
1866 | instance, set for processes spawned by it. </para></listitem> | |
1867 | </varlistentry> | |
1868 | ||
1869 | <varlistentry> | |
1870 | <term><varname>$LISTEN_FDS</varname></term> | |
1871 | <term><varname>$LISTEN_PID</varname></term> | |
5c019cf2 | 1872 | <term><varname>$LISTEN_FDNAMES</varname></term> |
798d3a52 ZJS |
1873 | |
1874 | <listitem><para>Information about file descriptors passed to a | |
1875 | service for socket activation. See | |
1876 | <citerefentry><refentrytitle>sd_listen_fds</refentrytitle><manvolnum>3</manvolnum></citerefentry>. | |
1877 | </para></listitem> | |
1878 | </varlistentry> | |
1879 | ||
5c019cf2 EV |
1880 | <varlistentry> |
1881 | <term><varname>$NOTIFY_SOCKET</varname></term> | |
1882 | ||
1883 | <listitem><para>The socket | |
1884 | <function>sd_notify()</function> talks to. See | |
1885 | <citerefentry><refentrytitle>sd_notify</refentrytitle><manvolnum>3</manvolnum></citerefentry>. | |
1886 | </para></listitem> | |
1887 | </varlistentry> | |
1888 | ||
1889 | <varlistentry> | |
1890 | <term><varname>$WATCHDOG_PID</varname></term> | |
1891 | <term><varname>$WATCHDOG_USEC</varname></term> | |
1892 | ||
1893 | <listitem><para>Information about watchdog keep-alive notifications. See | |
1894 | <citerefentry><refentrytitle>sd_watchdog_enabled</refentrytitle><manvolnum>3</manvolnum></citerefentry>. | |
1895 | </para></listitem> | |
1896 | </varlistentry> | |
1897 | ||
798d3a52 ZJS |
1898 | <varlistentry> |
1899 | <term><varname>$TERM</varname></term> | |
1900 | ||
1901 | <listitem><para>Terminal type, set only for units connected to | |
1902 | a terminal (<varname>StandardInput=tty</varname>, | |
1903 | <varname>StandardOutput=tty</varname>, or | |
1904 | <varname>StandardError=tty</varname>). See | |
1905 | <citerefentry project='man-pages'><refentrytitle>termcap</refentrytitle><manvolnum>5</manvolnum></citerefentry>. | |
1906 | </para></listitem> | |
1907 | </varlistentry> | |
7bce046b LP |
1908 | |
1909 | <varlistentry> | |
1910 | <term><varname>$JOURNAL_STREAM</varname></term> | |
1911 | ||
1912 | <listitem><para>If the standard output or standard error output of the executed processes are connected to the | |
1913 | journal (for example, by setting <varname>StandardError=journal</varname>) <varname>$JOURNAL_STREAM</varname> | |
1914 | contains the device and inode numbers of the connection file descriptor, formatted in decimal, separated by a | |
1915 | colon (<literal>:</literal>). This permits invoked processes to safely detect whether their standard output or | |
1916 | standard error output are connected to the journal. The device and inode numbers of the file descriptors should | |
1917 | be compared with the values set in the environment variable to determine whether the process output is still | |
1918 | connected to the journal. Note that it is generally not sufficient to only check whether | |
1919 | <varname>$JOURNAL_STREAM</varname> is set at all as services might invoke external processes replacing their | |
1920 | standard output or standard error output, without unsetting the environment variable.</para> | |
1921 | ||
1922 | <para>This environment variable is primarily useful to allow services to optionally upgrade their used log | |
1923 | protocol to the native journal protocol (using | |
1924 | <citerefentry><refentrytitle>sd_journal_print</refentrytitle><manvolnum>3</manvolnum></citerefentry> and other | |
1925 | functions) if their standard output or standard error output is connected to the journal anyway, thus enabling | |
1926 | delivery of structured metadata along with logged messages.</para></listitem> | |
1927 | </varlistentry> | |
136dc4c4 LP |
1928 | |
1929 | <varlistentry> | |
1930 | <term><varname>$SERVICE_RESULT</varname></term> | |
1931 | ||
1932 | <listitem><para>Only defined for the service unit type, this environment variable is passed to all | |
1933 | <varname>ExecStop=</varname> and <varname>ExecStopPost=</varname> processes, and encodes the service | |
e0c7d5f7 | 1934 | "result". Currently, the following values are defined: <literal>protocol</literal> (in case of a protocol |
7ed0a4c5 JW |
1935 | violation; if a service did not take the steps required by its unit configuration), <literal>timeout</literal> |
1936 | (in case of an operation timeout), <literal>exit-code</literal> (if a service process exited with a non-zero | |
1937 | exit code; see <varname>$EXIT_CODE</varname> below for the actual exit code returned), <literal>signal</literal> | |
e0c7d5f7 JW |
1938 | (if a service process was terminated abnormally by a signal; see <varname>$EXIT_CODE</varname> below for the |
1939 | actual signal used for the termination), <literal>core-dump</literal> (if a service process terminated | |
1940 | abnormally and dumped core), <literal>watchdog</literal> (if the watchdog keep-alive ping was enabled for the | |
1941 | service but it missed the deadline), or <literal>resources</literal> (a catch-all condition in case a system | |
1942 | operation failed).</para> | |
136dc4c4 LP |
1943 | |
1944 | <para>This environment variable is useful to monitor failure or successful termination of a service. Even | |
1945 | though this variable is available in both <varname>ExecStop=</varname> and <varname>ExecStopPost=</varname>, it | |
1946 | is usually a better choice to place monitoring tools in the latter, as the former is only invoked for services | |
1947 | that managed to start up correctly, and the latter covers both services that failed during their start-up and | |
1948 | those which failed during their runtime.</para></listitem> | |
1949 | </varlistentry> | |
1950 | ||
1951 | <varlistentry> | |
1952 | <term><varname>$EXIT_CODE</varname></term> | |
1953 | <term><varname>$EXIT_STATUS</varname></term> | |
1954 | ||
1955 | <listitem><para>Only defined for the service unit type, these environment variables are passed to all | |
1956 | <varname>ExecStop=</varname>, <varname>ExecStopPost=</varname> processes and contain exit status/code | |
1957 | information of the main process of the service. For the precise definition of the exit code and status, see | |
1958 | <citerefentry><refentrytitle>wait</refentrytitle><manvolnum>2</manvolnum></citerefentry>. <varname>$EXIT_CODE</varname> | |
1959 | is one of <literal>exited</literal>, <literal>killed</literal>, | |
1960 | <literal>dumped</literal>. <varname>$EXIT_STATUS</varname> contains the numeric exit code formatted as string | |
1961 | if <varname>$EXIT_CODE</varname> is <literal>exited</literal>, and the signal name in all other cases. Note | |
1962 | that these environment variables are only set if the service manager succeeded to start and identify the main | |
e64e1bfd ZJS |
1963 | process of the service.</para> |
1964 | ||
1965 | <table> | |
1966 | <title>Summary of possible service result variable values</title> | |
1967 | <tgroup cols='3'> | |
1968 | <colspec colname='result' /> | |
e64e1bfd | 1969 | <colspec colname='code' /> |
a4e26faf | 1970 | <colspec colname='status' /> |
e64e1bfd ZJS |
1971 | <thead> |
1972 | <row> | |
1973 | <entry><varname>$SERVICE_RESULT</varname></entry> | |
e64e1bfd | 1974 | <entry><varname>$EXIT_CODE</varname></entry> |
a4e26faf | 1975 | <entry><varname>$EXIT_STATUS</varname></entry> |
e64e1bfd ZJS |
1976 | </row> |
1977 | </thead> | |
1978 | ||
1979 | <tbody> | |
a4e26faf JW |
1980 | <row> |
1981 | <entry morerows="1" valign="top"><literal>protocol</literal></entry> | |
1982 | <entry valign="top">not set</entry> | |
1983 | <entry>not set</entry> | |
1984 | </row> | |
1985 | <row> | |
1986 | <entry><literal>exited</literal></entry> | |
1987 | <entry><literal>0</literal></entry> | |
1988 | </row> | |
1989 | ||
29df65f9 ZJS |
1990 | <row> |
1991 | <entry morerows="1" valign="top"><literal>timeout</literal></entry> | |
1992 | <entry valign="top"><literal>killed</literal></entry> | |
6757c06a | 1993 | <entry><literal>TERM</literal>, <literal>KILL</literal></entry> |
29df65f9 | 1994 | </row> |
29df65f9 ZJS |
1995 | <row> |
1996 | <entry valign="top"><literal>exited</literal></entry> | |
6757c06a LP |
1997 | <entry><literal>0</literal>, <literal>1</literal>, <literal>2</literal>, <literal |
1998 | >3</literal>, …, <literal>255</literal></entry> | |
29df65f9 ZJS |
1999 | </row> |
2000 | ||
e64e1bfd ZJS |
2001 | <row> |
2002 | <entry valign="top"><literal>exit-code</literal></entry> | |
2003 | <entry valign="top"><literal>exited</literal></entry> | |
6757c06a LP |
2004 | <entry><literal>0</literal>, <literal>1</literal>, <literal>2</literal>, <literal |
2005 | >3</literal>, …, <literal>255</literal></entry> | |
e64e1bfd ZJS |
2006 | </row> |
2007 | ||
2008 | <row> | |
2009 | <entry valign="top"><literal>signal</literal></entry> | |
2010 | <entry valign="top"><literal>killed</literal></entry> | |
6757c06a | 2011 | <entry><literal>HUP</literal>, <literal>INT</literal>, <literal>KILL</literal>, …</entry> |
e64e1bfd ZJS |
2012 | </row> |
2013 | ||
2014 | <row> | |
2015 | <entry valign="top"><literal>core-dump</literal></entry> | |
2016 | <entry valign="top"><literal>dumped</literal></entry> | |
6757c06a | 2017 | <entry><literal>ABRT</literal>, <literal>SEGV</literal>, <literal>QUIT</literal>, …</entry> |
e64e1bfd | 2018 | </row> |
136dc4c4 | 2019 | |
e64e1bfd ZJS |
2020 | <row> |
2021 | <entry morerows="2" valign="top"><literal>watchdog</literal></entry> | |
2022 | <entry><literal>dumped</literal></entry> | |
2023 | <entry><literal>ABRT</literal></entry> | |
2024 | </row> | |
2025 | <row> | |
2026 | <entry><literal>killed</literal></entry> | |
6757c06a | 2027 | <entry><literal>TERM</literal>, <literal>KILL</literal></entry> |
e64e1bfd ZJS |
2028 | </row> |
2029 | <row> | |
2030 | <entry><literal>exited</literal></entry> | |
6757c06a LP |
2031 | <entry><literal>0</literal>, <literal>1</literal>, <literal>2</literal>, <literal |
2032 | >3</literal>, …, <literal>255</literal></entry> | |
e64e1bfd ZJS |
2033 | </row> |
2034 | ||
2035 | <row> | |
2036 | <entry><literal>resources</literal></entry> | |
2037 | <entry>any of the above</entry> | |
2038 | <entry>any of the above</entry> | |
2039 | </row> | |
29df65f9 ZJS |
2040 | |
2041 | <row> | |
2042 | <entry namest="results" nameend="code">Note: the process may be also terminated by a signal not sent by systemd. In particular the process may send an arbitrary signal to itself in a handler for any of the non-maskable signals. Nevertheless, in the <literal>timeout</literal> and <literal>watchdog</literal> rows above only the signals that systemd sends have been included.</entry> | |
2043 | </row> | |
e64e1bfd ZJS |
2044 | </tbody> |
2045 | </tgroup> | |
2046 | </table> | |
2047 | ||
2048 | </listitem> | |
2049 | </varlistentry> | |
798d3a52 ZJS |
2050 | </variablelist> |
2051 | ||
2052 | <para>Additional variables may be configured by the following | |
2053 | means: for processes spawned in specific units, use the | |
5c019cf2 EV |
2054 | <varname>Environment=</varname>, <varname>EnvironmentFile=</varname> |
2055 | and <varname>PassEnvironment=</varname> options above; to specify | |
798d3a52 ZJS |
2056 | variables globally, use <varname>DefaultEnvironment=</varname> |
2057 | (see | |
2058 | <citerefentry><refentrytitle>systemd-system.conf</refentrytitle><manvolnum>5</manvolnum></citerefentry>) | |
2059 | or the kernel option <varname>systemd.setenv=</varname> (see | |
2060 | <citerefentry><refentrytitle>systemd</refentrytitle><manvolnum>1</manvolnum></citerefentry>). | |
2061 | Additional variables may also be set through PAM, | |
2062 | cf. <citerefentry project='man-pages'><refentrytitle>pam_env</refentrytitle><manvolnum>8</manvolnum></citerefentry>.</para> | |
2063 | </refsect1> | |
2064 | ||
2065 | <refsect1> | |
2066 | <title>See Also</title> | |
2067 | <para> | |
2068 | <citerefentry><refentrytitle>systemd</refentrytitle><manvolnum>1</manvolnum></citerefentry>, | |
2069 | <citerefentry><refentrytitle>systemctl</refentrytitle><manvolnum>1</manvolnum></citerefentry>, | |
869feb33 | 2070 | <citerefentry><refentrytitle>systemd-analyze</refentrytitle><manvolnum>1</manvolnum></citerefentry>, |
798d3a52 ZJS |
2071 | <citerefentry><refentrytitle>journalctl</refentrytitle><manvolnum>8</manvolnum></citerefentry>, |
2072 | <citerefentry><refentrytitle>systemd.unit</refentrytitle><manvolnum>5</manvolnum></citerefentry>, | |
2073 | <citerefentry><refentrytitle>systemd.service</refentrytitle><manvolnum>5</manvolnum></citerefentry>, | |
2074 | <citerefentry><refentrytitle>systemd.socket</refentrytitle><manvolnum>5</manvolnum></citerefentry>, | |
2075 | <citerefentry><refentrytitle>systemd.swap</refentrytitle><manvolnum>5</manvolnum></citerefentry>, | |
2076 | <citerefentry><refentrytitle>systemd.mount</refentrytitle><manvolnum>5</manvolnum></citerefentry>, | |
2077 | <citerefentry><refentrytitle>systemd.kill</refentrytitle><manvolnum>5</manvolnum></citerefentry>, | |
2078 | <citerefentry><refentrytitle>systemd.resource-control</refentrytitle><manvolnum>5</manvolnum></citerefentry>, | |
a4c18002 | 2079 | <citerefentry><refentrytitle>systemd.time</refentrytitle><manvolnum>7</manvolnum></citerefentry>, |
798d3a52 ZJS |
2080 | <citerefentry><refentrytitle>systemd.directives</refentrytitle><manvolnum>7</manvolnum></citerefentry>, |
2081 | <citerefentry><refentrytitle>tmpfiles.d</refentrytitle><manvolnum>5</manvolnum></citerefentry>, | |
2082 | <citerefentry project='man-pages'><refentrytitle>exec</refentrytitle><manvolnum>3</manvolnum></citerefentry> | |
2083 | </para> | |
2084 | </refsect1> | |
dd1eb43b | 2085 | |
e64e1bfd | 2086 | |
dd1eb43b | 2087 | </refentry> |