]> git.ipfire.org Git - thirdparty/systemd.git/blob - man/systemd.exec.xml
exec: optionally apply cgroup attributes to the cgroups we create
[thirdparty/systemd.git] / man / systemd.exec.xml
1 <?xml version='1.0'?> <!--*-nxml-*-->
2 <?xml-stylesheet type="text/xsl" href="http://docbook.sourceforge.net/release/xsl/current/xhtml/docbook.xsl"?>
3 <!DOCTYPE refentry PUBLIC "-//OASIS//DTD DocBook XML V4.2//EN"
4 "http://www.oasis-open.org/docbook/xml/4.2/docbookx.dtd">
5
6 <!--
7 This file is part of systemd.
8
9 Copyright 2010 Lennart Poettering
10
11 systemd is free software; you can redistribute it and/or modify it
12 under the terms of the GNU General Public License as published by
13 the Free Software Foundation; either version 2 of the License, or
14 (at your option) any later version.
15
16 systemd is distributed in the hope that it will be useful, but
17 WITHOUT ANY WARRANTY; without even the implied warranty of
18 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
19 General Public License for more details.
20
21 You should have received a copy of the GNU General Public License
22 along with systemd; If not, see <http://www.gnu.org/licenses/>.
23 -->
24
25 <refentry id="systemd.exec">
26 <refentryinfo>
27 <title>systemd.exec</title>
28 <productname>systemd</productname>
29
30 <authorgroup>
31 <author>
32 <contrib>Developer</contrib>
33 <firstname>Lennart</firstname>
34 <surname>Poettering</surname>
35 <email>lennart@poettering.net</email>
36 </author>
37 </authorgroup>
38 </refentryinfo>
39
40 <refmeta>
41 <refentrytitle>systemd.exec</refentrytitle>
42 <manvolnum>5</manvolnum>
43 </refmeta>
44
45 <refnamediv>
46 <refname>systemd.exec</refname>
47 <refpurpose>systemd execution environment configuration</refpurpose>
48 </refnamediv>
49
50 <refsynopsisdiv>
51 <para><filename>systemd.service</filename>,
52 <filename>systemd.socket</filename>,
53 <filename>systemd.mount</filename>,
54 <filename>systemd.swap</filename></para>
55 </refsynopsisdiv>
56
57 <refsect1>
58 <title>Description</title>
59
60 <para>Unit configuration files for services, sockets,
61 mount points and swap devices share a subset of
62 configuration options which define the execution
63 environment of spawned processes.</para>
64
65 <para>This man page lists the configuration options
66 shared by these four unit types. See
67 <citerefentry><refentrytitle>systemd.unit</refentrytitle><manvolnum>5</manvolnum></citerefentry>
68 for the common options of all unit configuration
69 files, and
70 <citerefentry><refentrytitle>systemd.service</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
71 <citerefentry><refentrytitle>systemd.socket</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
72 <citerefentry><refentrytitle>systemd.swap</refentrytitle><manvolnum>5</manvolnum></citerefentry>
73 and
74 <citerefentry><refentrytitle>systemd.mount</refentrytitle><manvolnum>5</manvolnum></citerefentry>
75 for more information on the specific unit
76 configuration files. The execution specific
77 configuration options are configured in the [Service],
78 [Socket], [Mount] resp. [Swap] section, depending on the unit
79 type.</para>
80 </refsect1>
81
82 <refsect1>
83 <title>Options</title>
84
85 <variablelist>
86
87 <varlistentry>
88 <term><varname>WorkingDirectory=</varname></term>
89
90 <listitem><para>Takes an absolute
91 directory path. Sets the working
92 directory for executed
93 processes.</para></listitem>
94 </varlistentry>
95
96 <varlistentry>
97 <term><varname>RootDirectory=</varname></term>
98
99 <listitem><para>Takes an absolute
100 directory path. Sets the root
101 directory for executed processes, with
102 the
103 <citerefentry><refentrytitle>chroot</refentrytitle><manvolnum>2</manvolnum></citerefentry>
104 system call. If this is used it must
105 be ensured that the process and all
106 its auxiliary files are available in
107 the <function>chroot()</function>
108 jail.</para></listitem>
109 </varlistentry>
110
111 <varlistentry>
112 <term><varname>User=</varname></term>
113 <term><varname>Group=</varname></term>
114
115 <listitem><para>Sets the Unix user
116 resp. group the processes are executed
117 as. Takes a single user resp. group
118 name or ID as argument. If no group is
119 set the default group of the user is
120 chosen.</para></listitem>
121 </varlistentry>
122
123 <varlistentry>
124 <term><varname>SupplementaryGroups=</varname></term>
125
126 <listitem><para>Sets the supplementary
127 Unix groups the processes are executed
128 as. This takes a space separated list
129 of group names or IDs. This option may
130 be specified more than once in which
131 case all listed groups are set as
132 supplementary groups. This option does
133 not override but extends the list of
134 supplementary groups configured in the
135 system group database for the
136 user.</para></listitem>
137 </varlistentry>
138
139 <varlistentry>
140 <term><varname>Nice=</varname></term>
141
142 <listitem><para>Sets the default nice
143 level (scheduling priority) for
144 executed processes. Takes an integer
145 between -20 (highest priority) and 19
146 (lowest priority). See
147 <citerefentry><refentrytitle>setpriority</refentrytitle><manvolnum>2</manvolnum></citerefentry>
148 for details.</para></listitem>
149 </varlistentry>
150
151 <varlistentry>
152 <term><varname>OOMScoreAdjust=</varname></term>
153
154 <listitem><para>Sets the adjustment
155 level for the Out-Of-Memory killer for
156 executed processes. Takes an integer
157 between -1000 (to disable OOM killing
158 for this process) and 1000 (to make
159 killing of this process under memory
160 pressure very likely). See <ulink
161 url="http://www.kernel.org/doc/Documentation/filesystems/proc.txt">proc.txt</ulink>
162 for details.</para></listitem>
163 </varlistentry>
164
165 <varlistentry>
166 <term><varname>IOSchedulingClass=</varname></term>
167
168 <listitem><para>Sets the IO scheduling
169 class for executed processes. Takes an
170 integer between 0 and 3 or one of the
171 strings <option>none</option>,
172 <option>realtime</option>,
173 <option>best-effort</option> or
174 <option>idle</option>. See
175 <citerefentry><refentrytitle>ioprio_set</refentrytitle><manvolnum>2</manvolnum></citerefentry>
176 for details.</para></listitem>
177 </varlistentry>
178
179 <varlistentry>
180 <term><varname>IOSchedulingPriority=</varname></term>
181
182 <listitem><para>Sets the IO scheduling
183 priority for executed processes. Takes
184 an integer between 0 (highest
185 priority) and 7 (lowest priority). The
186 available priorities depend on the
187 selected IO scheduling class (see
188 above). See
189 <citerefentry><refentrytitle>ioprio_set</refentrytitle><manvolnum>2</manvolnum></citerefentry>
190 for details.</para></listitem>
191 </varlistentry>
192
193 <varlistentry>
194 <term><varname>CPUSchedulingPolicy=</varname></term>
195
196 <listitem><para>Sets the CPU
197 scheduling policy for executed
198 processes. Takes one of
199 <option>other</option>,
200 <option>batch</option>,
201 <option>idle</option>,
202 <option>fifo</option> or
203 <option>rr</option>. See
204 <citerefentry><refentrytitle>sched_setscheduler</refentrytitle><manvolnum>2</manvolnum></citerefentry>
205 for details.</para></listitem>
206 </varlistentry>
207
208 <varlistentry>
209 <term><varname>CPUSchedulingPriority=</varname></term>
210
211 <listitem><para>Sets the CPU
212 scheduling priority for executed
213 processes. Takes an integer between 1
214 (lowest priority) and 99 (highest
215 priority). The available priority
216 range depends on the selected CPU
217 scheduling policy (see above). See
218 <citerefentry><refentrytitle>sched_setscheduler</refentrytitle><manvolnum>2</manvolnum></citerefentry>
219 for details.</para></listitem>
220 </varlistentry>
221
222 <varlistentry>
223 <term><varname>CPUSchedulingResetOnFork=</varname></term>
224
225 <listitem><para>Takes a boolean
226 argument. If true elevated CPU
227 scheduling priorities and policies
228 will be reset when the executed
229 processes fork, and can hence not leak
230 into child processes. See
231 <citerefentry><refentrytitle>sched_setscheduler</refentrytitle><manvolnum>2</manvolnum></citerefentry>
232 for details. Defaults to false.</para></listitem>
233 </varlistentry>
234
235 <varlistentry>
236 <term><varname>CPUAffinity=</varname></term>
237
238 <listitem><para>Controls the CPU
239 affinity of the executed
240 processes. Takes a space-separated
241 list of CPU indexes. See
242 <citerefentry><refentrytitle>sched_setaffinity</refentrytitle><manvolnum>2</manvolnum></citerefentry>
243 for details.</para></listitem>
244 </varlistentry>
245
246 <varlistentry>
247 <term><varname>UMask=</varname></term>
248
249 <listitem><para>Controls the file mode
250 creation mask. Takes an access mode in
251 octal notation. See
252 <citerefentry><refentrytitle>umask</refentrytitle><manvolnum>2</manvolnum></citerefentry>
253 for details. Defaults to
254 0022.</para></listitem>
255 </varlistentry>
256
257 <varlistentry>
258 <term><varname>Environment=</varname></term>
259
260 <listitem><para>Sets environment
261 variables for executed
262 processes. Takes a space-separated
263 list of variable assignments. This
264 option may be specified more than once
265 in which case all listed variables
266 will be set. If the same variable is
267 set twice the later setting will
268 override the earlier setting. See
269 <citerefentry><refentrytitle>environ</refentrytitle><manvolnum>7</manvolnum></citerefentry>
270 for details.</para></listitem>
271 </varlistentry>
272 <varlistentry>
273 <term><varname>EnvironmentFile=</varname></term>
274 <listitem><para>Similar to
275 <varname>Environment=</varname> but
276 reads the environment variables from a
277 text file. The text file should
278 contain new-line separated variable
279 assignments. Empty lines and lines
280 starting with ; or # will be ignored,
281 which may be used for commenting. The
282 argument passed should be an absolute
283 file name, optionally prefixed with
284 "-", which indicates that if the file
285 does not exist it won't be read and no
286 error or warning message is
287 logged. The files listed with this
288 directive will be read shortly before
289 the process is executed. Settings from
290 these files override settings made
291 with
292 <varname>Environment=</varname>. If
293 the same variable is set twice from
294 these files the files will be read in
295 the order they are specified and the
296 later setting will override the
297 earlier setting. </para></listitem>
298 </varlistentry>
299
300 <varlistentry>
301 <term><varname>StandardInput=</varname></term>
302 <listitem><para>Controls where file
303 descriptor 0 (STDIN) of the executed
304 processes is connected to. Takes one
305 of <option>null</option>,
306 <option>tty</option>,
307 <option>tty-force</option>,
308 <option>tty-fail</option> or
309 <option>socket</option>. If
310 <option>null</option> is selected
311 standard input will be connected to
312 <filename>/dev/null</filename>,
313 i.e. all read attempts by the process
314 will result in immediate EOF. If
315 <option>tty</option> is selected
316 standard input is connected to a TTY
317 (as configured by
318 <varname>TTYPath=</varname>, see
319 below) and the executed process
320 becomes the controlling process of the
321 terminal. If the terminal is already
322 being controlled by another process the
323 executed process waits until the current
324 controlling process releases the
325 terminal.
326 <option>tty-force</option>
327 is similar to <option>tty</option>,
328 but the executed process is forcefully
329 and immediately made the controlling
330 process of the terminal, potentially
331 removing previous controlling
332 processes from the
333 terminal. <option>tty-fail</option> is
334 similar to <option>tty</option> but if
335 the terminal already has a controlling
336 process start-up of the executed
337 process fails. The
338 <option>socket</option> option is only
339 valid in socket-activated services,
340 and only when the socket configuration
341 file (see
342 <citerefentry><refentrytitle>systemd.socket</refentrytitle><manvolnum>5</manvolnum></citerefentry>
343 for details) specifies a single socket
344 only. If this option is set standard
345 input will be connected to the socket
346 the service was activated from, which
347 is primarily useful for compatibility
348 with daemons designed for use with the
349 traditional
350 <citerefentry><refentrytitle>inetd</refentrytitle><manvolnum>8</manvolnum></citerefentry>
351 daemon. This setting defaults to
352 <option>null</option>.</para></listitem>
353 </varlistentry>
354 <varlistentry>
355 <term><varname>StandardOutput=</varname></term>
356 <listitem><para>Controls where file
357 descriptor 1 (STDOUT) of the executed
358 processes is connected to. Takes one
359 of <option>inherit</option>,
360 <option>null</option>,
361 <option>tty</option>,
362 <option>syslog</option>,
363 <option>kmsg</option>,
364 <option>kmsg+console</option>,
365 <option>syslog+console</option> or
366 <option>socket</option>. If set to
367 <option>inherit</option> the file
368 descriptor of standard input is
369 duplicated for standard output. If set
370 to <option>null</option> standard
371 output will be connected to
372 <filename>/dev/null</filename>,
373 i.e. everything written to it will be
374 lost. If set to <option>tty</option>
375 standard output will be connected to a
376 tty (as configured via
377 <varname>TTYPath=</varname>, see
378 below). If the TTY is used for output
379 only the executed process will not
380 become the controlling process of the
381 terminal, and will not fail or wait
382 for other processes to release the
383 terminal. <option>syslog</option>
384 connects standard output to the
385 <citerefentry><refentrytitle>syslog</refentrytitle><manvolnum>3</manvolnum></citerefentry>
386 system logger. <option>kmsg</option>
387 connects it with the kernel log buffer
388 which is accessible via
389 <citerefentry><refentrytitle>dmesg</refentrytitle><manvolnum>1</manvolnum></citerefentry>. <option>syslog+console</option>
390 and <option>kmsg+console</option> work
391 similarly but copy the output to the
392 system console as
393 well. <option>socket</option> connects
394 standard output to a socket from
395 socket activation, semantics are
396 similar to the respective option of
397 <varname>StandardInput=</varname>.
398 This setting defaults to
399 <option>inherit</option>.</para></listitem>
400 </varlistentry>
401 <varlistentry>
402 <term><varname>StandardError=</varname></term>
403 <listitem><para>Controls where file
404 descriptor 2 (STDERR) of the executed
405 processes is connected to. The
406 available options are identical to
407 those of
408 <varname>StandardOutput=</varname>,
409 with one exception: if set to
410 <option>inherit</option> the file
411 descriptor used for standard output is
412 duplicated for standard error. This
413 setting defaults to
414 <option>inherit</option>.</para></listitem>
415 </varlistentry>
416 <varlistentry>
417 <term><varname>TTYPath=</varname></term>
418 <listitem><para>Sets the terminal
419 device node to use if standard input,
420 output or stderr are connected to a
421 TTY (see above). Defaults to
422 <filename>/dev/console</filename>.</para></listitem>
423 </varlistentry>
424 <varlistentry>
425 <term><varname>TTYReset=</varname></term>
426 <listitem><para>Reset the terminal
427 device specified with
428 <varname>TTYPath=</varname> before and
429 after execution. Defaults to
430 <literal>no</literal>.</para></listitem>
431 </varlistentry>
432 <varlistentry>
433 <term><varname>TTYVHangup=</varname></term>
434 <listitem><para>Disconnect all clients
435 which have opened the terminal device
436 specified with
437 <varname>TTYPath=</varname>
438 before and after execution. Defaults
439 to
440 <literal>no</literal>.</para></listitem>
441 </varlistentry>
442 <varlistentry>
443 <term><varname>TTYVTDisallocate=</varname></term>
444 <listitem><para>If the the terminal
445 device specified with
446 <varname>TTYPath=</varname> is a
447 virtual console terminal try to
448 deallocate the TTY before and after
449 execution. This ensures that the
450 screen and scrollback buffer is
451 cleared. Defaults to
452 <literal>no</literal>.</para></listitem>
453 </varlistentry>
454 <varlistentry>
455 <term><varname>SyslogIdentifier=</varname></term>
456 <listitem><para>Sets the process name
457 to prefix log lines sent to syslog or
458 the kernel log buffer with. If not set
459 defaults to the process name of the
460 executed process. This option is only
461 useful when
462 <varname>StandardOutput=</varname> or
463 <varname>StandardError=</varname> are
464 set to <option>syslog</option> or
465 <option>kmsg</option>.</para></listitem>
466 </varlistentry>
467 <varlistentry>
468 <term><varname>SyslogFacility=</varname></term>
469 <listitem><para>Sets the syslog
470 facility to use when logging to
471 syslog. One of <option>kern</option>,
472 <option>user</option>,
473 <option>mail</option>,
474 <option>daemon</option>,
475 <option>auth</option>,
476 <option>syslog</option>,
477 <option>lpr</option>,
478 <option>news</option>,
479 <option>uucp</option>,
480 <option>cron</option>,
481 <option>authpriv</option>,
482 <option>ftp</option>,
483 <option>local0</option>,
484 <option>local1</option>,
485 <option>local2</option>,
486 <option>local3</option>,
487 <option>local4</option>,
488 <option>local5</option>,
489 <option>local6</option> or
490 <option>local7</option>. See
491 <citerefentry><refentrytitle>syslog</refentrytitle><manvolnum>3</manvolnum></citerefentry>
492 for details. This option is only
493 useful when
494 <varname>StandardOutput=</varname> or
495 <varname>StandardError=</varname> are
496 set to <option>syslog</option>.
497 Defaults to
498 <option>daemon</option>.</para></listitem>
499 </varlistentry>
500 <varlistentry>
501 <term><varname>SyslogLevel=</varname></term>
502 <listitem><para>Default syslog level
503 to use when logging to syslog or the
504 kernel log buffer. One of
505 <option>emerg</option>,
506 <option>alert</option>,
507 <option>crit</option>,
508 <option>err</option>,
509 <option>warning</option>,
510 <option>notice</option>,
511 <option>info</option>,
512 <option>debug</option>. See
513 <citerefentry><refentrytitle>syslog</refentrytitle><manvolnum>3</manvolnum></citerefentry>
514 for details. This option is only
515 useful when
516 <varname>StandardOutput=</varname> or
517 <varname>StandardError=</varname> are
518 set to <option>syslog</option> or
519 <option>kmsg</option>. Note that
520 individual lines output by the daemon
521 might be prefixed with a different log
522 level which can be used to override
523 the default log level specified
524 here. The interpretation of these
525 prefixes may be disabled with
526 <varname>SyslogLevelPrefix=</varname>,
527 see below. For details see
528 <citerefentry><refentrytitle>sd-daemon</refentrytitle><manvolnum>7</manvolnum></citerefentry>.
529
530 Defaults to
531 <option>info</option>.</para></listitem>
532 </varlistentry>
533
534 <varlistentry>
535 <term><varname>SyslogLevelPrefix=</varname></term>
536 <listitem><para>Takes a boolean
537 argument. If true and
538 <varname>StandardOutput=</varname> or
539 <varname>StandardError=</varname> are
540 set to <option>syslog</option> or
541 <option>kmsg</option> log lines
542 written by the executed process that
543 are prefixed with a log level will be
544 passed on to syslog with this log
545 level set but the prefix removed. If
546 set to false, the interpretation of
547 these prefixes is disabled and the
548 logged lines are passed on as-is. For
549 details about this prefixing see
550 <citerefentry><refentrytitle>sd-daemon</refentrytitle><manvolnum>7</manvolnum></citerefentry>.
551 Defaults to true.</para></listitem>
552 </varlistentry>
553
554 <varlistentry>
555 <term><varname>TimerSlackNSec=</varname></term>
556 <listitem><para>Sets the timer slack
557 in nanoseconds for the executed
558 processes. The timer slack controls the
559 accuracy of wake-ups triggered by
560 timers. See
561 <citerefentry><refentrytitle>prctl</refentrytitle><manvolnum>2</manvolnum></citerefentry>
562 for more information. Note that in
563 contrast to most other time span
564 definitions this parameter takes an
565 integer value in nano-seconds and does
566 not understand any other
567 units.</para></listitem>
568 </varlistentry>
569
570 <varlistentry>
571 <term><varname>LimitCPU=</varname></term>
572 <term><varname>LimitFSIZE=</varname></term>
573 <term><varname>LimitDATA=</varname></term>
574 <term><varname>LimitSTACK=</varname></term>
575 <term><varname>LimitCORE=</varname></term>
576 <term><varname>LimitRSS=</varname></term>
577 <term><varname>LimitNOFILE=</varname></term>
578 <term><varname>LimitAS=</varname></term>
579 <term><varname>LimitNPROC=</varname></term>
580 <term><varname>LimitMEMLOCK=</varname></term>
581 <term><varname>LimitLOCKS=</varname></term>
582 <term><varname>LimitSIGPENDING=</varname></term>
583 <term><varname>LimitMSGQUEUE=</varname></term>
584 <term><varname>LimitNICE=</varname></term>
585 <term><varname>LimitRTPRIO=</varname></term>
586 <term><varname>LimitRTTIME=</varname></term>
587 <listitem><para>These settings control
588 various resource limits for executed
589 processes. See
590 <citerefentry><refentrytitle>setrlimit</refentrytitle><manvolnum>2</manvolnum></citerefentry>
591 for details. Use the string
592 <varname>infinity</varname> to
593 configure no limit on a specific
594 resource.</para></listitem>
595 </varlistentry>
596
597 <varlistentry>
598 <term><varname>PAMName=</varname></term>
599 <listitem><para>Sets the PAM service
600 name to set up a session as. If set
601 the executed process will be
602 registered as a PAM session under the
603 specified service name. This is only
604 useful in conjunction with the
605 <varname>User=</varname> setting. If
606 not set no PAM session will be opened
607 for the executed processes. See
608 <citerefentry><refentrytitle>pam</refentrytitle><manvolnum>8</manvolnum></citerefentry>
609 for details.</para></listitem>
610 </varlistentry>
611
612 <varlistentry>
613 <term><varname>TCPWrapName=</varname></term>
614 <listitem><para>If this is a
615 socket-activated service this sets the
616 tcpwrap service name to check the
617 permission for the current connection
618 with. This is only useful in
619 conjunction with socket-activated
620 services, and stream sockets (TCP) in
621 particular. It has no effect on other
622 socket types (e.g. datagram/UDP) and on processes
623 unrelated to socket-based
624 activation. If the tcpwrap
625 verification fails daemon start-up
626 will fail and the connection is
627 terminated. See
628 <citerefentry><refentrytitle>tcpd</refentrytitle><manvolnum>8</manvolnum></citerefentry>
629 for details.</para></listitem>
630 </varlistentry>
631
632 <varlistentry>
633 <term><varname>CapabilityBoundingSet=</varname></term>
634
635 <listitem><para>Controls which
636 capabilities to include in the
637 capability bounding set for the
638 executed process. See
639 <citerefentry><refentrytitle>capabilities</refentrytitle><manvolnum>7</manvolnum></citerefentry>
640 for details. Takes a whitespace
641 separated list of capability names as
642 read by
643 <citerefentry><refentrytitle>cap_from_name</refentrytitle><manvolnum>3</manvolnum></citerefentry>.
644 Capabilities listed will be included
645 in the bounding set, all others are
646 removed. If the list of capabilities
647 is prefixed with ~ all but the listed
648 capabilities will be included, the
649 effect of the assignment
650 inverted. Note that this option does
651 not actually set or unset any
652 capabilities in the effective,
653 permitted or inherited capability
654 sets. That's what
655 <varname>Capabilities=</varname> is
656 for. If this option is not used the
657 capability bounding set is not
658 modified on process execution, hence
659 no limits on the capabilities of the
660 process are enforced.</para></listitem>
661 </varlistentry>
662
663 <varlistentry>
664 <term><varname>SecureBits=</varname></term>
665 <listitem><para>Controls the secure
666 bits set for the executed process. See
667 <citerefentry><refentrytitle>capabilities</refentrytitle><manvolnum>7</manvolnum></citerefentry>
668 for details. Takes a list of strings:
669 <option>keep-caps</option>,
670 <option>keep-caps-locked</option>,
671 <option>no-setuid-fixup</option>,
672 <option>no-setuid-fixup-locked</option>,
673 <option>no-setuid-noroot</option> and/or
674 <option>no-setuid-noroot-locked</option>.
675 </para></listitem>
676 </varlistentry>
677
678 <varlistentry>
679 <term><varname>Capabilities=</varname></term>
680 <listitem><para>Controls the
681 <citerefentry><refentrytitle>capabilities</refentrytitle><manvolnum>7</manvolnum></citerefentry>
682 set for the executed process. Take a
683 capability string describing the
684 effective, permitted and inherited
685 capability sets as documented in
686 <citerefentry><refentrytitle>cap_from_text</refentrytitle><manvolnum>3</manvolnum></citerefentry>.
687 Note that these capability sets are
688 usually influenced by the capabilities
689 attached to the executed file. Due to
690 that
691 <varname>CapabilityBoundingSet=</varname>
692 is probably the much more useful
693 setting.</para></listitem>
694 </varlistentry>
695
696 <varlistentry>
697 <term><varname>ControlGroup=</varname></term>
698
699 <listitem><para>Controls the control
700 groups the executed processes shall be
701 made members of. Takes a
702 space-separated list of cgroup
703 identifiers. A cgroup identifier has a
704 format like
705 <filename>cpu:/foo/bar</filename>,
706 where "cpu" identifies the kernel
707 control group controller used, and
708 <filename>/foo/bar</filename> is the
709 control group path. The controller
710 name and ":" may be omitted in which
711 case the named systemd control group
712 hierarchy is implied. Alternatively,
713 the path and ":" may be omitted, in
714 which case the default control group
715 path for this unit is implied. This
716 option may be used to place executed
717 processes in arbitrary groups in
718 arbitrary hierarchies -- which can be
719 configured externally with additional
720 execution limits. By default systemd
721 will place all executed processes in
722 separate per-unit control groups
723 (named after the unit) in the systemd
724 named hierarchy. Since every process
725 can be in one group per hierarchy only
726 overriding the control group path in
727 the named systemd hierarchy will
728 disable automatic placement in the
729 default group. This option is
730 primarily intended to place executed
731 processes in specific paths in
732 specific kernel controller
733 hierarchies. It is however not
734 recommended to manipulate the service
735 control group path in the systemd
736 named hierarchy. For details about
737 control groups see <ulink
738 url="http://www.kernel.org/doc/Documentation/cgroups/cgroups.txt">cgroups.txt</ulink>.</para></listitem>
739 </varlistentry>
740
741 <varlistentry>
742 <term><varname>ControlGroupModify=</varname></term>
743 <listitem><para>Takes a boolean
744 argument. If true, the control groups
745 created for this unit will be owned by
746 the user specified with
747 <varname>User=</varname> (and the
748 appropriate group), and he/she can create
749 subgroups as well as add processes to
750 the group.</para></listitem>
751 </varlistentry>
752
753 <varlistentry>
754 <term><varname>ControlGroupAttribute=</varname></term>
755
756 <listitem><para>Set a specific control
757 group attribute for executed
758 processes, and (if needed) add the the
759 executed processes to a cgroup in the
760 hierarchy of the controller the
761 attribute belongs to. Takes two
762 space-separated arguments: the
763 attribute name (syntax is
764 <literal>cpu.shares</literal> where
765 <literal>cpu</literal> refers to a
766 specific controller and
767 <literal>shares</literal> to the
768 attribute name), and the attribute
769 value. Example:
770 <literal>ControlGroupAttribute=cpu.shares
771 512</literal>. If this option is used
772 for an attribute that belongs to a
773 kernel controller hierarchy the unit
774 is not already configured to be added
775 to (for example via the
776 <literal>ControlGroup=</literal>
777 option) then the unit will be added to
778 the controller and the default unit
779 cgroup path is implied. Thus, using
780 <varname>ControlGroupAttribute=</varname>
781 is in most case sufficient to make use
782 of control group enforcements,
783 explicit
784 <varname>ControlGroup=</varname> are
785 only necessary in case the implied
786 default control group path for a
787 service is not desirable. For details
788 about control group attributes see
789 <ulink
790 url="http://www.kernel.org/doc/Documentation/cgroups/cgroups.txt">cgroups.txt</ulink>. This
791 option may appear more than once, in
792 order to set multiple control group
793 attributes.</para></listitem>
794 </varlistentry>
795
796 <varlistentry>
797 <term><varname>CPUShares=</varname></term>
798
799 <listitem><para>Assign the specified
800 overall CPU time shares to the processes executed. Takes
801 an integer value. This controls the
802 <literal>cpu.shares</literal> control
803 group attribute. For details about
804 this control group attribute see <ulink
805 url="http://www.kernel.org/doc/Documentation/scheduler/sched-design-CFS.txt">sched-design-CFS.txt</ulink>.</para></listitem>
806 </varlistentry>
807
808 <varlistentry>
809 <term><varname>MemoryLimit=</varname></term>
810 <term><varname>MemorySoftLimit=</varname></term>
811
812 <listitem><para>Limit the overall memory usage
813 of the executed processes to a certain
814 size. Takes a memory size in bytes. If
815 the value is suffixed with K, M, G or
816 T the specified memory size is parsed
817 as Kilobytes, Megabytes, Gigabytes
818 resp. Terabytes (to the base
819 1024). This controls the
820 <literal>memory.limit_in_bytes</literal>
821 and
822 <literal>memory.soft_limit_in_bytes</literal>
823 control group attributes. For details
824 about these control group attributes
825 see <ulink
826 url="http://www.kernel.org/doc/Documentation/cgroups/memory.txt">memory.txt</ulink>.</para></listitem>
827 </varlistentry>
828
829 <varlistentry>
830 <term><varname>DeviceAllow=</varname></term>
831 <term><varname>DeviceDeny=</varname></term>
832
833 <listitem><para>Control access to
834 specific device nodes by the executed processes. Takes two
835 space separated strings: a device node
836 path (such as
837 <filename>/dev/null</filename>)
838 followed by a combination of r, w, m
839 to control reading, writing resp.
840 creating of the specific device node
841 by the unit. This controls the
842 <literal>devices.allow</literal>
843 and
844 <literal>devices.deny</literal>
845 control group attributes. For details
846 about these control group attributes
847 see <ulink
848 url="http://www.kernel.org/doc/Documentation/cgroups/devices.txt">devices.txt</ulink>.</para></listitem>
849 </varlistentry>
850
851 <varlistentry>
852 <term><varname>ReadWriteDirectories=</varname></term>
853 <term><varname>ReadOnlyDirectories=</varname></term>
854 <term><varname>InaccessibleDirectories=</varname></term>
855
856 <listitem><para>Sets up a new
857 file-system name space for executed
858 processes. These options may be used
859 to limit access a process might have
860 to the main file-system
861 hierarchy. Each setting takes a
862 space-separated list of absolute
863 directory paths. Directories listed in
864 <varname>ReadWriteDirectories=</varname>
865 are accessible from within the
866 namespace with the same access rights
867 as from outside. Directories listed in
868 <varname>ReadOnlyDirectories=</varname>
869 are accessible for reading only,
870 writing will be refused even if the
871 usual file access controls would
872 permit this. Directories listed in
873 <varname>InaccessibleDirectories=</varname>
874 will be made inaccessible for processes
875 inside the namespace. Note that
876 restricting access with these options
877 does not extend to submounts of a
878 directory. You must list submounts
879 separately in these settings to
880 ensure the same limited access. These
881 options may be specified more than
882 once in which case all directories
883 listed will have limited access from
884 within the
885 namespace.</para></listitem>
886 </varlistentry>
887
888 <varlistentry>
889 <term><varname>PrivateTmp=</varname></term>
890
891 <listitem><para>Takes a boolean
892 argument. If true sets up a new file
893 system namespace for the executed
894 processes and mounts a private
895 <filename>/tmp</filename> directory
896 inside it, that is not shared by
897 processes outside of the
898 namespace. This is useful to secure
899 access to temporary files of the
900 process, but makes sharing between
901 processes via
902 <filename>/tmp</filename>
903 impossible. Defaults to
904 false.</para></listitem>
905 </varlistentry>
906
907 <varlistentry>
908 <term><varname>PrivateNetwork=</varname></term>
909
910 <listitem><para>Takes a boolean
911 argument. If true sets up a new
912 network namespace for the executed
913 processes and configures only the
914 loopback network device
915 <literal>lo</literal> inside it. No
916 other network devices will be
917 available to the executed process.
918 This is useful to securely turn off
919 network access by the executed
920 process. Defaults to
921 false.</para></listitem>
922 </varlistentry>
923
924 <varlistentry>
925 <term><varname>MountFlags=</varname></term>
926
927 <listitem><para>Takes a mount
928 propagation flag:
929 <option>shared</option>,
930 <option>slave</option> or
931 <option>private</option>, which
932 control whether namespaces set up with
933 <varname>ReadWriteDirectories=</varname>,
934 <varname>ReadOnlyDirectories=</varname>
935 and
936 <varname>InaccessibleDirectories=</varname>
937 receive or propagate new mounts
938 from/to the main namespace. See
939 <citerefentry><refentrytitle>mount</refentrytitle><manvolnum>1</manvolnum></citerefentry>
940 for details. Defaults to
941 <option>shared</option>, i.e. the new
942 namespace will both receive new mount
943 points from the main namespace as well
944 as propagate new mounts to
945 it.</para></listitem>
946 </varlistentry>
947
948 <varlistentry>
949 <term><varname>UtmpIdentifier=</varname></term>
950
951 <listitem><para>Takes a a four
952 character identifier string for an
953 utmp/wtmp entry for this service. This
954 should only be set for services such
955 as <command>getty</command>
956 implementations where utmp/wtmp
957 entries must be created and cleared
958 before and after execution. If the
959 configured string is longer than four
960 characters it is truncated and the
961 terminal four characters are
962 used. This setting interprets %I style
963 string replacements. This setting is
964 unset by default, i.e. no utmp/wtmp
965 entries are created or cleaned up for
966 this service.</para></listitem>
967 </varlistentry>
968
969 </variablelist>
970 </refsect1>
971
972 <refsect1>
973 <title>See Also</title>
974 <para>
975 <citerefentry><refentrytitle>systemd</refentrytitle><manvolnum>1</manvolnum></citerefentry>,
976 <citerefentry><refentrytitle>systemctl</refentrytitle><manvolnum>8</manvolnum></citerefentry>,
977 <citerefentry><refentrytitle>systemd.unit</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
978 <citerefentry><refentrytitle>systemd.service</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
979 <citerefentry><refentrytitle>systemd.socket</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
980 <citerefentry><refentrytitle>systemd.swap</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
981 <citerefentry><refentrytitle>systemd.mount</refentrytitle><manvolnum>5</manvolnum></citerefentry>
982 </para>
983 </refsect1>
984
985 </refentry>