1SYSTEMD-RUN(1) systemd-run SYSTEMD-RUN(1)
2
6 systemd-run - Run programs in transient scope units, service units, or
7 path-, socket-, or timer-triggered service units
8
10 systemd-run [OPTIONS...] COMMAND [ARGS...]
11 systemd-run [OPTIONS...] [PATH OPTIONS...] {COMMAND} [ARGS...]
13 systemd-run [OPTIONS...] [SOCKET OPTIONS...] {COMMAND} [ARGS...]
15 systemd-run [OPTIONS...] [TIMER OPTIONS...] {COMMAND} [ARGS...]
17
19 systemd-run may be used to create and start a transient .service or
20 .scope unit and run the specified COMMAND in it. It may also be used to
21 create and start a transient .path, .socket, or .timer unit, that
22 activates a .service unit when elapsing.
23 If a command is run as transient service unit, it will be started and
25 managed by the service manager like any other service, and thus shows
26 up in the output of systemctl list-units like any other unit. It will
27 run in a clean and detached execution environment, with the service
28 manager as its parent process. In this mode, systemd-run will start the
29 service asynchronously in the background and return after the command
30 has begun execution (unless --no-block or --wait are specified, see
31 below).
32 If a command is run as transient scope unit, it will be executed by
34 systemd-run itself as parent process and will thus inherit the
35 execution environment of the caller. However, the processes of the
36 command are managed by the service manager similar to normal services,
37 and will show up in the output of systemctl list-units. Execution in
38 this case is synchronous, and will return only when the command
39 finishes. This mode is enabled via the --scope switch (see below).
40 If a command is run with path, socket, or timer options such as
42 --on-calendar= (see below), a transient path, socket, or timer unit is
43 created alongside the service unit for the specified command. Only the
44 transient path, socket, or timer unit is started immediately, the
45 transient service unit will be triggered by the path, socket, or timer
46 unit. If the --unit= option is specified, the COMMAND may be omitted.
47 In this case, systemd-run creates only a .path, .socket, or .timer unit
48 that triggers the specified unit.
49
51 The following options are understood:
52 --no-ask-password
54 Do not query the user for authentication for privileged operations.
55 --scope
57 Create a transient .scope unit instead of the default transient
58 .service unit (see above).
59 --unit=
61 Use this unit name instead of an automatically generated one.
62 --property=, -p
64 Sets a property on the scope or service unit that is created. This
65 option takes an assignment in the same format as systemctl(1)'s
66 set-property command.
67 --description=
69 Provide a description for the service, scope, path, socket, or
70 timer unit. If not specified, the command itself will be used as a
71 description. See Description= in systemd.unit(5).
72 --slice=
74 Make the new .service or .scope unit part of the specified slice,
75 instead of system.slice.
76 -r, --remain-after-exit
78 After the service process has terminated, keep the service around
79 until it is explicitly stopped. This is useful to collect runtime
80 information about the service after it finished running. Also see
81 RemainAfterExit= in systemd.service(5).
82 --send-sighup
84 When terminating the scope or service unit, send a SIGHUP
85 immediately after SIGTERM. This is useful to indicate to shells and
86 shell-like processes that the connection has been severed. Also see
87 SendSIGHUP= in systemd.kill(5).
88 --service-type=
90 Sets the service type. Also see Type= in systemd.service(5). This
91 option has no effect in conjunction with --scope. Defaults to
92 simple.
93 --uid=, --gid=
95 Runs the service process under the specified UNIX user and group.
96 Also see User= and Group= in systemd.exec(5).
97 --nice=
99 Runs the service process with the specified nice level. Also see
100 Nice= in systemd.exec(5).
101 -E NAME=VALUE, --setenv=NAME=VALUE
103 Runs the service process with the specified environment variable
104 set. Also see Environment= in systemd.exec(5).
105 --pty, -t
107 When invoking the command, the transient service connects its
108 standard input, output and error to the terminal systemd-run is
109 invoked on, via a pseudo TTY device. This allows running programs
110 that expect interactive user input/output as services, such as
111 interactive command shells.
112 Note that machinectl(1)'s shell command is usually a better
114 alternative for requesting a new, interactive login session on the
115 local host or a local container.
116 See below for details on how this switch combines with --pipe.
118 --pipe, -P
120 If specified, standard input, output, and error of the transient
121 service are inherited from the systemd-run command itself. This
122 allows systemd-run to be used within shell pipelines. Note that
123 this mode is not suitable for interactive command shells and
124 similar, as the service process will not become a TTY controller
125 when invoked on a terminal. Use --pty instead in that case.
126 When both --pipe and --pty are used in combination the more
128 appropriate option is automatically determined and used.
129 Specifically, when invoked with standard input, output and error
130 connected to a TTY --pty is used, and otherwise --pipe.
131 When this option is used the original file descriptors systemd-run
133 receives are passed to the service processes as-is. If the service
134 runs with different privileges than systemd-run, this means the
135 service might not be able to re-open the passed file descriptors,
136 due to normal file descriptor access restrictions. If the invoked
137 process is a shell script that uses the echo "hello" > /dev/stderr
138 construct for writing messages to stderr, this might cause
139 problems, as this only works if stderr can be re-opened. To
140 mitigate this use the construct echo "hello" >&2 instead, which is
141 mostly equivalent and avoids this pitfall.
142 --quiet, -q
144 Suppresses additional informational output while running. This is
145 particularly useful in combination with --pty when it will suppress
146 the initial message explaining how to terminate the TTY connection.
147 --on-active=, --on-boot=, --on-startup=, --on-unit-active=,
149 --on-unit-inactive=
150 Defines a monotonic timer relative to different starting points for
151 starting the specified command. See OnActiveSec=, OnBootSec=,
152 OnStartupSec=, OnUnitActiveSec= and OnUnitInactiveSec= in
153 systemd.timer(5) for details. These options are shortcuts for
154 --timer-property= with the relevant properties. These options may
155 not be combined with --scope or --pty.
156 --on-calendar=
158 Defines a calendar timer for starting the specified command. See
159 OnCalendar= in systemd.timer(5). This option is a shortcut for
160 --timer-property=OnCalendar=. This option may not be combined with
161 --scope or --pty.
162 --path-property=, --socket-property=, --timer-property=
164 Sets a property on the path, socket, or timer unit that is created.
165 This option is similar to --property= but applies to the transient
166 path, socket, or timer unit rather than the transient service unit
167 created. This option takes an assignment in the same format as
168 systemctl(1)'s set-property command. These options may not be
169 combined with --scope or --pty.
170 --no-block
172 Do not synchronously wait for the unit start operation to finish.
173 If this option is not specified, the start request for the
174 transient unit will be verified, enqueued and systemd-run will wait
175 until the unit's start-up is completed. By passing this argument,
176 it is only verified and enqueued. This option may not be combined
177 with --wait.
178 --wait
180 Synchronously wait for the transient service to terminate. If this
181 option is specified, the start request for the transient unit is
182 verified, enqueued, and waited for. Subsequently the invoked unit
183 is monitored, and it is waited until it is deactivated again (most
184 likely because the specified command completed). On exit, terse
185 information about the unit's runtime is shown, including total
186 runtime (as well as CPU usage, if --property=CPUAccounting=1 was
187 set) and the exit code and status of the main process. This output
188 may be suppressed with --quiet. This option may not be combined
189 with --no-block, --scope or the various path, socket, or timer
190 options.
191 -G, --collect
193 Unload the transient unit after it completed, even if it failed.
194 Normally, without this option, all units that ran and failed are
195 kept in memory until the user explicitly resets their failure state
196 with systemctl reset-failed or an equivalent command. On the other
197 hand, units that ran successfully are unloaded immediately. If this
198 option is turned on the "garbage collection" of units is more
199 aggressive, and unloads units regardless if they exited
200 successfully or failed. This option is a shortcut for
201 --property=CollectMode=inactive-or-failed, see the explanation for
202 CollectMode= in systemd.unit(5) for further information.
203 --user
205 Talk to the service manager of the calling user, rather than the
206 service manager of the system.
207 --system
209 Talk to the service manager of the system. This is the implied
210 default.
211 -H, --host=
213 Execute the operation remotely. Specify a hostname, or a username
214 and hostname separated by "@", to connect to. The hostname may
215 optionally be suffixed by a container name, separated by ":", which
216 connects directly to a specific container on the specified host.
217 This will use SSH to talk to the remote machine manager instance.
218 Container names may be enumerated with machinectl -H HOST.
219 -M, --machine=
221 Execute operation on a local container. Specify a container name to
222 connect to.
223 -h, --help
225 Print a short help text and exit.
226 --version
228 Print a short version string and exit.
229 All command line arguments after the first non-option argument become
231 part of the command line of the launched process. If a command is run
232 as service unit, the first argument needs to be an absolute program
233 path.
234
236 On success, 0 is returned, a non-zero failure code otherwise.
237
239 Example 1. Logging environment variables provided by systemd to
240 services
241 # systemd-run env
243 Running as unit: run-19945.service
244 # journalctl -u run-19945.service
245 Sep 08 07:37:21 bupkis systemd[1]: Starting /usr/bin/env...
246 Sep 08 07:37:21 bupkis systemd[1]: Started /usr/bin/env.
247 Sep 08 07:37:21 bupkis env[19948]: PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin
248 Sep 08 07:37:21 bupkis env[19948]: LANG=en_US.UTF-8
249 Sep 08 07:37:21 bupkis env[19948]: BOOT_IMAGE=/vmlinuz-3.11.0-0.rc5.git6.2.fc20.x86_64
250 Example 2. Limiting resources available to a command
252 # systemd-run -p BlockIOWeight=10 updatedb
254 This command invokes the updatedb(8) tool, but lowers the block I/O
256 weight for it to 10. See systemd.resource-control(5) for more
257 information on the BlockIOWeight= property.
258 Example 3. Running commands at a specified time
260 The following command will touch a file after 30 seconds.
262 # date; systemd-run --on-active=30 --timer-property=AccuracySec=100ms /bin/touch /tmp/foo
264 Mon Dec 8 20:44:24 KST 2014
265 Running as unit: run-71.timer
266 Will run service as unit: run-71.service
267 # journalctl -b -u run-71.timer
268 -- Logs begin at Fri 2014-12-05 19:09:21 KST, end at Mon 2014-12-08 20:44:54 KST. --
269 Dec 08 20:44:38 container systemd[1]: Starting /bin/touch /tmp/foo.
270 Dec 08 20:44:38 container systemd[1]: Started /bin/touch /tmp/foo.
271 # journalctl -b -u run-71.service
272 -- Logs begin at Fri 2014-12-05 19:09:21 KST, end at Mon 2014-12-08 20:44:54 KST. --
273 Dec 08 20:44:48 container systemd[1]: Starting /bin/touch /tmp/foo...
274 Dec 08 20:44:48 container systemd[1]: Started /bin/touch /tmp/foo.
275 Example 4. Allowing access to the tty
277 The following command invokes /bin/bash as a service passing its
279 standard input, output and error to the calling TTY.
280 # systemd-run -t --send-sighup /bin/bash
282 Example 5. Start screen as a user service
284 $ systemd-run --scope --user screen
286 Running scope as unit run-r14b0047ab6df45bfb45e7786cc839e76.scope.
287 $ screen -ls
289 There is a screen on:
290 492..laptop (Detached)
291 1 Socket in /var/run/screen/S-fatima.
292 This starts the screen process as a child of the systemd --user process
294 that was started by user@.service, in a scope unit. A systemd.scope(5)
295 unit is used instead of a systemd.service(5) unit, because screen will
296 exit when detaching from the terminal, and a service unit would be
297 terminated. Running screen as a user unit has the advantage that it is
298 not part of the session scope. If KillUserProcesses=yes is configured
299 in logind.conf(5), the default, the session scope will be terminated
300 when the user logs out of that session.
301 The user@.service is started automatically when the user first logs in,
303 and stays around as long as at least one login session is open. After
304 the user logs out of the last session, user@.service and all services
305 underneath it are terminated. This behavior is the default, when
306 "lingering" is not enabled for that user. Enabling lingering means that
307 user@.service is started automatically during boot, even if the user is
308 not logged in, and that the service is not terminated when the user
309 logs out.
310 Enabling lingering allows the user to run processes without being
312 logged in, for example to allow screen to persist after the user logs
313 out, even if the session scope is terminated. In the default
314 configuration, users can enable lingering for themselves:
315 $ loginctl enable-linger
317
319 systemd(1), systemctl(1), systemd.unit(5), systemd.service(5),
320 systemd.scope(5), systemd.slice(5), systemd.exec(5), systemd.resource-
321 control(5), systemd.timer(5), systemd-mount(1), machinectl(1)
322systemd 239 SYSTEMD-RUN(1)