1podman-run(1) General Commands Manual podman-run(1)
2
6 podman-run - Run a command in a new container
7
10 podman run [options] image [command [arg ...]]
11 podman container run [options] image [command [arg ...]]
14
17 Run a process in a new container. podman run starts a process with its
18 own file system, its own networking, and its own isolated process tree.
19 The image which starts the process may define defaults related to the
20 process that will be run in the container, the networking to expose,
21 and more, but podman run gives final control to the operator or admin‐
22 istrator who starts the container from the image. For that reason pod‐
23 man run has more options than any other Podman command.
24 If the image is not already loaded then podman run will pull the image,
27 and all image dependencies, from the repository in the same way running
28 podman pull image , before it starts the container from that image.
29 Several files will be automatically created within the container. These
32 include /etc/hosts, /etc/hostname, and /etc/resolv.conf to manage net‐
33 working. These will be based on the host's version of the files,
34 though they can be customized with options (for example, --dns will
35 override the host's DNS servers in the created resolv.conf). Addition‐
36 ally, a container environment file is created in each container to in‐
37 dicate to programs they are running in a container. This file is lo‐
38 cated at /run/.containerenv (or /var/run/.containerenv for FreeBSD con‐
39 tainers). When using the --privileged flag the .containerenv contains
40 name/value pairs indicating the container engine version, whether the
41 engine is running in rootless mode, the container name and ID, as well
42 as the image name and ID that the container is based on. Note:
43 /run/.containerenv will not be created when a volume is mounted on
44 /run.
45 When running from a user defined network namespace, the
48 /etc/netns/NSNAME/resolv.conf will be used if it exists, otherwise
49 /etc/resolv.conf will be used.
50 Default settings are defined in containers.conf. Most settings for re‐
53 mote connections use the servers containers.conf, except when docu‐
54 mented in man pages.
55
58 The image is specified using transport:path format. If no transport is
59 specified, the docker (container registry) transport is used by de‐
60 fault. For remote Podman, including Mac and Windows (excluding WSL2)
61 machines, docker is the only allowed transport.
62 dir:path
65 An existing local directory path storing the manifest, layer tarballs
66 and signatures as individual files. This is a non-standardized format,
67 primarily useful for debugging or noninvasive container inspection.
68 $ podman save --format docker-dir fedora -o /tmp/fedora
70 $ podman run dir:/tmp/fedora echo hello
71 docker://docker-reference (Default)
75 An image reference stored in a remote container image registry. Exam‐
76 ple: "quay.io/podman/stable:latest". The reference can include a path
77 to a specific registry; if it does not, the registries listed in reg‐
78 istries.conf are queried to find a matching image. By default, creden‐
79 tials from podman login (stored at $XDG_RUNTIME_DIR/contain‐
80 ers/auth.json by default) are used to authenticate; otherwise it falls
81 back to using credentials in $HOME/.docker/config.json.
82 $ podman run registry.fedoraproject.org/fedora:latest echo hello
84 docker-archive:path[:docker-reference] An image stored in the docker
88 save formatted file. docker-reference is only used when creating such a
89 file, and it must not contain a digest.
90 $ podman save --format docker-archive fedora -o /tmp/fedora
92 $ podman run docker-archive:/tmp/fedora echo hello
93 docker-daemon:docker-reference
97 An image in docker-reference format stored in the docker daemon in‐
98 ternal storage. The docker-reference can also be an image ID (docker-
99 daemon:algo:digest).
100 $ sudo docker pull fedora
102 $ sudo podman run docker-daemon:docker.io/library/fedora echo hello
103 oci-archive:path:tag
107 An image in a directory compliant with the "Open Container Image Lay‐
108 out Specification" at the specified path and specified with a tag.
109 $ podman save --format oci-archive fedora -o /tmp/fedora
111 $ podman run oci-archive:/tmp/fedora echo hello
112
116 --add-host=host:ip
117 Add a custom host-to-IP mapping (host:ip)
118 Add a line to /etc/hosts. The format is hostname:ip. The --add-host op‐
121 tion can be set multiple times. Conflicts with the --no-hosts option.
122 --annotation=key=value
125 Add an annotation to the container. This option can be set multiple
126 times.
127 --arch=ARCH
130 Override the architecture, defaults to hosts, of the image to be
131 pulled. For example, arm. Unless overridden, subsequent lookups of the
132 same image in the local storage matches this architecture, regardless
133 of the host.
134 --attach, -a=stdin | stdout | stderr
137 Attach to STDIN, STDOUT or STDERR.
138 In foreground mode (the default when -d is not specified), podman run
141 can start the process in the container and attach the console to the
142 process's standard input, output, and error. It can even pretend to be
143 a TTY (this is what most command-line executables expect) and pass
144 along signals. The -a option can be set for each of stdin, stdout, and
145 stderr.
146 --authfile=path
149 Path of the authentication file. Default is ${XDG_RUNTIME_DIR}/contain‐
150 ers/auth.json on Linux, and $HOME/.config/containers/auth.json on Win‐
151 dows/macOS. The file is created by podman login. If the authorization
152 state is not found there, $HOME/.docker/config.json is checked, which
153 is set using docker login.
154 Note: There is also the option to override the default path of the au‐
157 thentication file by setting the REGISTRY_AUTH_FILE environment vari‐
158 able. This can be done with export REGISTRY_AUTH_FILE=path.
159 --blkio-weight=weight
162 Block IO relative weight. The weight is a value between 10 and 1000.
163 This option is not supported on cgroups V1 rootless systems.
166 --blkio-weight-device=device:weight
169 Block IO relative device weight.
170 --cap-add=capability
173 Add Linux capabilities.
174 --cap-drop=capability
177 Drop Linux capabilities.
178 --cgroup-conf=KEY=VALUE
181 When running on cgroup v2, specify the cgroup file to write to and its
182 value. For example --cgroup-conf=memory.high=1073741824 sets the mem‐
183 ory.high limit to 1GB.
184 --cgroup-parent=path
187 Path to cgroups under which the cgroup for the container is created. If
188 the path is not absolute, the path is considered to be relative to the
189 cgroups path of the init process. Cgroups are created if they do not
190 already exist.
191 --cgroupns=mode
194 Set the cgroup namespace mode for the container.
195 • host: use the host's cgroup namespace inside the container.
198 • container:id: join the namespace of the specified container.
200 • private: create a new cgroup namespace.
202 • ns:path: join the namespace at the specified path.
204 If the host uses cgroups v1, the default is set to host. On cgroups v2,
208 the default is private.
209 --cgroups=how
212 Determines whether the container creates CGroups.
213 Default is enabled.
216 The enabled option creates a new cgroup under the cgroup-parent. The
219 disabled option forces the container to not create CGroups, and thus
220 conflicts with CGroup options (--cgroupns and --cgroup-parent). The
221 no-conmon option disables a new CGroup only for the conmon process.
222 The split option splits the current CGroup in two sub-cgroups: one for
223 conmon and one for the container payload. It is not possible to set
224 --cgroup-parent with split.
225 --chrootdirs=path
228 Path to a directory inside the container that is treated as a chroot
229 directory. Any Podman managed file (e.g., /etc/resolv.conf,
230 /etc/hosts, etc/hostname) that is mounted into the root directory is
231 mounted into that location as well. Multiple directories are separated
232 with a comma.
233 --cidfile=file
236 Write the container ID to file. The file is removed along with the
237 container, except when used with podman --remote run on detached con‐
238 tainers.
239 --conmon-pidfile=file
242 Write the pid of the conmon process to a file. As conmon runs in a sep‐
243 arate process than Podman, this is necessary when using systemd to
244 restart Podman containers. (This option is not available with the re‐
245 mote Podman client, including Mac and Windows (excluding WSL2) ma‐
246 chines)
247 --cpu-period=limit
250 Set the CPU period for the Completely Fair Scheduler (CFS), which is a
251 duration in microseconds. Once the container's CPU quota is used up, it
252 will not be scheduled to run until the current period ends. Defaults to
253 100000 microseconds.
254 On some systems, changing the resource limits may not be allowed for
257 non-root users. For more details, see https://github.com/contain‐
258 ers/podman/blob/main/troubleshooting.md#26-running-containers-with-re‐
259 source-limits-fails-with-a-permissions-error
260 This option is not supported on cgroups V1 rootless systems.
263 --cpu-quota=limit
266 Limit the CPU Completely Fair Scheduler (CFS) quota.
267 Limit the container's CPU usage. By default, containers run with the
270 full CPU resource. The limit is a number in microseconds. If a number
271 is provided, the container is allowed to use that much CPU time until
272 the CPU period ends (controllable via --cpu-period).
273 On some systems, changing the resource limits may not be allowed for
276 non-root users. For more details, see https://github.com/contain‐
277 ers/podman/blob/main/troubleshooting.md#26-running-containers-with-re‐
278 source-limits-fails-with-a-permissions-error
279 This option is not supported on cgroups V1 rootless systems.
282 --cpu-rt-period=microseconds
285 Limit the CPU real-time period in microseconds.
286 Limit the container's Real Time CPU usage. This option tells the kernel
289 to restrict the container's Real Time CPU usage to the period speci‐
290 fied.
291 This option is only supported on cgroups V1 rootful systems.
294 --cpu-rt-runtime=microseconds
297 Limit the CPU real-time runtime in microseconds.
298 Limit the containers Real Time CPU usage. This option tells the kernel
301 to limit the amount of time in a given CPU period Real Time tasks may
302 consume. Ex: Period of 1,000,000us and Runtime of 950,000us means that
303 this container can consume 95% of available CPU and leave the remaining
304 5% to normal priority tasks.
305 The sum of all runtimes across containers cannot exceed the amount al‐
308 lotted to the parent cgroup.
309 This option is only supported on cgroups V1 rootful systems.
312 --cpu-shares, -c=shares
315 CPU shares (relative weight).
316 By default, all containers get the same proportion of CPU cycles. This
319 proportion can be modified by changing the container's CPU share
320 weighting relative to the combined weight of all the running contain‐
321 ers. Default weight is 1024.
322 The proportion only applies when CPU-intensive processes are running.
325 When tasks in one container are idle, other containers can use the
326 left-over CPU time. The actual amount of CPU time varies depending on
327 the number of containers running on the system.
328 For example, consider three containers, one has a cpu-share of 1024 and
331 two others have a cpu-share setting of 512. When processes in all three
332 containers attempt to use 100% of CPU, the first container receives 50%
333 of the total CPU time. If a fourth container is added with a cpu-share
334 of 1024, the first container only gets 33% of the CPU. The remaining
335 containers receive 16.5%, 16.5% and 33% of the CPU.
336 On a multi-core system, the shares of CPU time are distributed over all
339 CPU cores. Even if a container is limited to less than 100% of CPU
340 time, it can use 100% of each individual CPU core.
341 For example, consider a system with more than three cores. If the con‐
344 tainer C0 is started with --cpu-shares=512 running one process, and an‐
345 other container C1 with --cpu-shares=1024 running two processes, this
346 can result in the following division of CPU shares:
347 ┌────┬───────────┬─────┬──────────────┐
350 │PID │ container │ CPU │ CPU share │
351 ├────┼───────────┼─────┼──────────────┤
352 │100 │ C0 │ 0 │ 100% of CPU0 │
353 ├────┼───────────┼─────┼──────────────┤
354 │101 │ C1 │ 1 │ 100% of CPU1 │
355 ├────┼───────────┼─────┼──────────────┤
356 │102 │ C1 │ 2 │ 100% of CPU2 │
357 └────┴───────────┴─────┴──────────────┘
358 On some systems, changing the resource limits may not be allowed for
360 non-root users. For more details, see https://github.com/contain‐
361 ers/podman/blob/main/troubleshooting.md#26-running-containers-with-re‐
362 source-limits-fails-with-a-permissions-error
363 This option is not supported on cgroups V1 rootless systems.
366 --cpus=number
369 Number of CPUs. The default is 0.0 which means no limit. This is short‐
370 hand for --cpu-period and --cpu-quota, therefore the option cannot be
371 specified with --cpu-period or --cpu-quota.
372 On some systems, changing the CPU limits may not be allowed for non-
375 root users. For more details, see https://github.com/containers/pod‐
376 man/blob/main/troubleshooting.md#26-running-containers-with-resource-
377 limits-fails-with-a-permissions-error
378 This option is not supported on cgroups V1 rootless systems.
381 --cpuset-cpus=number
384 CPUs in which to allow execution. Can be specified as a comma-separated
385 list (e.g. 0,1), as a range (e.g. 0-3), or any combination thereof
386 (e.g. 0-3,7,11-15).
387 On some systems, changing the resource limits may not be allowed for
390 non-root users. For more details, see https://github.com/contain‐
391 ers/podman/blob/main/troubleshooting.md#26-running-containers-with-re‐
392 source-limits-fails-with-a-permissions-error
393 This option is not supported on cgroups V1 rootless systems.
396 --cpuset-mems=nodes
399 Memory nodes (MEMs) in which to allow execution (0-3, 0,1). Only effec‐
400 tive on NUMA systems.
401 If there are four memory nodes on the system (0-3), use --cpuset-
404 mems=0,1 then processes in the container only uses memory from the
405 first two memory nodes.
406 On some systems, changing the resource limits may not be allowed for
409 non-root users. For more details, see https://github.com/contain‐
410 ers/podman/blob/main/troubleshooting.md#26-running-containers-with-re‐
411 source-limits-fails-with-a-permissions-error
412 This option is not supported on cgroups V1 rootless systems.
415 --decryption-key=key[:passphrase]
418 The [key[:passphrase]] to be used for decryption of images. Key can
419 point to keys and/or certificates. Decryption is tried with all keys.
420 If the key is protected by a passphrase, it is required to be passed in
421 the argument and omitted otherwise.
422 --detach, -d
425 Detached mode: run the container in the background and print the new
426 container ID. The default is false.
427 At any time run podman ps in the other shell to view a list of the run‐
430 ning containers. Reattach to a detached container with podman attach
431 command.
432 When attached via tty mode, detach from the container (and leave it
435 running) using a configurable key sequence. The default sequence is
436 ctrl-p,ctrl-q. Specify the key sequence using the --detach-keys op‐
437 tion, or configure it in the containers.conf file: see contain‐
438 ers.conf(5) for more information.
439 --detach-keys=sequence
442 Specify the key sequence for detaching a container. Format is a single
443 character [a-Z] or one or more ctrl-<value> characters where <value> is
444 one of: a-z, @, ^, [, , or _. Specifying "" disables this feature. The
445 default is ctrl-p,ctrl-q.
446 This option can also be set in containers.conf(5) file.
449 --device=host-device[:container-device][:permissions]
452 Add a host device to the container. Optional permissions parameter can
453 be used to specify device permissions by combining r for read, w for
454 write, and m for mknod(2).
455 Example: --device=/dev/sdc:/dev/xvdc:rwm.
458 Note: if host-device is a symbolic link then it is resolved first. The
461 container only stores the major and minor numbers of the host device.
462 Podman may load kernel modules required for using the specified device.
465 The devices that Podman loads modules for when necessary are:
466 /dev/fuse.
467 In rootless mode, the new device is bind mounted in the container from
470 the host rather than Podman creating it within the container space. Be‐
471 cause the bind mount retains its SELinux label on SELinux systems, the
472 container can get permission denied when accessing the mounted device.
473 Modify SELinux settings to allow containers to use all device labels
474 via the following command:
475 $ sudo setsebool -P container_use_devices=true
478 Note: if the user only has access rights via a group, accessing the de‐
481 vice from inside a rootless container fails. Use the --group-add keep-
482 groups flag to pass the user's supplementary group access into the con‐
483 tainer.
484 --device-cgroup-rule="type major:minor mode"
487 Add a rule to the cgroup allowed devices list. The rule is expected to
488 be in the format specified in the Linux kernel documentation (Documen‐
489 tation/cgroup-v1/devices.txt):
490 - type: a (all), c (char), or b (block);
491 - major and minor: either a number, or * for all;
492 - mode: a composition of r (read), w (write), and m (mknod(2)).
493 --device-read-bps=path:rate
496 Limit read rate (in bytes per second) from a device (e.g. --device-
497 read-bps=/dev/sda:1mb).
498 On some systems, changing the resource limits may not be allowed for
501 non-root users. For more details, see https://github.com/contain‐
502 ers/podman/blob/main/troubleshooting.md#26-running-containers-with-re‐
503 source-limits-fails-with-a-permissions-error
504 This option is not supported on cgroups V1 rootless systems.
507 --device-read-iops=path:rate
510 Limit read rate (in IO operations per second) from a device (e.g. --de‐
511 vice-read-iops=/dev/sda:1000).
512 On some systems, changing the resource limits may not be allowed for
515 non-root users. For more details, see https://github.com/contain‐
516 ers/podman/blob/main/troubleshooting.md#26-running-containers-with-re‐
517 source-limits-fails-with-a-permissions-error
518 This option is not supported on cgroups V1 rootless systems.
521 --device-write-bps=path:rate
524 Limit write rate (in bytes per second) to a device (e.g. --device-
525 write-bps=/dev/sda:1mb).
526 On some systems, changing the resource limits may not be allowed for
529 non-root users. For more details, see https://github.com/contain‐
530 ers/podman/blob/main/troubleshooting.md#26-running-containers-with-re‐
531 source-limits-fails-with-a-permissions-error
532 This option is not supported on cgroups V1 rootless systems.
535 --device-write-iops=path:rate
538 Limit write rate (in IO operations per second) to a device (e.g. --de‐
539 vice-write-iops=/dev/sda:1000).
540 On some systems, changing the resource limits may not be allowed for
543 non-root users. For more details, see https://github.com/contain‐
544 ers/podman/blob/main/troubleshooting.md#26-running-containers-with-re‐
545 source-limits-fails-with-a-permissions-error
546 This option is not supported on cgroups V1 rootless systems.
549 --disable-content-trust
552 This is a Docker-specific option to disable image verification to a
553 container registry and is not supported by Podman. This option is a
554 NOOP and provided solely for scripting compatibility.
555 --dns=ipaddr
558 Set custom DNS servers.
559 This option can be used to override the DNS configuration passed to the
562 container. Typically this is necessary when the host DNS configuration
563 is invalid for the container (e.g., 127.0.0.1). When this is the case
564 the --dns flag is necessary for every run.
565 The special value none can be specified to disable creation of /etc/re‐
568 solv.conf in the container by Podman. The /etc/resolv.conf file in the
569 image is used without changes.
570 This option cannot be combined with --network that is set to none or
573 container:id.
574 --dns-option=option
577 Set custom DNS options. Invalid if using --dns-option with --network
578 that is set to none or container:id.
579 --dns-search=domain
582 Set custom DNS search domains. Invalid if using --dns-search with
583 --network that is set to none or container:id. Use --dns-search=. to
584 remove the search domain.
585 --entrypoint="command" | '["command", arg1 , ...]'
588 Override the default ENTRYPOINT from the image.
589 The ENTRYPOINT of an image is similar to a COMMAND because it specifies
592 what executable to run when the container starts, but it is (purposely)
593 more difficult to override. The ENTRYPOINT gives a container its de‐
594 fault nature or behavior. When the ENTRYPOINT is set, the container
595 runs as if it were that binary, complete with default options. More op‐
596 tions can be passed in via the COMMAND. But, if a user wants to run
597 something else inside the container, the --entrypoint option allows a
598 new ENTRYPOINT to be specified.
599 Specify multi option commands in the form of a json string.
602 --env, -e=env
605 Set environment variables.
606 This option allows arbitrary environment variables that are available
609 for the process to be launched inside of the container. If an environ‐
610 ment variable is specified without a value, Podman checks the host en‐
611 vironment for a value and set the variable only if it is set on the
612 host. As a special case, if an environment variable ending in * is
613 specified without a value, Podman searches the host environment for
614 variables starting with the prefix and adds those variables to the con‐
615 tainer.
616 See ⟨#environment⟩ note below for precedence and examples.
619 --env-file=file
622 Read in a line-delimited file of environment variables.
623 See ⟨#environment⟩ note below for precedence and examples.
626 --env-host
629 Use host environment inside of the container. See Environment note be‐
630 low for precedence. (This option is not available with the remote Pod‐
631 man client, including Mac and Windows (excluding WSL2) machines)
632 --env-merge=env
635 Preprocess default environment variables for the containers. For exam‐
636 ple if image contains environment variable hello=world user can prepro‐
637 cess it using --env-merge hello=${hello}-some so new value is
638 hello=world-some.
639 Please note that if the environment variable hello is not present in
642 the image, then it'll be replaced by an empty string and so using
643 --env-merge hello=${hello}-some would result in the new value of
644 hello=-some, notice the leading - delimiter.
645 --expose=port
648 Expose a port, or a range of ports (e.g. --expose=3300-3310) to set up
649 port redirection on the host system.
650 --gidmap=[flags]container_uid:from_uid[:amount]
653 Run the container in a new user namespace using the supplied GID map‐
654 ping. This option conflicts with the --userns and --subgidname options.
655 This option provides a way to map host GIDs to container GIDs in the
656 same way as --uidmap maps host UIDs to container UIDs. For details see
657 --uidmap.
658 Note: the --gidmap option cannot be called in conjunction with the
661 --pod option as a gidmap cannot be set on the container level when in a
662 pod.
663 --group-add=group | keep-groups
666 Assign additional groups to the primary user running within the con‐
667 tainer process.
668 • keep-groups is a special flag that tells Podman to keep the
671 supplementary group access.
672 Allows container to use the user's supplementary group access. If file
676 systems or devices are only accessible by the rootless user's group,
677 this flag tells the OCI runtime to pass the group access into the con‐
678 tainer. Currently only available with the crun OCI runtime. Note: keep-
679 groups is exclusive, other groups cannot be specified with this flag.
680 (Not available for remote commands, including Mac and Windows (exclud‐
681 ing WSL2) machines)
682 --group-entry=ENTRY
685 Customize the entry that is written to the /etc/group file within the
686 container when --user is used.
687 The variables $GROUPNAME, $GID, and $USERLIST are automatically re‐
690 placed with their value at runtime if present.
691 --health-cmd="command" | '["command", arg1 , ...]'
694 Set or alter a healthcheck command for a container. The command is a
695 command to be executed inside the container that determines the con‐
696 tainer health. The command is required for other healthcheck options to
697 be applied. A value of none disables existing healthchecks.
698 Multiple options can be passed in the form of a JSON array; otherwise,
701 the command is interpreted as an argument to /bin/sh -c.
702 --health-interval=interval
705 Set an interval for the healthchecks. An interval of disable results in
706 no automatic timer setup. The default is 30s.
707 --health-on-failure=action
710 Action to take once the container transitions to an unhealthy state.
711 The default is none.
712 • none: Take no action.
715 • kill: Kill the container.
717 • restart: Restart the container. Do not combine the restart
719 action with the --restart flag. When running inside of a sys‐
720 temd unit, consider using the kill or stop action instead to
721 make use of systemd's restart policy.
722 • stop: Stop the container.
724 --health-retries=retries
728 The number of retries allowed before a healthcheck is considered to be
729 unhealthy. The default value is 3.
730 --health-start-period=period
733 The initialization time needed for a container to bootstrap. The value
734 can be expressed in time format like 2m3s. The default value is 0s.
735 --health-startup-cmd="command" | '["command", arg1 , ...]'
738 Set a startup healthcheck command for a container. This command is exe‐
739 cuted inside the container and is used to gate the regular healthcheck.
740 When the startup command succeeds, the regular healthcheck begins and
741 the startup healthcheck ceases. Optionally, if the command fails for a
742 set number of attempts, the container is restarted. A startup
743 healthcheck can be used to ensure that containers with an extended
744 startup period are not marked as unhealthy until they are fully
745 started. Startup healthchecks can only be used when a regular
746 healthcheck (from the container's image or the --health-cmd option) is
747 also set.
748 --health-startup-interval=interval
751 Set an interval for the startup healthcheck. An interval of disable re‐
752 sults in no automatic timer setup. The default is 30s.
753 --health-startup-retries=retries
756 The number of attempts allowed before the startup healthcheck restarts
757 the container. If set to 0, the container is never restarted. The de‐
758 fault is 0.
759 --health-startup-success=retries
762 The number of successful runs required before the startup healthcheck
763 succeeds and the regular healthcheck begins. A value of 0 means that
764 any success begins the regular healthcheck. The default is 0.
765 --health-startup-timeout=timeout
768 The maximum time a startup healthcheck command has to complete before
769 it is marked as failed. The value can be expressed in a time format
770 like 2m3s. The default value is 30s.
771 --health-timeout=timeout
774 The maximum time allowed to complete the healthcheck before an interval
775 is considered failed. Like start-period, the value can be expressed in
776 a time format such as 1m22s. The default value is 30s.
777 --help
780 Print usage statement
781 --hostname, -h=name
784 Container host name
785 Sets the container host name that is available inside the container.
788 Can only be used with a private UTS namespace --uts=private (default).
789 If --pod is specified and the pod shares the UTS namespace (default)
790 the pod's hostname is used.
791 --hostuser=name
794 Add a user account to /etc/passwd from the host to the container. The
795 Username or UID must exist on the host system.
796 --http-proxy
799 By default proxy environment variables are passed into the container if
800 set for the Podman process. This can be disabled by setting the value
801 to false. The environment variables passed in include http_proxy,
802 https_proxy, ftp_proxy, no_proxy, and also the upper case versions of
803 those. This option is only needed when the host system must use a proxy
804 but the container does not use any proxy. Proxy environment variables
805 specified for the container in any other way overrides the values that
806 have been passed through from the host. (Other ways to specify the
807 proxy for the container include passing the values with the --env flag,
808 or hard coding the proxy environment at container build time.) When
809 used with the remote client it uses the proxy environment variables
810 that are set on the server process.
811 Defaults to true.
814 --image-volume=bind | tmpfs | ignore
817 Tells Podman how to handle the builtin image volumes. Default is bind.
818 • bind: An anonymous named volume is created and mounted into
821 the container.
822 • tmpfs: The volume is mounted onto the container as a tmpfs,
824 which allows the users to create content that disappears when
825 the container is stopped.
826 • ignore: All volumes are just ignored and no action is taken.
828 --init
832 Run an init inside the container that forwards signals and reaps pro‐
833 cesses. The container-init binary is mounted at /run/podman-init.
834 Mounting over /run breaks container execution.
835 --init-path=path
838 Path to the container-init binary.
839 --interactive, -i
842 When set to true, keep stdin open even if not attached. The default is
843 false.
844 --ip=ipv4
847 Specify a static IPv4 address for the container, for example
848 10.88.64.128. This option can only be used if the container is joined
849 to only a single network - i.e., --network=network-name is used at most
850 once - and if the container is not joining another container's network
851 namespace via --network=container:id. The address must be within the
852 network's IP address pool (default 10.88.0.0/16).
853 To specify multiple static IP addresses per container, set multiple
856 networks using the --network option with a static IP address specified
857 for each using the ip mode for that option.
858 --ip6=ipv6
861 Specify a static IPv6 address for the container, for example
862 fd46:db93:aa76:ac37::10. This option can only be used if the container
863 is joined to only a single network - i.e., --network=network-name is
864 used at most once - and if the container is not joining another con‐
865 tainer's network namespace via --network=container:id. The address
866 must be within the network's IPv6 address pool.
867 To specify multiple static IPv6 addresses per container, set multiple
870 networks using the --network option with a static IPv6 address speci‐
871 fied for each using the ip6 mode for that option.
872 --ipc=ipc
875 Set the IPC namespace mode for a container. The default is to create a
876 private IPC namespace.
877 • "": Use Podman's default, defined in containers.conf.
880 • container:id: reuses another container's shared memory, sema‐
882 phores, and message queues
883 • host: use the host's shared memory, semaphores, and message
885 queues inside the container. Note: the host mode gives the
886 container full access to local shared memory and is therefore
887 considered insecure.
888 • none: private IPC namespace, with /dev/shm not mounted.
890 • ns:path: path to an IPC namespace to join.
892 • private: private IPC namespace.
894 • shareable: private IPC namespace with a possibility to share
896 it with other containers.
897 --label, -l=key=value
901 Add metadata to a container.
902 --label-file=file
905 Read in a line-delimited file of labels.
906 --link-local-ip=ip
909 Not implemented.
910 --log-driver=driver
913 Logging driver for the container. Currently available options are k8s-
914 file, journald, none and passthrough, with json-file aliased to k8s-
915 file for scripting compatibility. (Default journald).
916 The podman info command below displays the default log-driver for the
919 system.
920 $ podman info --format '{{ .Host.LogDriver }}'
922 journald
923 The passthrough driver passes down the standard streams (stdin, stdout,
927 stderr) to the container. It is not allowed with the remote Podman
928 client, including Mac and Windows (excluding WSL2) machines, and on a
929 tty, since it is vulnerable to attacks via TIOCSTI.
930 --log-opt=name=value
933 Logging driver specific options.
934 Set custom logging configuration. The following *name*s are supported:
937 path: specify a path to the log file
940 (e.g. --log-opt path=/var/log/container/mycontainer.json);
941 max-size: specify a max size of the log file
944 (e.g. --log-opt max-size=10mb);
945 tag: specify a custom log tag for the container
948 (e.g. --log-opt tag="{{.ImageName}}". It supports the same keys as
949 podman inspect --format. This option is currently supported only by
950 the journald log driver.
951 --mac-address=address
954 Container network interface MAC address (e.g. 92:d0:c6:0a:29:33) This
955 option can only be used if the container is joined to only a single
956 network - i.e., --network=network-name is used at most once - and if
957 the container is not joining another container's network namespace via
958 --network=container:id.
959 Remember that the MAC address in an Ethernet network must be unique.
962 The IPv6 link-local address is based on the device's MAC address ac‐
963 cording to RFC4862.
964 To specify multiple static MAC addresses per container, set multiple
967 networks using the --network option with a static MAC address specified
968 for each using the mac mode for that option.
969 --memory, -m=number[unit]
972 Memory limit. A unit can be b (bytes), k (kibibytes), m (mebibytes), or
973 g (gibibytes).
974 Allows the memory available to a container to be constrained. If the
977 host supports swap memory, then the -m memory setting can be larger
978 than physical RAM. If a limit of 0 is specified (not using -m), the
979 container's memory is not limited. The actual limit may be rounded up
980 to a multiple of the operating system's page size (the value is very
981 large, that's millions of trillions).
982 This option is not supported on cgroups V1 rootless systems.
985 --memory-reservation=number[unit]
988 Memory soft limit. A unit can be b (bytes), k (kibibytes), m
989 (mebibytes), or g (gibibytes).
990 After setting memory reservation, when the system detects memory con‐
993 tention or low memory, containers are forced to restrict their consump‐
994 tion to their reservation. So always set the value below --memory, oth‐
995 erwise the hard limit takes precedence. By default, memory reservation
996 is the same as memory limit.
997 This option is not supported on cgroups V1 rootless systems.
1000 --memory-swap=number[unit]
1003 A limit value equal to memory plus swap. A unit can be b (bytes), k
1004 (kibibytes), m (mebibytes), or g (gibibytes).
1005 Must be used with the -m (--memory) flag. The argument value must be
1008 larger than that of
1009 -m (--memory) By default, it is set to double the value of --memory.
1010 Set number to -1 to enable unlimited swap.
1013 This option is not supported on cgroups V1 rootless systems.
1016 --memory-swappiness=number
1019 Tune a container's memory swappiness behavior. Accepts an integer be‐
1020 tween 0 and 100.
1021 This flag is only supported on cgroups V1 rootful systems.
1024 --mount=type=TYPE,TYPE-SPECIFIC-OPTION[,...]
1027 Attach a filesystem mount to the container
1028 Current supported mount TYPEs are bind, devpts, glob, image, ramfs,
1031 tmpfs and volume. [1] ⟨#Footnote1⟩
1032 e.g.
1034 type=bind,source=/path/on/host,destination=/path/in/container
1035 type=bind,src=/path/on/host,dst=/path/in/container,relabel=shared
1037 type=bind,src=/path/on/host,dst=/path/in/container,relabel=shared,U=true
1039 type=devpts,destination=/dev/pts
1041 type=glob,src=/usr/lib/libfoo*,destination=/usr/lib,ro=true
1043 type=image,source=fedora,destination=/fedora-image,rw=true
1045 type=ramfs,tmpfs-size=512M,destination=/path/in/container
1047 type=tmpfs,tmpfs-size=512M,destination=/path/in/container
1049 type=tmpfs,destination=/path/in/container,noswap
1051 type=volume,source=vol1,destination=/path/in/container,ro=true
1053 Common Options:
1055 · src, source: mount source spec for bind, glob, and volume. Mandatory for bind and glob.
1057 · dst, destination, target: mount destination spec.
1059 When source globs are specified without the destination directory,
1061 the files and directories are mounted with their complete path
1062 within the container. When the destination is specified, the
1063 files and directories matching the glob on the base file name
1064 on the destination directory are mounted. The option
1065 `type=glob,src=/foo*,destination=/tmp/bar` tells container engines
1066 to mount host files matching /foo* to the /tmp/bar/
1067 directory in the container.
1068 Options specific to volume:
1070 · ro, readonly: true or false (default).
1072 . U, chown: true or false (default). Change recursively the owner and group of the source volume based on the UID and GID of the container.
1074 · idmap: true or false (default). If specified, create an idmapped mount to the target user namespace in the container.
1076 The idmap option supports a custom mapping that can be different than the user namespace used by the container.
1077 The mapping can be specified after the idmap option like: `idmap=uids=0-1-10#10-11-10;gids=0-100-10`. For each triplet, the first value is the
1078 start of the backing file system IDs that are mapped to the second value on the host. The length of this mapping is given in the third value.
1079 Multiple ranges are separated with #. If the specified mapping is prepended with a '@' then the mapping is considered relative to the container
1080 user namespace. The host ID for the mapping is changed to account for the relative position of the container user in the container user namespace.
1081 Options specific to image:
1083 · rw, readwrite: true or false (default).
1085 Options specific to bind and glob:
1087 · ro, readonly: true or false (default).
1089 · bind-propagation: shared, slave, private, unbindable, rshared, rslave, runbindable, or rprivate(default). See also mount(2).
1091 . bind-nonrecursive: do not set up a recursive bind mount. By default it is recursive.
1093 . relabel: shared, private.
1095 · idmap: true or false (default). If specified, create an idmapped mount to the target user namespace in the container.
1097 . U, chown: true or false (default). Change recursively the owner and group of the source volume based on the UID and GID of the container.
1099 Options specific to tmpfs and ramfs:
1101 · ro, readonly: true or false (default).
1103 · tmpfs-size: Size of the tmpfs/ramfs mount in bytes. Unlimited by default in Linux.
1105 · tmpfs-mode: File mode of the tmpfs/ramfs in octal. (e.g. 700 or 0700.) Defaults to 1777 in Linux.
1107 · tmpcopyup: Enable copyup from the image directory at the same location to the tmpfs/ramfs. Used by default.
1109 · notmpcopyup: Disable copying files from the image to the tmpfs/ramfs.
1111 . U, chown: true or false (default). Change recursively the owner and group of the source volume based on the UID and GID of the container.
1113 Options specific to devpts:
1115 · uid: UID of the file owner (default 0).
1117 · gid: GID of the file owner (default 0).
1119 · mode: permission mask for the file (default 600).
1121 · max: maximum number of PTYs (default 1048576).
1123 --name=name
1127 Assign a name to the container.
1128 The operator can identify a container in three ways:
1131 • UUID long identifier
1134 (“f78375b1c487e03c9438c729345e54db9d20cfa2ac1fc3494b6eb60872e74778”);
1135 • UUID short identifier (“f78375b1c487”);
1137 • Name (“jonah”).
1139 Podman generates a UUID for each container, and if a name is not as‐
1143 signed to the container with --name then it generates a random string
1144 name. The name can be useful as a more human-friendly way to identify
1145 containers. This works for both background and foreground containers.
1146 --network=mode, --net
1149 Set the network mode for the container.
1150 Valid mode values are:
1153 • bridge[:OPTIONS,...]: Create a network stack on the default
1156 bridge. This is the default for rootful containers. It is pos‐
1157 sible to specify these additional options:
1158 • alias=name: Add network-scoped alias for the container.
1160 • ip=IPv4: Specify a static ipv4 address for this container.
1162 • ip=IPv6: Specify a static ipv6 address for this container.
1164 • mac=MAC: Specify a static mac address for this container.
1166 • interface_name: Specify a name for the created network in‐
1168 terface inside the container.
1169 For example to set a static ipv4 address and a static mac ad‐
1171 dress, use --network bridge:ip=10.88.0.10,mac=44:33:22:11:00:99.
1172 • <network name or ID>[:OPTIONS,...]: Connect to a user-defined
1174 network; this is the network name or ID from a network created
1175 by podman network create. Using the network name implies the
1176 bridge network mode. It is possible to specify the same op‐
1177 tions described under the bridge mode above. Use the --network
1178 option multiple times to specify additional networks. For
1179 backwards compatibility it is also possible to specify net‐
1180 works comma separated on the first --network argument, however
1181 this prevents you from using the options described under the
1182 bridge section above.
1183 • none: Create a network namespace for the container but do not
1185 configure network interfaces for it, thus the container has no
1186 network connectivity.
1187 • container:id: Reuse another container's network stack.
1189 • host: Do not create a network namespace, the container uses
1191 the host's network. Note: The host mode gives the container
1192 full access to local system services such as D-bus and is
1193 therefore considered insecure.
1194 • ns:path: Path to a network namespace to join.
1196 • private: Create a new namespace for the container. This uses
1198 the bridge mode for rootful containers and slirp4netns for
1199 rootless ones.
1200 • slirp4netns[:OPTIONS,...]: use slirp4netns(1) to create a user
1202 network stack. This is the default for rootless containers. It
1203 is possible to specify these additional options, they can also
1204 be set with network_cmd_options in containers.conf:
1205 • allow_host_loopback=true|false: Allow slirp4netns to reach
1207 the host loopback IP (default is 10.0.2.2 or the second IP
1208 from slirp4netns cidr subnet when changed, see the cidr op‐
1209 tion below). The default is false.
1210 • mtu=MTU: Specify the MTU to use for this network. (Default
1212 is 65520).
1213 • cidr=CIDR: Specify ip range to use for this network. (De‐
1215 fault is 10.0.2.0/24).
1216 • enable_ipv6=true|false: Enable IPv6. Default is true. (Re‐
1218 quired for outbound_addr6).
1219 • outbound_addr=INTERFACE: Specify the outbound interface
1221 slirp binds to (ipv4 traffic only).
1222 • outbound_addr=IPv4: Specify the outbound ipv4 address slirp
1224 binds to.
1225 • outbound_addr6=INTERFACE: Specify the outbound interface
1227 slirp binds to (ipv6 traffic only).
1228 • outbound_addr6=IPv6: Specify the outbound ipv6 address slirp
1230 binds to.
1231 • port_handler=rootlesskit: Use rootlesskit for port forward‐
1233 ing. Default. Note: Rootlesskit changes the source IP ad‐
1234 dress of incoming packets to an IP address in the container
1235 network namespace, usually 10.0.2.100. If the application
1236 requires the real source IP address, e.g. web server logs,
1237 use the slirp4netns port handler. The rootlesskit port han‐
1238 dler is also used for rootless containers when connected to
1239 user-defined networks.
1240 • port_handler=slirp4netns: Use the slirp4netns port forward‐
1242 ing, it is slower than rootlesskit but preserves the correct
1243 source IP address. This port handler cannot be used for
1244 user-defined networks.
1245 • pasta[:OPTIONS,...]: use pasta(1) to create a user-mode net‐
1249 working stack.
1250 This is only supported in rootless mode.
1251 By default, IPv4 and IPv6 addresses and routes, as well as the
1252 pod interface name, are copied from the host. If port forward‐
1253 ing isn't configured, ports are forwarded dynamically as ser‐
1254 vices are bound on either side (init namespace or container
1255 namespace). Port forwarding preserves the original source IP
1256 address. Options described in pasta(1) can be specified as
1257 comma-separated arguments.
1258 In terms of pasta(1) options, --config-net is given by de‐
1259 fault, in order to configure networking when the container is
1260 started, and --no-map-gw is also assumed by default, to avoid
1261 direct access from container to host using the gateway ad‐
1262 dress. The latter can be overridden by passing --map-gw in the
1263 pasta-specific options (despite not being an actual pasta(1)
1264 option).
1265 Also, -t none and -u none are passed if, respectively, no TCP
1266 or UDP port forwarding from host to container is configured,
1267 to disable automatic port forwarding based on bound ports.
1268 Similarly, -T none and -U none are given to disable the same
1269 functionality from container to host.
1270 Some examples:
1271 • pasta:--map-gw: Allow the container to directly reach the
1273 host using the gateway address.
1274 • pasta:--mtu,1500: Specify a 1500 bytes MTU for the tap in‐
1276 terface in the container.
1277 • pasta:--ipv4-only,-a,10.0.2.0,-n,24,-g,10.0.2.2,--dns-for‐
1279 ward,10.0.2.3,-m,1500,--no-ndp,--no-dhcpv6,--no-dhcp, equiv‐
1280 alent to default slirp4netns(1) options: disable IPv6, as‐
1281 sign 10.0.2.0/24 to the tap0 interface in the container,
1282 with gateway 10.0.2.3, enable DNS forwarder reachable at
1283 10.0.2.3, set MTU to 1500 bytes, disable NDP, DHCPv6 and
1284 DHCP support.
1285 • pasta:-I,tap0,--ipv4-only,-a,10.0.2.0,-n,24,-g,10.0.2.2,--dns-
1287 forward,10.0.2.3,--no-ndp,--no-dhcpv6,--no-dhcp, equivalent
1288 to default slirp4netns(1) options with Podman overrides:
1289 same as above, but leave the MTU to 65520 bytes
1290 • pasta:-t,auto,-u,auto,-T,auto,-U,auto: enable automatic port
1292 forwarding based on observed bound ports from both host and
1293 container sides
1294 • pasta:-T,5201: enable forwarding of TCP port 5201 from con‐
1296 tainer to host, using the loopback interface instead of the
1297 tap interface for improved performance
1298 NOTE: For backward compatibility reasons, if there is an exist‐
1300 ing network named pasta, Podman uses it instead of the pasta
1301 mode."?
1302 Invalid if using --dns, --dns-option, or --dns-search with --network
1306 set to none or container:id.
1307 If used together with --pod, the container joins the pod's network
1310 namespace.
1311 --network-alias=alias
1314 Add a network-scoped alias for the container, setting the alias for all
1315 networks that the container joins. To set a name only for a specific
1316 network, use the alias option as described under the --network option.
1317 If the network has DNS enabled (podman network inspect -f {{.DNSEn‐
1318 abled}} <name>), these aliases can be used for name resolution on the
1319 given network. This option can be specified multiple times. NOTE: When
1320 using CNI a container only has access to aliases on the first network
1321 that it joins. This limitation does not exist with netavark/aardvark-
1322 dns.
1323 --no-healthcheck
1326 Disable any defined healthchecks for container.
1327 --no-hosts
1330 Do not create /etc/hosts for the container. By default, Podman manages
1331 /etc/hosts, adding the container's own IP address and any hosts from
1332 --add-host. --no-hosts disables this, and the image's /etc/hosts is
1333 preserved unmodified.
1334 This option conflicts with --add-host.
1337 --oom-kill-disable
1340 Whether to disable OOM Killer for the container or not.
1341 This flag is not supported on cgroups V2 systems.
1344 --oom-score-adj=num
1347 Tune the host's OOM preferences for containers (accepts values from
1348 -1000 to 1000).
1349 When running in rootless mode, the specified value can't be lower than
1352 the oom_score_adj for the current process. In this case, the oom-score-
1353 adj is clamped to the current process value.
1354 --os=OS
1357 Override the OS, defaults to hosts, of the image to be pulled. For ex‐
1358 ample, windows. Unless overridden, subsequent lookups of the same im‐
1359 age in the local storage matches this OS, regardless of the host.
1360 --passwd
1363 Allow Podman to add entries to /etc/passwd and /etc/group when used in
1364 conjunction with the --user option. This is used to override the Pod‐
1365 man provided user setup in favor of entrypoint configurations such as
1366 libnss-extrausers.
1367 --passwd-entry=ENTRY
1370 Customize the entry that is written to the /etc/passwd file within the
1371 container when --passwd is used.
1372 The variables $USERNAME, $UID, $GID, $NAME, $HOME are automatically re‐
1375 placed with their value at runtime.
1376 --personality=persona
1379 Personality sets the execution domain via Linux personality(2).
1380 --pid=mode
1383 Set the PID namespace mode for the container. The default is to create
1384 a private PID namespace for the container.
1385 • container:id: join another container's PID namespace;
1388 • host: use the host's PID namespace for the container. Note the
1390 host mode gives the container full access to local PID and is
1391 therefore considered insecure;
1392 • ns:path: join the specified PID namespace;
1394 • private: create a new namespace for the container (default).
1396 --pidfile=path
1400 When the pidfile location is specified, the container process' PID is
1401 written to the pidfile. (This option is not available with the remote
1402 Podman client, including Mac and Windows (excluding WSL2) machines) If
1403 the pidfile option is not specified, the container process' PID is
1404 written to /run/containers/storage/${storage-driver}-contain‐
1405 ers/$CID/userdata/pidfile.
1406 After the container is started, the location for the pidfile can be
1409 discovered with the following podman inspect command:
1410 $ podman inspect --format '{{ .PidFile }}' $CID
1412 /run/containers/storage/${storage-driver}-containers/$CID/userdata/pidfile
1413 --pids-limit=limit
1417 Tune the container's pids limit. Set to -1 to have unlimited pids for
1418 the container. The default is 2048 on systems that support "pids"
1419 cgroup controller.
1420 --platform=OS/ARCH
1423 Specify the platform for selecting the image. (Conflicts with --arch
1424 and --os) The --platform option can be used to override the current ar‐
1425 chitecture and operating system. Unless overridden, subsequent lookups
1426 of the same image in the local storage matches this platform, regard‐
1427 less of the host.
1428 --pod=name
1431 Run container in an existing pod. Podman makes the pod automatically if
1432 the pod name is prefixed with new:. To make a pod with more granular
1433 options, use the podman pod create command before creating a container.
1434 When a container is run with a pod with an infra-container, the infra-
1435 container is started first.
1436 --pod-id-file=file
1439 Run container in an existing pod and read the pod's ID from the speci‐
1440 fied file. When a container is run within a pod which has an infra-
1441 container, the infra-container starts first.
1442 --preserve-fds=N
1445 Pass down to the process N additional file descriptors (in addition to
1446 0, 1, 2). The total FDs are 3+N. (This option is not available with
1447 the remote Podman client, including Mac and Windows (excluding WSL2)
1448 machines)
1449 --privileged
1452 Give extended privileges to this container. The default is false.
1453 By default, Podman containers are unprivileged (=false) and cannot, for
1456 example, modify parts of the operating system. This is because by de‐
1457 fault a container is only allowed limited access to devices. A "privi‐
1458 leged" container is given the same access to devices as the user
1459 launching the container, with the exception of virtual consoles
1460 (/dev/tty\d+) when running in systemd mode (--systemd=always).
1461 A privileged container turns off the security features that isolate the
1464 container from the host. Dropped Capabilities, limited devices, read-
1465 only mount points, Apparmor/SELinux separation, and Seccomp filters are
1466 all disabled. Due to the disabled security features, the privileged
1467 field should almost never be set as containers can easily break out of
1468 confinement.
1469 Containers running in a user namespace (e.g., rootless containers) can‐
1472 not have more privileges than the user that launched them.
1473 --publish, -p=[[ip:][hostPort]:]containerPort[/protocol]
1476 Publish a container's port, or range of ports, to the host.
1477 Both hostPort and containerPort can be specified as a range of ports.
1480 When specifying ranges for both, the number of container ports in the
1481 range must match the number of host ports in the range.
1482 If host IP is set to 0.0.0.0 or not set at all, the port is bound on
1485 all IPs on the host.
1486 By default, Podman publishes TCP ports. To publish a UDP port instead,
1489 give udp as protocol. To publish both TCP and UDP ports, set --publish
1490 twice, with tcp, and udp as protocols respectively. Rootful containers
1491 can also publish ports using the sctp protocol.
1492 Host port does not have to be specified (e.g. podman run -p
1495 127.0.0.1::80). If it is not, the container port is randomly assigned
1496 a port on the host.
1497 Use podman port to see the actual mapping: podman port $CONTAINER $CON‐
1500 TAINERPORT.
1501 Note that the network drivers macvlan and ipvlan do not support port
1504 forwarding, it will have no effect on these networks.
1505 Note: If a container runs within a pod, it is not necessary to publish
1508 the port for the containers in the pod. The port must only be published
1509 by the pod itself. Pod network stacks act like the network stack on the
1510 host - meaning a variety of containers in the pod and programs in the
1511 container all share a single interface, IP address, and associated
1512 ports. If one container binds to a port, no other container can use
1513 that port within the pod while it is in use. Containers in the pod can
1514 also communicate over localhost by having one container bind to local‐
1515 host in the pod, and another connect to that port.
1516 --publish-all, -P
1519 Publish all exposed ports to random ports on the host interfaces. The
1520 default is false.
1521 When set to true, publish all exposed ports to the host interfaces. If
1524 the operator uses -P (or -p) then Podman makes the exposed port acces‐
1525 sible on the host and the ports are available to any client that can
1526 reach the host.
1527 When using this option, Podman binds any exposed port to a random port
1530 on the host within an ephemeral port range defined by
1531 /proc/sys/net/ipv4/ip_local_port_range. To find the mapping between
1532 the host ports and the exposed ports, use podman port.
1533 --pull=policy
1536 Pull image policy. The default is missing.
1537 • always: Always pull the image and throw an error if the pull
1540 fails.
1541 • missing: Pull the image only when the image is not in the lo‐
1543 cal containers storage. Throw an error if no image is found
1544 and the pull fails.
1545 • never: Never pull the image but use the one from the local
1547 containers storage. Throw an error if no image is found.
1548 • newer: Pull if the image on the registry is newer than the one
1550 in the local containers storage. An image is considered to be
1551 newer when the digests are different. Comparing the time
1552 stamps is prone to errors. Pull errors are suppressed if a
1553 local image was found.
1554 --quiet, -q
1558 Suppress output information when pulling images
1559 --rdt-class=intel-rdt-class-of-service
1562 Rdt-class sets the class of service (CLOS or COS) for the container to
1563 run in. Based on the Cache Allocation Technology (CAT) feature that is
1564 part of Intel's Resource Director Technology (RDT) feature set, all
1565 container processes will run within the pre-configured COS, represent‐
1566 ing a part of the cache. The COS has to be created and configured using
1567 a pseudo file system (usually mounted at /sys/fs/resctrl) that the
1568 resctrl kernel driver provides. Assigning the container to a COS re‐
1569 quires root privileges and thus doesn't work in a rootless environment.
1570 Currently, the feature is only supported using runc as a runtime. See
1571 ⟨https://docs.kernel.org/arch/x86/resctrl.html⟩ for more details on
1572 creating a COS before a container can be assigned to it.
1573 --read-only
1576 Mount the container's root filesystem as read-only.
1577 By default, container root filesystems are writable, allowing processes
1580 to write files anywhere. By specifying the --read-only flag, the con‐
1581 tainers root filesystem are mounted read-only prohibiting any writes.
1582 --read-only-tmpfs
1585 When running --read-only containers, mount a read-write tmpfs on /dev,
1586 /dev/shm, /run, /tmp, and /var/tmp. The default is true.
1587 ┌────────────┬───────────────────┬─────┬──────────────────────┐
1590 │--read-only │ --read-only-tmpfs │ / │ /run, /tmp, /var/tmp │
1591 ├────────────┼───────────────────┼─────┼──────────────────────┤
1592 │true │ true │ r/o │ r/w │
1593 ├────────────┼───────────────────┼─────┼──────────────────────┤
1594 │true │ false │ r/o │ r/o │
1595 ├────────────┼───────────────────┼─────┼──────────────────────┤
1596 │false │ false │ r/w │ r/w │
1597 ├────────────┼───────────────────┼─────┼──────────────────────┤
1598 │false │ true │ r/w │ r/w │
1599 └────────────┴───────────────────┴─────┴──────────────────────┘
1600 When --read-only=true and --read-only-tmpfs=true additional tmpfs are
1602 mounted on the /tmp, /run, and /var/tmp directories.
1603 When --read-only=true and --read-only-tmpfs=false /dev and /dev/shm are
1606 marked Read/Only and no tmpfs are mounted on /tmp, /run and /var/tmp.
1607 The directories are exposed from the underlying image, meaning they are
1608 read-only by default. This makes the container totally read-only. No
1609 writable directories exist within the container. In this mode writable
1610 directories need to be added via external volumes or mounts.
1611 By default, when --read-only=false, the /dev and /dev/shm are
1614 read/write, and the /tmp, /run, and /var/tmp are read/write directories
1615 from the container image.
1616 --replace
1619 If another container with the same name already exists, replace and re‐
1620 move it. The default is false.
1621 --requires=container
1624 Specify one or more requirements. A requirement is a dependency con‐
1625 tainer that is started before this container. Containers can be speci‐
1626 fied by name or ID, with multiple containers being separated by commas.
1627 --restart=policy
1630 Restart policy to follow when containers exit. Restart policy does not
1631 take effect if a container is stopped via the podman kill or podman
1632 stop commands.
1633 Valid policy values are:
1636 • no : Do not restart containers on exit
1639 • never : Synonym for no; do not restart con‐
1641 tainers on exit
1642 • on-failure[:max_retries] : Restart containers when they exit
1644 with a non-zero exit code, retrying indefinitely or until the
1645 optional max_retries count is hit
1646 • always : Restart containers when they exit,
1648 regardless of status, retrying indefinitely
1649 • unless-stopped : Identical to always
1651 Podman provides a systemd unit file, podman-restart.service, which
1655 restarts containers after a system reboot.
1656 When running containers in systemd services, use the restart function‐
1659 ality provided by systemd. In other words, do not use this option in a
1660 container unit, instead set the Restart= systemd directive in the [Ser‐
1661 vice] section. See podman-systemd.unit(5) and systemd.service(5).
1662 --rm
1665 Automatically remove the container and any anonymous unnamed volume as‐
1666 sociated with the container when it exits. The default is false.
1667 --rmi
1670 After exit of the container, remove the image unless another container
1671 is using it. Implies --rm on the new container. The default is false.
1672 --rootfs
1675 If specified, the first argument refers to an exploded container on the
1676 file system.
1677 This is useful to run a container without requiring any image manage‐
1680 ment, the rootfs of the container is assumed to be managed externally.
1681 Overlay Rootfs Mounts
1684 The :O flag tells Podman to mount the directory from the rootfs path as
1687 storage using the overlay file system. The container processes can mod‐
1688 ify content within the mount point which is stored in the container
1689 storage in a separate directory. In overlay terms, the source directory
1690 is the lower, and the container storage directory is the upper. Modifi‐
1691 cations to the mount point are destroyed when the container finishes
1692 executing, similar to a tmpfs mount point being unmounted.
1693 Note: On SELinux systems, the rootfs needs the correct label, which is
1696 by default unconfined_u:object_r:container_file_t:s0.
1697 idmap
1700 If idmap is specified, create an idmapped mount to the target user
1703 namespace in the container. The idmap option supports a custom mapping
1704 that can be different than the user namespace used by the container.
1705 The mapping can be specified after the idmap option like:
1706 idmap=uids=0-1-10#10-11-10;gids=0-100-10. For each triplet, the first
1707 value is the start of the backing file system IDs that are mapped to
1708 the second value on the host. The length of this mapping is given in
1709 the third value. Multiple ranges are separated with #.
1710 --sdnotify=container | conmon | healthy | ignore
1713 Determines how to use the NOTIFY_SOCKET, as passed with systemd and
1714 Type=notify.
1715 Default is container, which means allow the OCI runtime to proxy the
1718 socket into the container to receive ready notification. Podman sets
1719 the MAINPID to conmon's pid. The conmon option sets MAINPID to con‐
1720 mon's pid, and sends READY when the container has started. The socket
1721 is never passed to the runtime or the container. The healthy option
1722 sets MAINPID to conmon's pid, and sends READY when the container has
1723 turned healthy; requires a healthcheck to be set. The socket is never
1724 passed to the runtime or the container. The ignore option removes NO‐
1725 TIFY_SOCKET from the environment for itself and child processes, for
1726 the case where some other process above Podman uses NOTIFY_SOCKET and
1727 Podman does not use it.
1728 --seccomp-policy=policy
1731 Specify the policy to select the seccomp profile. If set to image, Pod‐
1732 man looks for a "io.containers.seccomp.profile" label in the container-
1733 image config and use its value as a seccomp profile. Otherwise, Podman
1734 follows the default policy by applying the default profile unless spec‐
1735 ified otherwise via --security-opt seccomp as described below.
1736 Note that this feature is experimental and may change in the future.
1739 --secret=secret[,opt=opt ...]
1742 Give the container access to a secret. Can be specified multiple times.
1743 A secret is a blob of sensitive data which a container needs at runtime
1746 but is not stored in the image or in source control, such as usernames
1747 and passwords, TLS certificates and keys, SSH keys or other important
1748 generic strings or binary content (up to 500 kb in size).
1749 When secrets are specified as type mount, the secrets are copied and
1752 mounted into the container when a container is created. When secrets
1753 are specified as type env, the secret is set as an environment variable
1754 within the container. Secrets are written in the container at the time
1755 of container creation, and modifying the secret using podman secret
1756 commands after the container is created affects the secret inside the
1757 container.
1758 Secrets and its storage are managed using the podman secret command.
1761 Secret Options
1764 • type=mount|env : How the secret is exposed to the con‐
1767 tainer.
1768 mount mounts the secret into the container
1769 as a file.
1770 env exposes the secret as an environment
1771 variable.
1772 Defaults to mount.
1773 • target=target : Target of secret.
1775 For mounted secrets, this is the path to
1776 the secret inside the container.
1777 If a fully qualified path is provided, the
1778 secret is mounted at that location.
1779 Otherwise, the secret is mounted to
1780 /run/secrets/target for linux containers
1781 or
1782 /var/run/secrets/target for freebsd con‐
1783 tainers.
1784 If the target is not set, the secret is
1785 mounted to /run/secrets/secretname by default.
1786 For env secrets, this is the environment
1787 variable key. Defaults to secretname.
1788 • uid=0 : UID of secret. Defaults to 0. Mount secret
1790 type only.
1791 • gid=0 : GID of secret. Defaults to 0. Mount secret
1793 type only.
1794 • mode=0 : Mode of secret. Defaults to 0444. Mount
1796 secret type only.
1797 Examples
1801 Mount at /my/location/mysecret with UID 1:
1804 --secret mysecret,target=/my/location/mysecret,uid=1
1806 Mount at /run/secrets/customtarget with mode 0777:
1810 --secret mysecret,target=customtarget,mode=0777
1812 Create a secret environment variable called ENVSEC:
1816 --secret mysecret,type=env,target=ENVSEC
1818 --security-opt=option
1822 Security Options
1823 • apparmor=unconfined : Turn off apparmor confinement for the
1826 container
1827 • apparmor=alternate-profile : Set the apparmor confinement pro‐
1829 file for the container
1830 • label=user:USER: Set the label user for the container pro‐
1832 cesses
1833 • label=role:ROLE: Set the label role for the container pro‐
1835 cesses
1836 • label=type:TYPE: Set the label process type for the container
1838 processes
1839 • label=level:LEVEL: Set the label level for the container pro‐
1841 cesses
1842 • label=filetype:TYPE: Set the label file type for the container
1844 files
1845 • label=disable: Turn off label separation for the container
1847 Note: Labeling can be disabled for all containers by setting la‐
1851 bel=false in the containers.conf (/etc/containers/containers.conf or
1852 $HOME/.config/containers/containers.conf) file.
1853 • label=nested: Allows SELinux modifications within the con‐
1856 tainer. Containers are allowed to modify SELinux labels on
1857 files and processes, as long as SELinux policy allows. Without
1858 nested, containers view SELinux as disabled, even when it is
1859 enabled on the host. Containers are prevented from setting any
1860 labels.
1861 • mask=/path/1:/path/2: The paths to mask separated by a colon.
1863 A masked path cannot be accessed inside the container.
1864 • no-new-privileges: Disable container processes from gaining
1866 additional privileges.
1867 • seccomp=unconfined: Turn off seccomp confinement for the con‐
1869 tainer.
1870 • seccomp=profile.json: JSON file to be used as a seccomp fil‐
1872 ter. Note that the io.podman.annotations.seccomp annotation is
1873 set with the specified value as shown in podman inspect.
1874 • proc-opts=OPTIONS : Comma-separated list of options to use for
1876 the /proc mount. More details for the possible mount options
1877 are specified in the proc(5) man page.
1878 • unmask=ALL or /path/1:/path/2, or shell expanded paths
1880 (/proc/*): Paths to unmask separated by a colon. If set to
1881 ALL, it unmasks all the paths that are masked or made read-
1882 only by default. The default masked paths are /proc/acpi,
1883 /proc/kcore, /proc/keys, /proc/latency_stats, /proc/sched_de‐
1884 bug, /proc/scsi, /proc/timer_list, /proc/timer_stats,
1885 /sys/firmware, and /sys/fs/selinux, /sys/devices/virtual/pow‐
1886 ercap. The default paths that are read-only are /proc/asound,
1887 /proc/bus, /proc/fs, /proc/irq, /proc/sys, /proc/sysrq-trig‐
1888 ger, /sys/fs/cgroup.
1889 Note: Labeling can be disabled for all containers by setting la‐
1893 bel=false in the containers.conf(5) file.
1894 --shm-size=number[unit]
1897 Size of /dev/shm. A unit can be b (bytes), k (kibibytes), m
1898 (mebibytes), or g (gibibytes). If the unit is omitted, the system uses
1899 bytes. If the size is omitted, the default is 64m. When size is 0,
1900 there is no limit on the amount of memory used for IPC by the con‐
1901 tainer. This option conflicts with --ipc=host.
1902 --shm-size-systemd=number[unit]
1905 Size of systemd-specific tmpfs mounts such as /run, /run/lock,
1906 /var/log/journal and /tmp. A unit can be b (bytes), k (kibibytes), m
1907 (mebibytes), or g (gibibytes). If the unit is omitted, the system uses
1908 bytes. If the size is omitted, the default is 64m. When size is 0, the
1909 usage is limited to 50% of the host's available memory.
1910 --sig-proxy
1913 Proxy received signals to the container process. SIGCHLD, SIGURG,
1914 SIGSTOP, and SIGKILL are not proxied.
1915 The default is true.
1918 --stop-signal=signal
1921 Signal to stop a container. Default is SIGTERM.
1922 --stop-timeout=seconds
1925 Timeout to stop a container. Default is 10. Remote connections use lo‐
1926 cal containers.conf for defaults.
1927 --subgidname=name
1930 Run the container in a new user namespace using the map with name in
1931 the /etc/subgid file. If running rootless, the user needs to have the
1932 right to use the mapping. See subgid(5). This flag conflicts with
1933 --userns and --gidmap.
1934 --subuidname=name
1937 Run the container in a new user namespace using the map with name in
1938 the /etc/subuid file. If running rootless, the user needs to have the
1939 right to use the mapping. See subuid(5). This flag conflicts with
1940 --userns and --uidmap.
1941 --sysctl=name=value
1944 Configure namespaced kernel parameters at runtime.
1945 For the IPC namespace, the following sysctls are allowed:
1948 • kernel.msgmax
1951 • kernel.msgmnb
1953 • kernel.msgmni
1955 • kernel.sem
1957 • kernel.shmall
1959 • kernel.shmmax
1961 • kernel.shmmni
1963 • kernel.shm_rmid_forced
1965 • Sysctls beginning with fs.mqueue.*
1967 Note: if using the --ipc=host option, the above sysctls are not al‐
1971 lowed.
1972 For the network namespace, only sysctls beginning with net.* are al‐
1975 lowed.
1976 Note: if using the --network=host option, the above sysctls are not al‐
1979 lowed.
1980 --systemd=true | false | always
1983 Run container in systemd mode. The default is true.
1984 • true enables systemd mode only when the command executed in‐
1987 side the container is systemd, /usr/sbin/init, /sbin/init or
1988 /usr/local/sbin/init.
1989 • false disables systemd mode.
1991 • always enforces the systemd mode to be enabled.
1993 Running the container in systemd mode causes the following changes:
1997 • Podman mounts tmpfs file systems on the following directories
2000 • /run
2002 • /run/lock
2004 • /tmp
2006 • /sys/fs/cgroup/systemd (on a cgroup v1 system)
2008 • /var/lib/journal
2010 • Podman sets the default stop signal to SIGRTMIN+3.
2014 • Podman sets container_uuid environment variable in the con‐
2016 tainer to the first 32 characters of the container ID.
2017 • Podman does not mount virtual consoles (/dev/tty\d+) when run‐
2019 ning with --privileged.
2020 • On cgroup v2, /sys/fs/cgroup is mounted writeable.
2022 This allows systemd to run in a confined container without any modifi‐
2026 cations.
2027 Note that on SELinux systems, systemd attempts to write to the cgroup
2030 file system. Containers writing to the cgroup file system are denied by
2031 default. The container_manage_cgroup boolean must be enabled for this
2032 to be allowed on an SELinux separated system.
2033 setsebool -P container_manage_cgroup true
2035 --timeout=seconds
2039 Maximum time a container is allowed to run before conmon sends it the
2040 kill signal. By default containers run until they exit or are stopped
2041 by podman stop.
2042 --tls-verify
2045 Require HTTPS and verify certificates when contacting registries (de‐
2046 fault: true). If explicitly set to true, TLS verification is used. If
2047 set to false, TLS verification is not used. If not specified, TLS ver‐
2048 ification is used unless the target registry is listed as an insecure
2049 registry in containers-registries.conf(5)
2050 --tmpfs=fs
2053 Create a tmpfs mount.
2054 Mount a temporary filesystem (tmpfs) mount into a container, for exam‐
2057 ple:
2058 $ podman run -d --tmpfs /tmp:rw,size=787448k,mode=1777 my_image
2060 This command mounts a tmpfs at /tmp within the container. The supported
2064 mount options are the same as the Linux default mount flags. If no op‐
2065 tions are specified, the system uses the following options:
2066 rw,noexec,nosuid,nodev.
2067 --tty, -t
2070 Allocate a pseudo-TTY. The default is false.
2071 When set to true, Podman allocates a pseudo-tty and attach to the stan‐
2074 dard input of the container. This can be used, for example, to run a
2075 throwaway interactive shell.
2076 NOTE: The --tty flag prevents redirection of standard output. It com‐
2079 bines STDOUT and STDERR, it can insert control characters, and it can
2080 hang pipes. This option is only used when run interactively in a termi‐
2081 nal. When feeding input to Podman, use -i only, not -it.
2082 echo "asdf" | podman run --rm -i someimage /bin/cat
2084 --tz=timezone
2088 Set timezone in container. This flag takes area-based timezones, GMT
2089 time, as well as local, which sets the timezone in the container to
2090 match the host machine. See /usr/share/zoneinfo/ for valid timezones.
2091 Remote connections use local containers.conf for defaults
2092 --uidmap=[flags]container_uid:from_uid[:amount]
2095 Run the container in a new user namespace using the supplied UID map‐
2096 ping. This option conflicts with the --userns and --subuidname options.
2097 This option provides a way to map host UIDs to container UIDs. It can
2098 be passed several times to map different ranges.
2099 The possible values of the optional flags are discussed further down on
2102 this page. The amount value is optional and assumed to be 1 if not
2103 given.
2104 The from_uid value is based upon the user running the command, either
2107 rootful or rootless users.
2108 • rootful user: [flags]container_uid:host_uid[:amount]
2111 • rootless user: [flags]container_uid:intermediate_uid[:amount]
2113 Rootful mappings
2117 When podman run is called by a privileged user, the option --uidmap
2120 works as a direct mapping between host UIDs and container UIDs.
2121 host UID -> container UID
2124 The amount specifies the number of consecutive UIDs that is mapped. If
2127 for example amount is 4 the mapping looks like:
2128 ┌─────────────┬───────────────────┐
2131 │host UID │ container UID │
2132 ├─────────────┼───────────────────┤
2133 │from_uid │ container_uid │
2134 ├─────────────┼───────────────────┤
2135 │from_uid + 1 │ container_uid + 1 │
2136 ├─────────────┼───────────────────┤
2137 │from_uid + 2 │ container_uid + 2 │
2138 ├─────────────┼───────────────────┤
2139 │from_uid + 3 │ container_uid + 3 │
2140 └─────────────┴───────────────────┘
2141 Rootless mappings
2143 When podman run is called by an unprivileged user (i.e. running root‐
2146 less), the value from_uid is interpreted as an "intermediate UID". In
2147 the rootless case, host UIDs are not mapped directly to container UIDs.
2148 Instead the mapping happens over two mapping steps:
2149 host UID -> intermediate UID -> container UID
2152 The --uidmap option only influences the second mapping step.
2155 The first mapping step is derived by Podman from the contents of the
2158 file /etc/subuid and the UID of the user calling Podman.
2159 First mapping step:
2162 ┌────────────────────┬──────────────────┐
2165 │host UID │ intermediate UID │
2166 ├────────────────────┼──────────────────┤
2167 │UID for Podman user │ 0 │
2168 ├────────────────────┼──────────────────┤
2169 │1st subordinate UID │ 1 │
2170 ├────────────────────┼──────────────────┤
2171 │2nd subordinate UID │ 2 │
2172 ├────────────────────┼──────────────────┤
2173 │3rd subordinate UID │ 3 │
2174 ├────────────────────┼──────────────────┤
2175 │nth subordinate UID │ n │
2176 └────────────────────┴──────────────────┘
2177 To be able to use intermediate UIDs greater than zero, the user needs
2179 to have subordinate UIDs configured in /etc/subuid. See subuid(5).
2180 The second mapping step is configured with --uidmap.
2183 If for example amount is 5 the second mapping step looks like:
2186 ┌─────────────────┬───────────────────┐
2189 │intermediate UID │ container UID │
2190 ├─────────────────┼───────────────────┤
2191 │from_uid │ container_uid │
2192 ├─────────────────┼───────────────────┤
2193 │from_uid + 1 │ container_uid + 1 │
2194 ├─────────────────┼───────────────────┤
2195 │from_uid + 2 │ container_uid + 2 │
2196 ├─────────────────┼───────────────────┤
2197 │from_uid + 3 │ container_uid + 3 │
2198 ├─────────────────┼───────────────────┤
2199 │from_uid + 4 │ container_uid + 4 │
2200 └─────────────────┴───────────────────┘
2201 When running as rootless, Podman uses all the ranges configured in the
2203 /etc/subuid file.
2204 The current user ID is mapped to UID=0 in the rootless user namespace.
2207 Every additional range is added sequentially afterward:
2208 ┌──────────────────────┬─────────────────────────┬──────────────────────┐
2211 │host │ rootless user namespace │ length │
2212 ├──────────────────────┼─────────────────────────┼──────────────────────┤
2213 │$UID │ 0 │ 1 │
2214 ├──────────────────────┼─────────────────────────┼──────────────────────┤
2215 │1 │ $FIRST_RANGE_ID │ $FIRST_RANGE_LENGTH │
2216 ├──────────────────────┼─────────────────────────┼──────────────────────┤
2217 │1+$FIRST_RANGE_LENGTH │ $SECOND_RANGE_ID │ $SECOND_RANGE_LENGTH │
2218 └──────────────────────┴─────────────────────────┴──────────────────────┘
2219 Referencing a host ID from the parent namespace
2221 As a rootless user, the given host ID in --uidmap or --gidmap is mapped
2224 from the intermediate namespace generated by Podman. Sometimes it is
2225 desirable to refer directly at the host namespace. It is possible to
2226 manually do so, by running podman unshare cat /proc/self/gid_map, find‐
2227 ing the desired host id at the second column of the output, and getting
2228 the corresponding intermediate id from the first column.
2229 Podman can perform all that by preceding the host id in the mapping
2232 with the @ symbol. For instance, by specifying --gidmap 100000:@2000:1,
2233 podman will look up the intermediate id corresponding to host id 2000
2234 and it will map the found intermediate id to the container id 100000.
2235 The given host id must have been subordinated (otherwise it would not
2236 be mapped into the intermediate space in the first place).
2237 If the length is greater than one, for instance with --gidmap
2240 100000:@2000:2, Podman will map host ids 2000 and 2001 to 100000 and
2241 100001, respectively, regardless of how the intermediate mapping is de‐
2242 fined.
2243 Extending previous mappings
2246 Some mapping modifications may be cumbersome. For instance, a user
2249 starts with a mapping such as --gidmap="0:0:65000", that needs to be
2250 changed such as the parent id 1000 is mapped to container id 100000 in‐
2251 stead, leaving container id 1 unassigned. The corresponding --gidmap
2252 becomes --gidmap="0:0:1" --gidmap="2:2:65534" --gidmap="100000:1:1".
2253 This notation can be simplified using the + flag, that takes care of
2256 breaking previous mappings removing any conflicting assignment with the
2257 given mapping. The flag is given before the container id as follows:
2258 --gidmap="0:0:65000" --gidmap="+100000:1:1"
2259 ┌─────┬─────────────┬─────────────────────────────┐
2262 │Flag │ Example │ Description │
2263 ├─────┼─────────────┼─────────────────────────────┤
2264 │+ │ +100000:1:1 │ Extend the previous mapping │
2265 └─────┴─────────────┴─────────────────────────────┘
2266 This notation leads to gaps in the assignment, so it may be convenient
2268 to fill those gaps afterwards: --gidmap="0:0:65000"
2269 --gidmap="+100000:1:1" --gidmap="1:65001:1"
2270 One specific use case for this flag is in the context of rootless
2273 users. A rootless user may specify mappings with the + flag as in
2274 --gidmap="+100000:1:1". Podman will then "fill the gaps" starting from
2275 zero with all the remaining intermediate ids. This is convenient when a
2276 user wants to map a specific intermediate id to a container id, leaving
2277 the rest of subordinate ids to be mapped by Podman at will.
2278 Passing only one of --uidmap or --gidmap
2281 Usually, subordinated user and group ids are assigned simultaneously,
2284 and for any user the subordinated user ids match the subordinated group
2285 ids. For convenience, if only one of --uidmap or --gidmap is given,
2286 podman assumes the mapping refers to both UIDs and GIDs and applies the
2287 given mapping to both. If only one value of the two needs to be
2288 changed, the mappings should include the u or the g flags to specify
2289 that they only apply to UIDs or GIDs and should not be copied over.
2290 ┌─────┬───────────────┬─────────────────────┐
2293 │flag │ Example │ Description │
2294 ├─────┼───────────────┼─────────────────────┤
2295 │u │ u20000:2000:1 │ The mapping only │
2296 │ │ │ applies to UIDs │
2297 ├─────┼───────────────┼─────────────────────┤
2298 │g │ g10000:1000:1 │ The mapping only │
2299 │ │ │ applies to GIDs │
2300 └─────┴───────────────┴─────────────────────┘
2301 For instance given the command
2303 podman run --gidmap "0:0:1000" --gidmap "g2000:2000:1"
2305 Since no --uidmap is given, the --gidmap is copied to --uidmap, giving
2309 a command equivalent to
2310 podman run --gidmap "0:0:1000" --gidmap "2000:2000:1" --uidmap "0:0:1000"
2312 The --gidmap "g2000:2000:1" used the g flag and therefore it was not
2316 copied to --uidmap.
2317 Rootless mapping of additional host GIDs
2320 A rootless user may desire to map a specific host group that has al‐
2323 ready been subordinated within /etc/subgid without specifying the rest
2324 of the mapping.
2325 This can be done with --gidmap "+gcontainer_gid:@host_gid"
2328 Where:
2331 • The host GID is given through the @ symbol
2334 • The mapping of this GID is not copied over to --usermap thanks
2336 to the g flag.
2337 • The rest of the container IDs will be mapped starting from 0
2339 to n, with all the remaining subordinated GIDs, thanks to the
2340 + flag.
2341 For instance, if a user belongs to the group 2000 and that group is
2345 subordinated to that user (with usermod --add-subgids 2000-2000 $USER),
2346 the user can map the group into the container with:
2347 --gidmap=+g100000:@2000.
2348 If this mapping is combined with the option, --group-add=keep-groups,
2351 the process in the container will belong to group 100000, and files be‐
2352 longing to group 2000 in the host will appear as being owned by group
2353 100000 inside the container.
2354 podman run --group-add=keep-groups --gidmap="+g100000:@2000" ...
2356 No subordinate UIDs
2360 Even if a user does not have any subordinate UIDs in /etc/subuid,
2363 --uidmap can be used to map the normal UID of the user to a container
2364 UID by running podman run --uidmap $container_uid:0:1 --user $con‐
2365 tainer_uid ....
2366 Pods
2369 The --uidmap option cannot be called in conjunction with the --pod op‐
2372 tion as a uidmap cannot be set on the container level when in a pod.
2373 --ulimit=option
2376 Ulimit options. Sets the ulimits values inside of the container.
2377 --ulimit with a soft and hard limit in the format =[:]. For example:
2380 $ podman run --ulimit nofile=1024:1024 --rm ubi9 ulimit -n 1024
2383 Set -1 for the soft or hard limit to set the limit to the maximum limit
2386 of the current process. In rootful mode this is often unlimited.
2387 Use host to copy the current configuration from the host.
2390 Don't use nproc with the ulimit flag as Linux uses nproc to set the
2393 maximum number of processes available to a user, not to a container.
2394 Use the --pids-limit option to modify the cgroup control to limit the
2397 number of processes within a container.
2398 --umask=umask
2401 Set the umask inside the container. Defaults to 0022. Remote connec‐
2402 tions use local containers.conf for defaults
2403 --unsetenv=env
2406 Unset default environment variables for the container. Default environ‐
2407 ment variables include variables provided natively by Podman, environ‐
2408 ment variables configured by the image, and environment variables from
2409 containers.conf.
2410 --unsetenv-all
2413 Unset all default environment variables for the container. Default en‐
2414 vironment variables include variables provided natively by Podman, en‐
2415 vironment variables configured by the image, and environment variables
2416 from containers.conf.
2417 --user, -u=user[:group]
2420 Sets the username or UID used and, optionally, the groupname or GID for
2421 the specified command. Both user and group may be symbolic or numeric.
2422 Without this argument, the command runs as the user specified in the
2425 container image. Unless overridden by a USER command in the Container‐
2426 file or by a value passed to this option, this user generally defaults
2427 to root.
2428 When a user namespace is not in use, the UID and GID used within the
2431 container and on the host match. When user namespaces are in use, how‐
2432 ever, the UID and GID in the container may correspond to another UID
2433 and GID on the host. In rootless containers, for example, a user name‐
2434 space is always used, and root in the container by default corresponds
2435 to the UID and GID of the user invoking Podman.
2436 --userns=mode
2439 Set the user namespace mode for the container.
2440 If --userns is not set, the default value is determined as follows. -
2443 If --pod is set, --userns is ignored and the user namespace of the pod
2444 is used. - If the environment variable PODMAN_USERNS is set its value
2445 is used. - If userns is specified in containers.conf this value is
2446 used. - Otherwise, --userns=host is assumed.
2447 --userns="" (i.e., an empty string) is an alias for --userns=host.
2450 This option is incompatible with --gidmap, --uidmap, --subuidname and
2453 --subgidname.
2454 Rootless user --userns=Key mappings:
2457 ┌────────────────────────┬───────────┬─────────────────────┐
2460 │Key │ Host User │ Container User │
2461 ├────────────────────────┼───────────┼─────────────────────┤
2462 │auto │ $UID │ nil (Host User UID │
2463 │ │ │ is not mapped into │
2464 │ │ │ container.) │
2465 ├────────────────────────┼───────────┼─────────────────────┤
2466 │host │ $UID │ 0 (Default User ac‐ │
2467 │ │ │ count mapped to │
2468 │ │ │ root user in con‐ │
2469 │ │ │ tainer.) │
2470 ├────────────────────────┼───────────┼─────────────────────┤
2471 │keep-id │ $UID │ $UID (Map user ac‐ │
2472 │ │ │ count to same UID │
2473 │ │ │ within container.) │
2474 ├────────────────────────┼───────────┼─────────────────────┤
2475 │keep-id:uid=200,gid=210 │ $UID │ 200:210 (Map user │
2476 │ │ │ account to speci‐ │
2477 │ │ │ fied UID, GID value │
2478 │ │ │ within container.) │
2479 ├────────────────────────┼───────────┼─────────────────────┤
2480 │nomap │ $UID │ nil (Host User UID │
2481 │ │ │ is not mapped into │
2482 │ │ │ container.) │
2483 └────────────────────────┴───────────┴─────────────────────┘
2484 Valid mode values are:
2486 auto[:OPTIONS,...]: automatically create a unique user namespace.
2489 • rootful mode: The --userns=auto flag requires that the user
2492 name containers be specified in the /etc/subuid and /etc/sub‐
2493 gid files, with an unused range of subordinate user IDs that
2494 Podman containers are allowed to allocate.
2495 Example: containers:2147483647:2147483648.
2499 • rootless mode: The users range from the /etc/subuid and
2502 /etc/subgid files will be used. Note running a single con‐
2503 tainer without using --userns=auto will use the entire range
2504 of UIDs and not allow further subdividing. See subuid(5).
2505 Podman allocates unique ranges of UIDs and GIDs from the containers
2509 subordinate user IDs. The size of the ranges is based on the number of
2510 UIDs required in the image. The number of UIDs and GIDs can be overrid‐
2511 den with the size option.
2512 The option --userns=keep-id uses all the subuids and subgids of the
2515 user. The option --userns=nomap uses all the subuids and subgids of
2516 the user except the user's own ID. Using --userns=auto when starting
2517 new containers does not work as long as any containers exist that were
2518 started with --userns=keep-id or --userns=nomap.
2519 Valid auto options:
2522 • gidmapping=CONTAINER_GID:HOST_GID:SIZE: to force a GID mapping
2525 to be present in the user namespace.
2526 • size=SIZE: to specify an explicit size for the automatic user
2528 namespace. e.g. --userns=auto:size=8192. If size is not speci‐
2529 fied, auto estimates a size for the user namespace.
2530 • uidmapping=CONTAINER_UID:HOST_UID:SIZE: to force a UID mapping
2532 to be present in the user namespace.
2533 container:id: join the user namespace of the specified container.
2537 host or "" (empty string): run in the user namespace of the caller. The
2540 processes running in the container have the same privileges on the host
2541 as any other process launched by the calling user.
2542 keep-id: creates a user namespace where the current user's UID:GID are
2545 mapped to the same values in the container. For containers created by
2546 root, the current mapping is created into a new user namespace.
2547 Valid keep-id options:
2550 • uid=UID: override the UID inside the container that is used to
2553 map the current user to.
2554 • gid=GID: override the GID inside the container that is used to
2556 map the current user to.
2557 nomap: creates a user namespace where the current rootless user's
2561 UID:GID are not mapped into the container. This option is not allowed
2562 for containers created by the root user.
2563 ns:namespace: run the container in the given existing user namespace.
2566 --uts=mode
2569 Set the UTS namespace mode for the container. The following values are
2570 supported:
2571 • host: use the host's UTS namespace inside the container.
2574 • private: create a new namespace for the container (default).
2576 • ns:[path]: run the container in the given existing UTS name‐
2578 space.
2579 • container:[container]: join the UTS namespace of the specified
2581 container.
2582 --variant=VARIANT
2586 Use VARIANT instead of the default architecture variant of the con‐
2587 tainer image. Some images can use multiple variants of the arm archi‐
2588 tectures, such as arm/v5 and arm/v7.
2589 --volume, -v=[[SOURCE-VOLUME|HOST-DIR:]CONTAINER-DIR[:OPTIONS]]
2592 Create a bind mount. If -v /HOST-DIR:/CONTAINER-DIR is specified, Pod‐
2593 man bind mounts /HOST-DIR from the host into /CONTAINER-DIR in the Pod‐
2594 man container. Similarly, -v SOURCE-VOLUME:/CONTAINER-DIR mounts the
2595 named volume from the host into the container. If no such named volume
2596 exists, Podman creates one. If no source is given, the volume is cre‐
2597 ated as an anonymously named volume with a randomly generated name, and
2598 is removed when the container is removed via the --rm flag or the pod‐
2599 man rm --volumes command.
2600 (Note when using the remote client, including Mac and Windows (exclud‐
2603 ing WSL2) machines, the volumes are mounted from the remote server, not
2604 necessarily the client machine.)
2605 The OPTIONS is a comma-separated list and can be: [1] ⟨#Footnote1⟩
2608 • rw|ro
2611 • z|Z
2613 • [O]
2615 • [U]
2617 • [no]copy
2619 • [no]dev
2621 • [no]exec
2623 • [no]suid
2625 • [r]bind
2627 • [r]shared|[r]slave|[r]private[r]unbindable
2629 • idmap[=options]
2631 The CONTAINER-DIR must be an absolute path such as /src/docs. The vol‐
2635 ume is mounted into the container at this directory.
2636 If a volume source is specified, it must be a path on the host or the
2639 name of a named volume. Host paths are allowed to be absolute or rela‐
2640 tive; relative paths are resolved relative to the directory Podman is
2641 run in. If the source does not exist, Podman returns an error. Users
2642 must pre-create the source files or directories.
2643 Any source that does not begin with a . or / is treated as the name of
2646 a named volume. If a volume with that name does not exist, it is cre‐
2647 ated. Volumes created with names are not anonymous, and they are not
2648 removed by the --rm option and the podman rm --volumes command.
2649 Specify multiple -v options to mount one or more volumes into a con‐
2652 tainer.
2653 Write Protected Volume Mounts
2656 Add :ro or :rw option to mount a volume in read-only or read-write
2659 mode, respectively. By default, the volumes are mounted read-write.
2660 See examples.
2661 Chowning Volume Mounts
2664 By default, Podman does not change the owner and group of source volume
2667 directories mounted into containers. If a container is created in a new
2668 user namespace, the UID and GID in the container may correspond to an‐
2669 other UID and GID on the host.
2670 The :U suffix tells Podman to use the correct host UID and GID based on
2673 the UID and GID within the container, to change recursively the owner
2674 and group of the source volume. Chowning walks the file system under
2675 the volume and changes the UID/GID on each file. If the volume has
2676 thousands of inodes, this process takes a long time, delaying the start
2677 of the container.
2678 Warning use with caution since this modifies the host filesystem.
2681 Labeling Volume Mounts
2684 Labeling systems like SELinux require that proper labels are placed on
2687 volume content mounted into a container. Without a label, the security
2688 system might prevent the processes running inside the container from
2689 using the content. By default, Podman does not change the labels set by
2690 the OS.
2691 To change a label in the container context, add either of two suffixes
2694 :z or :Z to the volume mount. These suffixes tell Podman to relabel
2695 file objects on the shared volumes. The z option tells Podman that two
2696 or more containers share the volume content. As a result, Podman labels
2697 the content with a shared content label. Shared volume labels allow all
2698 containers to read/write content. The Z option tells Podman to label
2699 the content with a private unshared label Only the current container
2700 can use a private volume. Relabeling walks the file system under the
2701 volume and changes the label on each file, if the volume has thousands
2702 of inodes, this process takes a long time, delaying the start of the
2703 container. If the volume was previously relabeled with the z option,
2704 Podman is optimized to not relabel a second time. If files are moved
2705 into the volume, then the labels can be manually change with the chcon
2706 -Rt container_file_t PATH command.
2707 Note: Do not relabel system files and directories. Relabeling system
2710 content might cause other confined services on the machine to fail.
2711 For these types of containers we recommend disabling SELinux separa‐
2712 tion. The option --security-opt label=disable disables SELinux separa‐
2713 tion for the container. For example if a user wanted to volume mount
2714 their entire home directory into a container, they need to disable
2715 SELinux separation.
2716 $ podman run --security-opt label=disable -v $HOME:/home/user fedora touch /home/user/file
2718 Overlay Volume Mounts
2722 The :O flag tells Podman to mount the directory from the host as a tem‐
2725 porary storage using the overlay file system. The container processes
2726 can modify content within the mountpoint which is stored in the con‐
2727 tainer storage in a separate directory. In overlay terms, the source
2728 directory is the lower, and the container storage directory is the up‐
2729 per. Modifications to the mount point are destroyed when the container
2730 finishes executing, similar to a tmpfs mount point being unmounted.
2731 For advanced users, the overlay option also supports custom non-
2734 volatile upperdir and workdir for the overlay mount. Custom upperdir
2735 and workdir can be fully managed by the users themselves, and Podman
2736 does not remove it on lifecycle completion. Example :O,up‐
2737 perdir=/some/upper,workdir=/some/work
2738 Subsequent executions of the container sees the original source direc‐
2741 tory content, any changes from previous container executions no longer
2742 exist.
2743 One use case of the overlay mount is sharing the package cache from the
2746 host into the container to allow speeding up builds.
2747 Note: The O flag conflicts with other options listed above.
2750 Content mounted into the container is labeled with the private label.
2753 On SELinux systems, labels in the source directory must be read‐
2754 able by the container label. Usually containers can read/execute con‐
2755 tainer_share_t and can read/write container_file_t. If unable to change
2756 the labels on a source volume, SELinux container separation must be
2757 disabled for the container to work.
2758 - Do not modify the source directory mounted into the container
2759 with an overlay mount, it can cause unexpected failures. Only modify
2760 the directory after the container finishes running.
2761 Mounts propagation
2764 By default bind mounted volumes are private. That means any mounts done
2767 inside the container is not visible on host and vice versa. One can
2768 change this behavior by specifying a volume mount propagation property.
2769 Making a volume shared mounts done under that volume inside the con‐
2770 tainer is visible on host and vice versa. Making a volume slave enables
2771 only one way mount propagation and that is mounts done on host under
2772 that volume is visible inside container but not the other way around.
2773 [1] ⟨#Footnote1⟩
2774 To control mount propagation property of a volume one can use the
2777 [r]shared, [r]slave, [r]private or the [r]unbindable propagation flag.
2778 Propagation property can be specified only for bind mounted volumes and
2779 not for internal volumes or named volumes. For mount propagation to
2780 work the source mount point (the mount point where source dir is
2781 mounted on) has to have the right propagation properties. For shared
2782 volumes, the source mount point has to be shared. And for slave vol‐
2783 umes, the source mount point has to be either shared or slave. [1]
2784 ⟨#Footnote1⟩
2785 To recursively mount a volume and all of its submounts into a con‐
2788 tainer, use the rbind option. By default the bind option is used, and
2789 submounts of the source directory is not mounted into the container.
2790 Mounting the volume with a copy option tells podman to copy content
2793 from the underlying destination directory onto newly created internal
2794 volumes. The copy only happens on the initial creation of the volume.
2795 Content is not copied up when the volume is subsequently used on dif‐
2796 ferent containers. The copy option is ignored on bind mounts and has no
2797 effect.
2798 Mounting volumes with the nosuid options means that SUID executables on
2801 the volume can not be used by applications to change their privilege.
2802 By default volumes are mounted with nosuid.
2803 Mounting the volume with the noexec option means that no executables on
2806 the volume can be executed within the container.
2807 Mounting the volume with the nodev option means that no devices on the
2810 volume can be used by processes within the container. By default vol‐
2811 umes are mounted with nodev.
2812 If the HOST-DIR is a mount point, then dev, suid, and exec options are
2815 ignored by the kernel.
2816 Use df HOST-DIR to figure out the source mount, then use findmnt -o
2819 TARGET,PROPAGATION source-mount-dir to figure out propagation proper‐
2820 ties of source mount. If findmnt[4m(1) utility is not available, then one
2821 can look at the mount entry for the source mount point in
2822 /proc/self/mountinfo. Look at the "optional fields" and see if any
2823 propagation properties are specified. In there, shared:N means the
2824 mount is shared, master:N means mount is slave, and if nothing is
2825 there, the mount is private. [1] ⟨#Footnote1⟩
2826 To change propagation properties of a mount point, use mount(8) com‐
2829 mand. For example, if one wants to bind mount source directory /foo,
2830 one can do mount --bind /foo /foo and mount --make-private --make-
2831 shared /foo. This converts /foo into a shared mount point. Alterna‐
2832 tively, one can directly change propagation properties of source mount.
2833 Say / is source mount for /foo, then use mount --make-shared / to con‐
2834 vert / into a shared mount.
2835 Note: if the user only has access rights via a group, accessing the
2838 volume from inside a rootless container fails.
2839 Idmapped mount
2842 If idmap is specified, create an idmapped mount to the target user
2845 namespace in the container. The idmap option supports a custom mapping
2846 that can be different than the user namespace used by the container.
2847 The mapping can be specified after the idmap option like:
2848 idmap=uids=0-1-10#10-11-10;gids=0-100-10. For each triplet, the first
2849 value is the start of the backing file system IDs that are mapped to
2850 the second value on the host. The length of this mapping is given in
2851 the third value. Multiple ranges are separated with #.
2852 Use the --group-add keep-groups option to pass the user's supplementary
2855 group access into the container.
2856 --volumes-from=CONTAINER[:OPTIONS]
2859 Mount volumes from the specified container(s). Used to share volumes
2860 between containers. The options is a comma-separated list with the fol‐
2861 lowing available elements:
2862 • rw|ro
2865 • z
2867 Mounts already mounted volumes from a source container onto another
2871 container. CONTAINER may be a name or ID. To share a volume, use the
2872 --volumes-from option when running the target container. Volumes can be
2873 shared even if the source container is not running.
2874 By default, Podman mounts the volumes in the same mode (read-write or
2877 read-only) as it is mounted in the source container. This can be
2878 changed by adding a ro or rw option.
2879 Labeling systems like SELinux require that proper labels are placed on
2882 volume content mounted into a container. Without a label, the security
2883 system might prevent the processes running inside the container from
2884 using the content. By default, Podman does not change the labels set by
2885 the OS.
2886 To change a label in the container context, add z to the volume mount.
2889 This suffix tells Podman to relabel file objects on the shared volumes.
2890 The z option tells Podman that two entities share the volume content.
2891 As a result, Podman labels the content with a shared content label.
2892 Shared volume labels allow all containers to read/write content.
2893 If the location of the volume from the source container overlaps with
2896 data residing on a target container, then the volume hides that data on
2897 the target.
2898 --workdir, -w=dir
2901 Working directory inside the container.
2902 The default working directory for running binaries within a container
2905 is the root directory (/). The image developer can set a different de‐
2906 fault with the WORKDIR instruction. The operator can override the work‐
2907 ing directory by using the -w option.
2908
2911 The exit code from podman run gives information about why the container
2912 failed to run or why it exited. When podman run exits with a non-zero
2913 code, the exit codes follow the chroot(1) standard, see below:
2914 125 The error is with Podman itself
2917 $ podman run --foo busybox; echo $?
2919 Error: unknown flag: --foo
2920 125
2921 126 The contained command cannot be invoked
2925 $ podman run busybox /etc; echo $?
2927 Error: container_linux.go:346: starting container process caused "exec: \"/etc\": permission denied": OCI runtime error
2928 126
2929 127 The contained command cannot be found
2933 $ podman run busybox foo; echo $?
2935 Error: container_linux.go:346: starting container process caused "exec: \"foo\": executable file not found in $PATH": OCI runtime error
2936 127
2937 Exit code contained command exit code
2941 $ podman run busybox /bin/sh -c 'exit 3'; echo $?
2943 3
2944
2948 Running container in read-only mode
2949 During container image development, containers often need to write to
2950 the image content. Installing packages into /usr, for example. In pro‐
2951 duction, applications seldom need to write to the image. Container ap‐
2952 plications write to volumes if they need to write to file systems at
2953 all. Applications can be made more secure by running them in read-only
2954 mode using the --read-only switch. This protects the container's image
2955 from modification. By default read-only containers can write to tempo‐
2956 rary data. Podman mounts a tmpfs on /run and /tmp within the container.
2957 If the container does not write to any file system within the con‐
2958 tainer, including tmpfs, set --read-only-tmpfs=false.
2959 $ podman run --read-only -i -t fedora /bin/bash
2961 $ podman run --read-only --read-only-tmpfs=false --tmpfs /run -i -t fedora /bin/bash
2963 Exposing shared libraries inside of container as read-only using a glob
2967 $ podman run --mount type=glob,src=/usr/lib64/libnvidia\*,ro=true -i -t fedora /bin/bash
2968 Exposing log messages from the container to the host's log
2972 Bind mount the /dev/log directory to have messages that are logged in
2973 the container show up in the host's syslog/journal.
2974 $ podman run -v /dev/log:/dev/log -i -t fedora /bin/bash
2976 From inside the container test this by sending a message to the log.
2980 (bash)# logger "Hello from my container"
2982 Then exit and check the journal.
2986 (bash)# exit
2988 $ journalctl -b | grep Hello
2990 This lists the message sent to the logger.
2994 Attaching to one or more from STDIN, STDOUT, STDERR
2997 Without specifying the -a option, Podman attaches everything (stdin,
2998 stdout, stderr). Override the default by specifying -a (stdin, stdout,
2999 stderr), as in:
3000 $ podman run -a stdin -a stdout -i -t fedora /bin/bash
3002 Sharing IPC between containers
3006 Using shm_server.c available here:
3007 https://www.cs.cf.ac.uk/Dave/C/node27.html
3008 Testing --ipc=host mode:
3011 Host shows a shared memory segment with 7 pids attached, happens to be
3014 from httpd:
3015 $ sudo ipcs -m
3017 ------ Shared Memory Segments --------
3019 key shmid owner perms bytes nattch status
3020 0x01128e25 0 root 600 1000 7
3021 Now run a regular container, and it correctly does NOT see the shared
3025 memory segment from the host:
3026 $ podman run -it shm ipcs -m
3028 ------ Shared Memory Segments --------
3030 key shmid owner perms bytes nattch status
3031 Run a container with the new --ipc=host option, and it now sees the
3035 shared memory segment from the host httpd:
3036 $ podman run -it --ipc=host shm ipcs -m
3038 ------ Shared Memory Segments --------
3040 key shmid owner perms bytes nattch status
3041 0x01128e25 0 root 600 1000 7
3042 Testing --ipc=container:id mode:
3046 Start a container with a program to create a shared memory segment:
3049 $ podman run -it shm bash
3051 $ sudo shm/shm_server &
3052 $ sudo ipcs -m
3053 ------ Shared Memory Segments --------
3055 key shmid owner perms bytes nattch status
3056 0x0000162e 0 root 666 27 1
3057 Create a 2nd container correctly shows no shared memory segment from
3061 1st container:
3062 $ podman run shm ipcs -m
3064 ------ Shared Memory Segments --------
3066 key shmid owner perms bytes nattch status
3067 Create a 3rd container using the --ipc=container:id option, now it
3071 shows the shared memory segment from the first:
3072 $ podman run -it --ipc=container:ed735b2264ac shm ipcs -m
3074 $ sudo ipcs -m
3075 ------ Shared Memory Segments --------
3077 key shmid owner perms bytes nattch status
3078 0x0000162e 0 root 666 27 1
3079 Mapping Ports for External Usage
3083 The exposed port of an application can be mapped to a host port using
3084 the -p flag. For example, an httpd port 80 can be mapped to the host
3085 port 8080 using the following:
3086 $ podman run -p 8080:80 -d -i -t fedora/httpd
3088 Mounting External Volumes
3092 To mount a host directory as a container volume, specify the absolute
3093 path to the directory and the absolute path for the container directory
3094 separated by a colon. If the source is a named volume maintained by
3095 Podman, it is recommended to use its name rather than the path to the
3096 volume. Otherwise the volume is considered an orphan and wiped by the
3097 podman volume prune command:
3098 $ podman run -v /var/db:/data1 -i -t fedora bash
3100 $ podman run -v data:/data2 -i -t fedora bash
3102 $ podman run -v /var/cache/dnf:/var/cache/dnf:O -ti fedora dnf -y update
3104 If the container needs a writeable mounted volume by a non root user
3108 inside the container, use the U option. This option tells Podman to
3109 chown the source volume to match the default UID and GID used within
3110 the container.
3111 $ podman run -d -e MYSQL_ROOT_PASSWORD=root --user mysql --userns=keep-id -v ~/data:/var/lib/mysql:Z,U mariadb
3113 Alternatively if the container needs a writable volume by a non root
3117 user inside of the container, the --userns=keep-id option allows users
3118 to specify the UID and GID of the user executing Podman to specific
3119 UIDs and GIDs within the container. Since the processes running in the
3120 container run as the user's UID, they can read/write files owned by the
3121 user.
3122 $ podman run -d -e MYSQL_ROOT_PASSWORD=root --user mysql --userns=keep-id:uid=999,gid=999 -v ~/data:/var/lib/mysql:Z mariadb
3124 Using --mount flags to mount a host directory as a container folder,
3128 specify the absolute path to the directory or the volume name, and the
3129 absolute path within the container directory:
3130 $ podman run --mount type=bind,src=/var/db,target=/data1 busybox sh
3132 $ podman run --mount type=bind,src=volume-name,target=/data1 busybox sh
3134 When using SELinux, be aware that the host has no knowledge of con‐
3138 tainer SELinux policy. Therefore, in the above example, if SELinux pol‐
3139 icy is enforced, the /var/db directory is not writable to the con‐
3140 tainer. A "Permission Denied" message occurs, and an avc: message is
3141 added to the host's syslog.
3142 To work around this, at time of writing this man page, the following
3145 command needs to be run in order for the proper SELinux policy type la‐
3146 bel to be attached to the host directory:
3147 $ chcon -Rt svirt_sandbox_file_t /var/db
3149 Now, writing to the /data1 volume in the container is allowed and the
3153 changes are reflected on the host in /var/db.
3154 Using alternative security labeling
3157 Override the default labeling scheme for each container by specifying
3158 the --security-opt flag. For example, specify the MCS/MLS level, a re‐
3159 quirement for MLS systems. Specifying the level in the following com‐
3160 mand allows the same content to be shared between containers.
3161 podman run --security-opt label=level:s0:c100,c200 -i -t fedora bash
3163 An MLS example might be:
3167 $ podman run --security-opt label=level:TopSecret -i -t rhel7 bash
3169 To disable the security labeling for this container versus running with
3173 the
3174 --permissive flag, use the following command:
3177 $ podman run --security-opt label=disable -i -t fedora bash
3178 Tighten the security policy on the processes within a container by
3182 specifying an alternate type for the container. For example, run a con‐
3183 tainer that is only allowed to listen on Apache ports by executing the
3184 following command:
3185 $ podman run --security-opt label=type:svirt_apache_t -i -t centos bash
3187 Note that an SELinux policy defining a svirt_apache_t type must be
3191 written.
3192 To mask additional specific paths in the container, specify the paths
3195 separated by a colon using the mask option with the --security-opt
3196 flag.
3197 $ podman run --security-opt mask=/foo/bar:/second/path fedora bash
3199 To unmask all the paths that are masked by default, set the unmask op‐
3203 tion to ALL. Or to only unmask specific paths, specify the paths as
3204 shown above with the mask option.
3205 $ podman run --security-opt unmask=ALL fedora bash
3207 To unmask all the paths that start with /proc, set the unmask option to
3211 /proc/*.
3212 $ podman run --security-opt unmask=/proc/* fedora bash
3214 $ podman run --security-opt unmask=/foo/bar:/sys/firmware fedora bash
3217 Setting device weight via --blkio-weight-device flag.
3221 $ podman run -it --blkio-weight-device "/dev/sda:200" ubuntu
3222 Using a podman container with input from a pipe
3226 $ echo "asdf" | podman run --rm -i --entrypoint /bin/cat someimage
3227 asdf
3228 Setting automatic user namespace separated containers
3232 # podman run --userns=auto:size=65536 ubi8-micro cat /proc/self/uid_map
3233 0 2147483647 65536
3234 # podman run --userns=auto:size=65536 ubi8-micro cat /proc/self/uid_map
3235 0 2147549183 65536
3236 Setting Namespaced Kernel Parameters (Sysctls)
3240 The --sysctl sets namespaced kernel parameters (sysctls) in the con‐
3241 tainer. For example, to turn on IP forwarding in the containers network
3242 namespace, run this command:
3243 $ podman run --sysctl net.ipv4.ip_forward=1 someimage
3245 Note that not all sysctls are namespaced. Podman does not support
3249 changing sysctls inside of a container that also modify the host sys‐
3250 tem. As the kernel evolves we expect to see more sysctls become names‐
3251 paced.
3252 See the definition of the --sysctl option above for the current list of
3255 supported sysctls.
3256 Set UID/GID mapping in a new user namespace
3259 Running a container in a new user namespace requires a mapping of the
3260 UIDs and GIDs from the host.
3261 $ podman run --uidmap 0:30000:7000 --gidmap 0:30000:7000 fedora echo hello
3263 Configuring Storage Options from the command line
3267 Podman allows for the configuration of storage by changing the values
3268 in the /etc/container/storage.conf or by using global options. This
3269 shows how to set up and use fuse-overlayfs for a one-time run of busy‐
3270 box using global options.
3271 podman --log-level=debug --storage-driver overlay --storage-opt "overlay.mount_program=/usr/bin/fuse-overlayfs" run busybox /bin/sh
3273 Configure timezone in a container
3277 $ podman run --tz=local alpine date
3278 $ podman run --tz=Asia/Shanghai alpine date
3279 $ podman run --tz=US/Eastern alpine date
3280 Adding dependency containers
3284 The first container, container1, is not started initially, but must be
3285 running before container2 starts. The podman run command starts the
3286 container automatically before starting container2.
3287 $ podman create --name container1 -t -i fedora bash
3289 $ podman run --name container2 --requires container1 -t -i fedora bash
3290 Multiple containers can be required.
3294 $ podman create --name container1 -t -i fedora bash
3296 $ podman create --name container2 -t -i fedora bash
3297 $ podman run --name container3 --requires container1,container2 -t -i fedora bash
3298 Configure keep supplemental groups for access to volume
3302 $ podman run -v /var/lib/design:/var/lib/design --group-add keep-groups ubi8
3303 Configure execution domain for containers using personality flag
3307 $ podman run --name container1 --personality=LINUX32 fedora bash
3308 Run a container with external rootfs mounted as an overlay
3312 $ podman run --name container1 --rootfs /path/to/rootfs:O bash
3313 Handling Timezones in java applications in a container.
3317 In order to use a timezone other than UTC when running a Java applica‐
3318 tion within a container, the TZ environment variable must be set within
3319 the container. Java applications ignores the value set with the --tz
3320 option.
3321 # Example run
3323 podman run -ti --rm -e TZ=EST mytzimage
3324 lrwxrwxrwx. 1 root root 29 Nov 3 08:51 /etc/localtime -> ../usr/share/zoneinfo/Etc/UTC
3325 Now with default timezone:
3326 Fri Nov 19 18:10:55 EST 2021
3327 Java default sees the following timezone:
3328 2021-11-19T18:10:55.651130-05:00
3329 Forcing UTC:
3330 Fri Nov 19 23:10:55 UTC 2021
3331 Run a container connected to two networks (called net1 and net2) with a
3335 static ip
3336 $ podman run --network net1:ip=10.89.1.5 --network net2:ip=10.89.10.10 alpine ip addr
3337 Rootless Containers
3341 Podman runs as a non-root user on most systems. This feature requires
3342 that a new enough version of shadow-utils be installed. The shadow-
3343 utils package must include the newuidmap(1) and newgidmap(1) executa‐
3344 bles.
3345 In order for users to run rootless, there must be an entry for their
3348 username in /etc/subuid and /etc/subgid which lists the UIDs for their
3349 user namespace.
3350 Rootless Podman works better if the fuse-overlayfs and slirp4netns
3353 packages are installed. The fuse-overlayfs package provides a
3354 userspace overlay storage driver, otherwise users need to use the vfs
3355 storage driver, which can be disk space expensive and less performant
3356 than other drivers.
3357 To enable VPN on the container, slirp4netns or pasta needs to be speci‐
3360 fied; without either, containers need to be run with the --network=host
3361 flag.
3362
3365 Environment variables within containers can be set using multiple dif‐
3366 ferent options, in the following order of precedence (later entries
3367 override earlier entries):
3368 • Container image: Any environment variables specified in the
3371 container image.
3372 • --http-proxy: By default, several environment variables are
3374 passed in from the host, such as http_proxy and no_proxy. See
3375 --http-proxy for details.
3376 • --env-host: Host environment of the process executing Podman
3378 is added.
3379 • --env-file: Any environment variables specified via env-files.
3381 If multiple files are specified, then they override each other
3382 in order of entry.
3383 • --env: Any environment variables specified overrides previous
3385 settings.
3386 Run containers and set the environment ending with a *. The trailing *
3390 glob functionality is only active when no value is specified:
3391 $ export ENV1=a
3393 $ podman run --env 'ENV*' alpine env | grep ENV
3394 ENV1=a
3395 $ podman run --env 'ENV*=b' alpine env | grep ENV
3396 ENV*=b
3397
3401 When Podman starts a container it actually executes the conmon program,
3402 which then executes the OCI Runtime. Conmon is the container monitor.
3403 It is a small program whose job is to watch the primary process of the
3404 container, and if the container dies, save the exit code. It also
3405 holds open the tty of the container, so that it can be attached to
3406 later. This is what allows Podman to run in detached mode (back‐
3407 grounded), so Podman can exit but conmon continues to run. Each con‐
3408 tainer has their own instance of conmon. Conmon waits for the container
3409 to exit, gathers and saves the exit code, and then launches a Podman
3410 process to complete the container cleanup, by shutting down the network
3411 and storage. For more information about conmon, see the conmon(8) man
3412 page.
3413
3416 /etc/subuid
3417 /etc/subgid
3420 NOTE: Use the environment variable TMPDIR to change the temporary stor‐
3423 age location of downloaded container images. Podman defaults to use
3424 /var/tmp.
3425
3428 podman(1), podman-save(1), podman-ps(1), podman-attach(1), podman-pod-
3429 create(1), podman-port(1), podman-start(1), podman-kill(1), podman-
3430 stop(1), podman-generate-systemd(1), podman-rm(1), subgid(5), sub‐
3431 uid(5), containers.conf(5), systemd.unit(5), setsebool(8),
3432 slirp4netns(1), pasta(1), fuse-overlayfs(1), proc(5), conmon(8), per‐
3433 sonality(2)
3434
3437 September 2018, updated by Kunal Kushwaha <kushwaha_ku‐
3438 nal_v7@lab.ntt.co.jp>
3439 October 2017, converted from Docker documentation to Podman by Dan
3442 Walsh for Podman <dwalsh@redhat.com>
3443 November 2015, updated by Sally O'Malley <somalley@redhat.com>
3446 June 2014, updated by Sven Dowideit <SvenDowideit@home.org.au>
3449 April 2014, Originally compiled by William Henry <whenry@redhat.com>
3452 based on docker.com source material and internal work.
3453
3456 1: The Podman project is committed to inclusivity, a core value of open
3457 source. The master and slave mount propagation terminology used here is
3458 problematic and divisive, and needs to be changed. However, these terms
3459 are currently used within the Linux kernel and must be used as-is at
3460 this time. When the kernel maintainers rectify this usage, Podman will
3461 follow suit immediately.
3462 podman-run(1)