1podman-run(1)() 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. When using the --privileged flag the .con‐
39 tainerenv contains name/value pairs indicating the container engine
40 version, whether the engine is running in rootless mode, the container
41 name and id, as well as the image name and id that the container is
42 based on.
43 When running from a user defined network namespace, the
46 /etc/netns/NSNAME/resolv.conf will be used if it exists, otherwise
47 /etc/resolv.conf will be used.
48 Default settings are defined in containers.conf. Most settings for re‐
51 mote connections use the servers containers.conf, except when docu‐
52 mented in man pages.
53
56 The image is specified using transport:path format. If no transport is
57 specified, the docker (container registry) transport will be used by
58 default. For remote Podman, docker is the only allowed transport.
59 dir:path
62 An existing local directory path storing the manifest, layer tarballs
63 and signatures as individual files. This is a non-standardized format,
64 primarily useful for debugging or noninvasive container inspection.
65 $ podman save --format docker-dir fedora -o /tmp/fedora
68 $ podman run dir:/tmp/fedora echo hello
69 docker://docker-reference (Default)
73 An image reference stored in a remote container image registry. Exam‐
74 ple: "quay.io/podman/stable:latest". The reference can include a path
75 to a specific registry; if it does not, the registries listed in reg‐
76 istries.conf will be queried to find a matching image. By default,
77 credentials from podman login (stored at $XDG_RUNTIME_DIR/contain‐
78 ers/auth.json by default) will be used to authenticate; otherwise it
79 falls back to using credentials in $HOME/.docker/config.json.
80 $ podman run registry.fedoraproject.org/fedora:latest echo hello
83 docker-archive:path[:docker-reference] An image stored in the docker
87 save formatted file. docker-reference is only used when creating such a
88 file, and it must not contain a digest.
89 $ 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
99 (docker-daemon:algo:digest).
100 $ sudo docker pull fedora
103 $ sudo podman run docker-daemon:docker.io/library/fedora echo hello
104 oci-archive:path:tag
108 An image in a directory compliant with the "Open Container Image Lay‐
109 out Specification" at the specified path and specified with a tag.
110 $ podman save --format oci-archive fedora -o /tmp/fedora
113 $ podman run oci-archive:/tmp/fedora echo hello
114
118 --add-host=host:ip
119 Add a line to container's /etc/hosts for custom host-to-IP mapping.
120 This option can be set multiple times.
121 --annotation=key=value
124 Add an annotation to the container. This option can be set multiple
125 times.
126 --arch=ARCH
129 Override the architecture, defaults to hosts, of the image to be
130 pulled. For example, arm.
131 --attach, -a=stdin|stdout|stderr
134 Attach to STDIN, STDOUT or STDERR.
135 In foreground mode (the default when -d is not specified), podman run
138 can start the process in the container and attach the console to the
139 process's standard input, output, and error. It can even pretend to be
140 a TTY (this is what most commandline executables expect) and pass along
141 signals. The -a option can be set for each of stdin, stdout, and
142 stderr.
143 --authfile[=path]
146 Path to the authentication file. Default is ${XDG_RUNTIME_DIR}/contain‐
147 ers/auth.json.
148 Note: You can also override the default path of the authentication file
151 by setting the REGISTRY_AUTH_FILE environment variable.
152 --blkio-weight=weight
155 Block IO relative weight. The weight is a value between 10 and 1000.
156 --blkio-weight-device=device:weight
159 Block IO relative device weight.
160 --cap-add=capability
163 Add Linux capabilities.
164 --cap-drop=capability
167 Drop Linux capabilities.
168 --cgroupns=mode
171 Set the cgroup namespace mode for the container.
172 • host: use the host's cgroup namespace inside the container.
175 • container:id: join the namespace of the specified container.
177 • private: create a new cgroup namespace.
179 • ns:path: join the namespace at the specified path.
181 If the host uses cgroups v1, the default is set to host. On cgroups v2,
185 the default is private.
186 --cgroups=enabled|disabled|no-conmon|split
189 Determines whether the container will create CGroups.
190 Default is enabled.
193 The enabled option will create a new cgroup under the cgroup-parent.
196 The disabled option will force the container to not create CGroups, and
197 thus conflicts with CGroup options (--cgroupns and --cgroup-parent).
198 The no-conmon option disables a new CGroup only for the conmon process.
199 The split option splits the current CGroup in two sub-cgroups: one for
200 conmon and one for the container payload. It is not possible to set
201 --cgroup-parent with split.
202 --cgroup-parent=path
205 Path to cgroups under which the cgroup for the container will be cre‐
206 ated. If the path is not absolute, the path is considered to be rela‐
207 tive to the cgroups path of the init process. Cgroups will be created
208 if they do not already exist.
209 --cgroup-conf=KEY=VALUE
212 When running on cgroup v2, specify the cgroup file to write to and its
213 value. For example --cgroup-conf=memory.high=1073741824 sets the mem‐
214 ory.high limit to 1GB.
215 --cidfile=file
218 Write the container ID to file.
219 --conmon-pidfile=file
222 Write the pid of the conmon process to a file. As conmon runs in a sep‐
223 arate process than Podman, this is necessary when using systemd to
224 restart Podman containers. (This option is not available with the re‐
225 mote Podman client)
226 --cpu-period=limit
229 Set the CPU period for the Completely Fair Scheduler (CFS), which is a
230 duration in microseconds. Once the container's CPU quota is used up, it
231 will not be scheduled to run until the current period ends. Defaults to
232 100000 microseconds.
233 On some systems, changing the CPU limits may not be allowed for
236 non-root users. For more details, see https://github.com/contain‐
237 ers/podman/blob/master/troubleshooting.md#26-running-contain‐
238 ers-with-cpu-limits-fails-with-a-permissions-error
239 --cpu-quota=limit
242 Limit the CPU Completely Fair Scheduler (CFS) quota.
243 Limit the container's CPU usage. By default, containers run with the
246 full CPU resource. The limit is a number in microseconds. If you pro‐
247 vide a number, the container will be allowed to use that much CPU time
248 until the CPU period ends (controllable via --cpu-period).
249 On some systems, changing the CPU limits may not be allowed for
252 non-root users. For more details, see https://github.com/contain‐
253 ers/podman/blob/master/troubleshooting.md#26-running-contain‐
254 ers-with-cpu-limits-fails-with-a-permissions-error
255 --cpu-rt-period=microseconds
258 Limit the CPU real-time period in microseconds.
259 Limit the container's Real Time CPU usage. This flag tell the kernel to
262 restrict the container's Real Time CPU usage to the period you specify.
263 --cpu-rt-runtime=microseconds
266 Limit the CPU real-time runtime in microseconds.
267 Limit the containers Real Time CPU usage. This flag tells the kernel to
270 limit the amount of time in a given CPU period Real Time tasks may con‐
271 sume. Ex: Period of 1,000,000us and Runtime of 950,000us means that
272 this container could consume 95% of available CPU and leave the remain‐
273 ing 5% to normal priority tasks.
274 The sum of all runtimes across containers cannot exceed the amount al‐
277 lotted to the parent cgroup.
278 --cpu-shares=shares
281 CPU shares (relative weight).
282 By default, all containers get the same proportion of CPU cycles. This
285 proportion can be modified by changing the container's CPU share
286 weighting relative to the combined weight of all the running contain‐
287 ers. Default weight is 1024.
288 The proportion will only apply when CPU-intensive processes are run‐
291 ning. When tasks in one container are idle, other containers can use
292 the left-over CPU time. The actual amount of CPU time will vary depend‐
293 ing on the number of containers running on the system.
294 For example, consider three containers, one has a cpu-share of 1024 and
297 two others have a cpu-share setting of 512. When processes in all three
298 containers attempt to use 100% of CPU, the first container would re‐
299 ceive 50% of the total CPU time. If you add a fourth container with a
300 cpu-share of 1024, the first container only gets 33% of the CPU. The
301 remaining containers receive 16.5%, 16.5% and 33% of the CPU.
302 On a multi-core system, the shares of CPU time are distributed over all
305 CPU cores. Even if a container is limited to less than 100% of CPU
306 time, it can use 100% of each individual CPU core.
307 For example, consider a system with more than three cores. If you start
310 one container {C0} with --cpu-shares=512 running one process, and an‐
311 other container {C1} with --cpu-shares=1024 running two processes, this
312 can result in the following division of CPU shares:
313 ┌────┬───────────┬─────┬──────────────┐
316 │PID │ container │ CPU │ CPU share │
317 ├────┼───────────┼─────┼──────────────┤
318 │100 │ {C0} │ 0 │ 100% of CPU0 │
319 ├────┼───────────┼─────┼──────────────┤
320 │101 │ {C1} │ 1 │ 100% of CPU1 │
321 ├────┼───────────┼─────┼──────────────┤
322 │102 │ {C1} │ 2 │ 100% of CPU2 │
323 ├────┼───────────┼─────┼──────────────┤
324 │ │ │ │ │
325 └────┴───────────┴─────┴──────────────┘
326 --cpus=number
328 Number of CPUs. The default is 0.0 which means no limit. This is short‐
329 hand for --cpu-period and --cpu-quota, so you may only set either
330 --cpus or --cpu-period and --cpu-quota.
333 On some systems, changing the CPU limits may not be allowed for
334 non-root users. For more details, see https://github.com/contain‐
335 ers/podman/blob/master/troubleshooting.md#26-running-contain‐
336 ers-with-cpu-limits-fails-with-a-permissions-error
337 --cpuset-cpus=number
340 CPUs in which to allow execution. Can be specified as a comma-separated
341 list (e.g. 0,1), as a range (e.g. 0-3), or any combination thereof
342 (e.g. 0-3,7,11-15).
343 --cpuset-mems=nodes
346 Memory nodes (MEMs) in which to allow execution. Only effective on NUMA
347 systems.
348 For example, if you have four memory nodes (0-3) on your system, use
351 --cpuset-mems=0,1 to only use memory from the first two memory nodes.
352 --detach, -d=true|false
355 Detached mode: run the container in the background and print the new
356 container ID. The default is false.
357 At any time you can run podman ps in the other shell to view a list of
360 the running containers. You can reattach to a detached container with
361 podman attach.
362 When attached in the tty mode, you can detach from the container (and
365 leave it running) using a configurable key sequence. The default se‐
366 quence is ctrl-p,ctrl-q. Configure the keys sequence using the --de‐
367 tach-keys option, or specifying it in the containers.conf file: see
368 containers.conf(5) for more information.
369 --detach-keys=sequence
372 Specify the key sequence for detaching a container. Format is a single
373 character [a-Z] or one or more ctrl-<value> characters where <value> is
374 one of: a-z, @, ^, [, , or _. Specifying "" will disable this feature.
375 The default is ctrl-p,ctrl-q.
376 This option can also be set in containers.conf(5) file.
379 Specifying "" will disable this feature. The default is ctrl-p,ctrl-q.
382 --device=host-device[:container-device][:permissions]
385 Add a host device to the container. Optional permissions parameter can
386 be used to specify device permissions, it is combination of r for read,
387 w for write, and m for mknod(2).
388 Example: --device=/dev/sdc:/dev/xvdc:rwm.
391 Note: if _hostdevice is a symbolic link then it will be resolved first.
394 The container will only store the major and minor numbers of the host
395 device.
396 Note: if the user only has access rights via a group, accessing the de‐
399 vice from inside a rootless container will fail. Use the --group-add
400 keep-groups flag to pass the user's supplementary group access into the
401 container.
402 Podman may load kernel modules required for using the specified device.
405 The devices that Podman will load modules when necessary are:
406 /dev/fuse.
407 --device-cgroup-rule=rule
410 Add a rule to the cgroup allowed devices list
411 --device-read-bps=path:rate
414 Limit read rate (in bytes per second) from a device (e.g. --de‐
415 vice-read-bps=/dev/sda:1mb).
416 --device-read-iops=path:rate
419 Limit read rate (in IO operations per second) from a device (e.g. --de‐
420 vice-read-iops=/dev/sda:1000).
421 --device-write-bps=path:rate
424 Limit write rate (in bytes per second) to a device (e.g. --de‐
425 vice-write-bps=/dev/sda:1mb).
426 --device-write-iops=path:rate
429 Limit write rate (in IO operations per second) to a device (e.g. --de‐
430 vice-write-iops=/dev/sda:1000).
431 --disable-content-trust
434 This is a Docker specific option to disable image verification to a
435 Docker registry and is not supported by Podman. This flag is a NOOP and
436 provided solely for scripting compatibility.
437 --dns=ipaddr
440 Set custom DNS servers. Invalid if using --dns with --network that is
441 set to none or container:id.
442 This option can be used to override the DNS configuration passed to the
445 container. Typically this is necessary when the host DNS configuration
446 is invalid for the container (e.g., 127.0.0.1). When this is the case
447 the --dns flags is necessary for every run.
448 The special value none can be specified to disable creation of /etc/re‐
451 solv.conf in the container by Podman. The /etc/resolv.conf file in the
452 image will be used without changes.
453 --dns-opt=option
456 Set custom DNS options. Invalid if using --dns-opt with --network that
457 is set to none or container:id.
458 --dns-search=domain
461 Set custom DNS search domains. Invalid if using --dns-search and --net‐
462 work that is set to none or container:id. Use --dns-search=. if you
463 don't wish to set the search domain.
464 --entrypoint="command" | '["command", arg1 , ...]'
467 Overwrite the default ENTRYPOINT of the image.
468 This option allows you to overwrite the default entrypoint of the im‐
471 age.
472 The ENTRYPOINT of an image is similar to a COMMAND because it specifies
475 what executable to run when the container starts, but it is (purposely)
476 more difficult to override. The ENTRYPOINT gives a container its de‐
477 fault nature or behavior, so that when you set an ENTRYPOINT you can
478 run the container as if it were that binary, complete with default op‐
479 tions, and you can pass in more options via the COMMAND. But, sometimes
480 an operator may want to run something else inside the container, so you
481 can override the default ENTRYPOINT at runtime by using a --entrypoint
482 and a string to specify the new ENTRYPOINT.
483 You need to specify multi option commands in the form of a json string.
486 --env, -e=env
489 Set environment variables.
490 This option allows arbitrary environment variables that are available
493 for the process to be launched inside of the container. If an environ‐
494 ment variable is specified without a value, Podman will check the host
495 environment for a value and set the variable only if it is set on the
496 host. If an environment variable ending in * is specified, Podman will
497 search the host environment for variables starting with the prefix and
498 will add those variables to the container. If an environment variable
499 with a trailing ***** is specified, then a value must be supplied.
500 See Environment ⟨#environment⟩ note below for precedence and examples.
503 --env-host=true|false
506 Use host environment inside of the container. See Environment note be‐
507 low for precedence. (This option is not available with the remote Pod‐
508 man client)
509 --env-file=file
512 Read in a line delimited file of environment variables. See Environment
513 note below for precedence.
514 --expose=port
517 Expose a port, or a range of ports (e.g. --expose=3300-3310) to set up
518 port redirection on the host system.
519 --gidmap=container_gid:host_gid:amount
522 Run the container in a new user namespace using the supplied mapping.
523 This option conflicts with the --userns and --subgidname flags. This
524 option can be passed several times to map different ranges. If calling
525 podman run as an unprivileged user, the user needs to have the right to
526 use the mapping. See subuid(5). The example maps gids 0-1999 in the
527 container to the gids 30000-31999 on the host: --gidmap=0:30000:2000.
528 Important note: The new user namespace mapping based on --gidmap is
531 based on the initial mapping made in the /etc/subgid file. Assuming
532 there is a /etc/subgid mapping groupname:100000:65536, then groupname
533 is initially mapped to a namespace starting with gid 100000 for 65536
534 ids. From here the --gidmap mapping to the new namespace starts from 0
535 again, but is based on the initial mapping. Meaning groupname is ini‐
536 tially mapped to gid 100000 which is referenced as 0 in the following
537 --gidmap mapping. In terms of the example above: The group groupname is
538 mapped to group 100000 of the initial namespace then the 30000st id of
539 this namespace (which is gid 130000 in this namespace) is mapped to
540 container namespace group id 0. (groupname -> 100000 / 30000 -> 0)
541 --group-add=group|keep-groups
544 Add additional groups to assign to primary user running within the con‐
545 tainer process.
546 • keep-groups is a special flag that tells Podman to keep the
549 supplementary group access.
550 Allows container to use the user's supplementary group access. If file
554 systems or devices are only accessible by the rootless user's group,
555 this flag tells the OCI runtime to pass the group access into the con‐
556 tainer. Currently only available with the crun OCI runtime. Note:
557 keep-groups is exclusive, you cannot add any other groups with this
558 flag. (Not available for remote commands)
559 --health-cmd="command" | '["command", arg1 , ...]'
562 Set or alter a healthcheck command for a container. The command is a
563 command to be executed inside your container that determines your con‐
564 tainer health. The command is required for other healthcheck options to
565 be applied. A value of none disables existing healthchecks.
566 Multiple options can be passed in the form of a JSON array; otherwise,
569 the command will be interpreted as an argument to /bin/sh -c.
570 --health-interval=interval
573 Set an interval for the healthchecks. An interval of disable results in
574 no automatic timer setup. The default is 30s.
575 --health-retries=retries
578 The number of retries allowed before a healthcheck is considered to be
579 unhealthy. The default value is 3.
580 --health-start-period=period
583 The initialization time needed for a container to bootstrap. The value
584 can be expressed in time format like 2m3s. The default value is 0s.
585 --health-timeout=timeout
588 The maximum time allowed to complete the healthcheck before an interval
589 is considered failed. Like start-period, the value can be expressed in
590 a time format such as 1m22s. The default value is 30s.
591 --help
594 Print usage statement
595 --hostname=name, -h
598 Container host name
599 Sets the container host name that is available inside the container.
602 Can only be used with a private UTS namespace --uts=private (default).
603 If --pod is specified and the pod shares the UTS namespace (default)
604 the pod's hostname will be used.
605 --http-proxy=true|false
608 By default proxy environment variables are passed into the container if
609 set for the Podman process. This can be disabled by setting the value
610 to false. The environment variables passed in include http_proxy,
611 https_proxy, ftp_proxy, no_proxy, and also the upper case versions of
612 those. This option is only needed when the host system must use a proxy
613 but the container should not use any proxy. Proxy environment variables
614 specified for the container in any other way will override the values
615 that would have been passed through from the host. (Other ways to spec‐
616 ify the proxy for the container include passing the values with the
617 --env flag, or hard coding the proxy environment at container build
618 time.) (This option is not available with the remote Podman client)
619 Defaults to true.
622 --image-volume, builtin-volume=bind|tmpfs|ignore
625 Tells Podman how to handle the builtin image volumes. Default is bind.
626 • bind: An anonymous named volume will be created and mounted
629 into the container.
630 • tmpfs: The volume is mounted onto the container as a tmpfs,
632 which allows the users to create content that disappears when
633 the container is stopped.
634 • ignore: All volumes are just ignored and no action is taken.
636 --init
640 Run an init inside the container that forwards signals and reaps pro‐
641 cesses.
642 --init-path=path
645 Path to the container-init binary.
646 --interactive, -i=true|false
649 When set to true, keep stdin open even if not attached. The default is
650 false.
651 --ip6=ip
654 Not implemented.
655 --ip=ip
658 Specify a static IP address for the container, for example
659 10.88.64.128. This option can only be used if the container is joined
660 to only a single network - i.e., --network=_network-name_ is used at
661 most once and if the container is not joining another container's net‐
662 work namespace via --network=container:_id_. The address must be
663 within the CNI network's IP address pool (default 10.88.0.0/16).
664 --ipc=mode
667 Set the IPC namespace mode for a container. The default is to create a
668 private IPC namespace.
669 • container:id: reuses another container shared memory, sema‐
672 phores and message queues
673 • host: use the host shared memory,semaphores and message queues
675 inside the container. Note: the host mode gives the container
676 full access to local shared memory and is therefore considered
677 insecure.
678 • ns:path: path to an IPC namespace to join.
680 --kernel-memory=number[unit]
684 Kernel memory limit. A unit can be b (bytes), k (kilobytes), m
685 (megabytes), or g (gigabytes).
686 Constrains the kernel memory available to a container. If a limit of 0
689 is specified (not using --kernel-memory), the container's kernel memory
690 is not limited. If you specify a limit, it may be rounded up to a mul‐
691 tiple of the operating system's page size and the value can be very
692 large, millions of trillions.
693 --label, -l=key=value
696 Add metadata to a container.
697 --label-file=file
700 Read in a line-delimited file of labels.
701 --link-local-ip=ip
704 Not implemented.
705 --log-driver="driver"
708 Logging driver for the container. Currently available options are
709 k8s-file, journald, and none, with json-file aliased to k8s-file for
710 scripting compatibility.
711 --log-opt=name=value
714 Logging driver specific options.
715 Set custom logging configuration. The following *name*s are supported:
718 path: specify a path to the log file
721 (e.g. --log-opt path=/var/log/container/mycontainer.json);
722 max-size: specify a max size of the log file
725 (e.g. --log-opt max-size=10mb);
726 tag: specify a custom log tag for the container
729 (e.g. --log-opt tag="{{.ImageName}}".
730 This option is currently supported only by the journald log driver.
733 --mac-address=address
736 Container MAC address (e.g. 92:d0:c6:0a:29:33).
737 Remember that the MAC address in an Ethernet network must be unique.
740 The IPv6 link-local address will be based on the device's MAC address
741 according to RFC4862.
742 --memory, -m=number[unit]
745 Memory limit. A unit can be b (bytes), k (kilobytes), m (megabytes), or
746 g (gigabytes).
747 Allows you to constrain the memory available to a container. If the
750 host supports swap memory, then the -m memory setting can be larger
751 than physical RAM. If a limit of 0 is specified (not using -m), the
752 container's memory is not limited. The actual limit may be rounded up
753 to a multiple of the operating system's page size (the value would be
754 very large, that's millions of trillions).
755 --memory-reservation=number[unit]
758 Memory soft limit. A unit can be b (bytes), k (kilobytes), m
759 (megabytes), or g (gigabytes).
760 After setting memory reservation, when the system detects memory con‐
763 tention or low memory, containers are forced to restrict their consump‐
764 tion to their reservation. So you should always set the value below
765 --memory, otherwise the hard limit will take precedence. By default,
766 memory reservation will be the same as memory limit.
767 --memory-swap=number[unit]
770 A limit value equal to memory plus swap. A unit can be b (bytes), k
771 (kilobytes), m (megabytes), or g (gigabytes).
772 Must be used with the -m (--memory) flag. The argument value should
775 always be larger than that of
776 -m (--memory) By default, it is set to double the value of --memory.
777 Set number to -1 to enable unlimited swap.
780 --memory-swappiness=number
783 Tune a container's memory swappiness behavior. Accepts an integer be‐
784 tween 0 and 100.
785 --mount=type=TYPE,TYPE-SPECIFIC-OPTION[,...]
788 Attach a filesystem mount to the container
789 Current supported mount TYPEs are bind, volume, image, tmpfs and de‐
792 vpts. [1] ⟨#Footnote1⟩
793 e.g.
796 type=bind,source=/path/on/host,destination=/path/in/container
798 type=bind,src=/path/on/host,dst=/path/in/container,relabel=shared
800 type=volume,source=vol1,destination=/path/in/container,ro=true
802 type=tmpfs,tmpfs-size=512M,destination=/path/in/container
804 type=image,source=fedora,destination=/fedora-image,rw=true
806 type=devpts,destination=/dev/pts
808 Common Options:
810 · src, source: mount source spec for bind and volume. Mandatory for bind.
812 · dst, destination, target: mount destination spec.
814 Options specific to volume:
816 · ro, readonly: true or false (default).
818 Options specific to image:
820 · rw, readwrite: true or false (default).
822 Options specific to bind:
824 · ro, readonly: true or false (default).
826 · bind-propagation: shared, slave, private, unbindable, rshared, rslave, runbindable, or rprivate(default). See also mount(2).
828 . bind-nonrecursive: do not setup a recursive bind mount. By default it is recursive.
830 . relabel: shared, private.
832 Options specific to tmpfs:
834 · ro, readonly: true or false (default).
836 · tmpfs-size: Size of the tmpfs mount in bytes. Unlimited by default in Linux.
838 · tmpfs-mode: File mode of the tmpfs in octal. (e.g. 700 or 0700.) Defaults to 1777 in Linux.
840 · tmpcopyup: Enable copyup from the image directory at the same location to the tmpfs. Used by default.
842 · notmpcopyup: Disable copying files from the image to the tmpfs.
844 --name=name
848 Assign a name to the container.
849 The operator can identify a container in three ways:
852 • UUID long identifier
855 (“f78375b1c487e03c9438c729345e54db9d20cfa2ac1fc3494b6eb60872e74778”);
856 • UUID short identifier (“f78375b1c487”);
858 • Name (“jonah”).
860 Podman generates a UUID for each container, and if a name is not as‐
864 signed to the container with --name then it will generate a random
865 string name. The name is useful any place you need to identify a con‐
866 tainer. This works for both background and foreground containers.
867 --network=mode, --net
870 Set the network mode for the container. Invalid if using --dns,
871 --dns-opt, or --dns-search with --network that is set to none or con‐
872 tainer:id. If used together with --pod, the container will not join the
873 pods network namespace.
874 Valid mode values are:
877 • bridge: create a network stack on the default bridge;
880 • none: no networking;
882 • container:id: reuse another container's network stack;
884 • host: use the Podman host network stack. Note: the host mode
886 gives the container full access to local system services such
887 as D-bus and is therefore considered insecure;
888 • network-id: connect to a user-defined network, multiple net‐
890 works should be comma-separated;
891 • ns:path: path to a network namespace to join;
893 • private: create a new namespace for the container (default)
895 • slirp4netns[:OPTIONS,...]: use slirp4netns(1) to create a user
897 network stack. This is the default for rootless containers. It
898 is possible to specify these additional options:
899 • allow_host_loopback=true|false: Allow the slirp4netns to
901 reach the host loopback IP (10.0.2.2). Default is false.
902 • mtu=MTU: Specify the MTU to use for this network. (Default
904 is 65520).
905 • cidr=CIDR: Specify ip range to use for this network. (De‐
907 fault is 10.0.2.0/24).
908 • enable_ipv6=true|false: Enable IPv6. Default is false. (Re‐
910 quired for outbound_addr6).
911 • outbound_addr=INTERFACE: Specify the outbound interface
913 slirp should bind to (ipv4 traffic only).
914 • outbound_addr=IPv4: Specify the outbound ipv4 address slirp
916 should bind to.
917 • outbound_addr6=INTERFACE: Specify the outbound interface
919 slirp should bind to (ipv6 traffic only).
920 • outbound_addr6=IPv6: Specify the outbound ipv6 address slirp
922 should bind to.
923 • port_handler=rootlesskit: Use rootlesskit for port forward‐
925 ing. Default.
926 • port_handler=slirp4netns: Use the slirp4netns port forward‐
928 ing.
929 --network-alias=alias
935 Add network-scoped alias for the container
936 --no-healthcheck=true|false
939 Disable any defined healthchecks for container.
940 --no-hosts=true|false
943 Do not create /etc/hosts for the container.
944 By default, Podman will manage /etc/hosts, adding the container's own
947 IP address and any hosts from --add-host.
948 --no-hosts disables this, and the image's /etc/hosts will be preserved un‐
951 modified.
952 This option conflicts with --add-host.
953 --oom-kill-disable=true|false
956 Whether to disable OOM Killer for the container or not.
957 --oom-score-adj=num
960 Tune the host's OOM preferences for containers (accepts values from
961 -1000 to 1000).
962 --os=OS
965 Override the OS, defaults to hosts, of the image to be pulled. For ex‐
966 ample, windows.
967 --pid=mode
970 Set the PID namespace mode for the container. The default is to create
971 a private PID namespace for the container.
972 • container:id: join another container's PID namespace;
975 • host: use the host's PID namespace for the container. Note the
977 host mode gives the container full access to local PID and is
978 therefore considered insecure;
979 • private: create a new namespace for the container (default)
981 • ns:path: join the specified PID namespace.
983 --pids-limit=limit
987 Tune the container's pids limit. Set to 0 to have unlimited pids for
988 the container. The default is 4096 on systems that support "pids"
989 cgroup controller.
990 --platform=OS/ARCH
993 Specify the platform for selecting the image. (Conflicts with --arch
994 and --os) The --platform option can be used to override the current ar‐
995 chitecture and operating system.
996 --pod=name
999 Run container in an existing pod. If you want Podman to make the pod
1000 for you, prefix the pod name with new:. To make a pod with more granu‐
1001 lar options, use the podman pod create command before creating a con‐
1002 tainer. If a container is run with a pod, and the pod has an in‐
1003 fra-container, the infra-container will be started before the container
1004 is.
1005 --pod-id-file=path
1008 Run container in an existing pod and read the pod's ID from the speci‐
1009 fied file. If a container is run within a pod, and the pod has an in‐
1010 fra-container, the infra-container will be started before the container
1011 is.
1012 --preserve-fds=N
1015 Pass down to the process N additional file descriptors (in addition to
1016 0, 1, 2). The total FDs will be 3+N. (This option is not available
1017 with the remote Podman client)
1018 --privileged=true|false
1021 Give extended privileges to this container. The default is false.
1022 By default, Podman containers are unprivileged (=false) and cannot, for
1025 example, modify parts of the operating system. This is because by de‐
1026 fault a container is only allowed limited access to devices. A "privi‐
1027 leged" container is given the same access to devices as the user
1028 launching the container.
1029 A privileged container turns off the security features that isolate the
1032 container from the host. Dropped Capabilities, limited devices,
1033 read-only mount points, Apparmor/SELinux separation, and Seccomp fil‐
1034 ters are all disabled.
1035 Rootless containers cannot have more privileges than the account that
1038 launched them.
1039 --publish, -p=ip:hostPort:containerPort | ip::containerPort | hostPort:con‐
1042 tainerPort | containerPort
1043 Publish a container's port, or range of ports, to the host.
1044 Both hostPort and containerPort can be specified as a range of ports.
1047 When specifying ranges for both, the number of container ports in the
1050 range must match the number of host ports in the range.
1051 If host IP is set to 0.0.0.0 or not set at all, the port will be bound
1054 on all IPs on the host.
1055 Host port does not have to be specified (e.g. podman run -p
1058 127.0.0.1::80). If it is not, the container port will be randomly as‐
1059 signed a port on the host.
1060 Use podman port to see the actual mapping: podman port $CONTAINER $CON‐
1063 TAINERPORT.
1064 Note: if a container will be run within a pod, it is not necessary to
1067 publish the port for the containers in the pod. The port must only be
1068 published by the pod itself. Pod network stacks act like the network
1069 stack on the host - you have a variety of containers in the pod, and
1070 programs in the container, all sharing a single interface and IP ad‐
1071 dress, and associated ports. If one container binds to a port, no other
1072 container can use that port within the pod while it is in use. Contain‐
1073 ers in the pod can also communicate over localhost by having one con‐
1074 tainer bind to localhost in the pod, and another connect to that port.
1075 --publish-all, -P=true|false
1078 Publish all exposed ports to random ports on the host interfaces. The
1079 default is false.
1080 When set to true, publish all exposed ports to the host interfaces. The
1083 default is false. If the operator uses -P (or -p) then Podman will make
1084 the exposed port accessible on the host and the ports will be available
1085 to any client that can reach the host.
1086 When using this option, Podman will bind any exposed port to a random
1089 port on the host within an ephemeral port range defined by
1090 /proc/sys/net/ipv4/ip_local_port_range. To find the mapping between
1091 the host ports and the exposed ports, use podman port.
1092 --pull=always|missing|never
1095 Pull image before running. The default is missing.
1096 • missing: attempt to pull the latest image from the registries
1099 listed in registries.conf if a local image does not ex‐
1100 ist.Raise an error if the image is not in any listed registry
1101 and is not present locally.
1102 • always: Pull the image from the first registry it is found in
1104 as listed in registries.conf. Raise an error if not found in
1105 the registries, even if the image is present locally.
1106 • never: do not pull the image from the registry, use only the
1108 local version. Raise an error if the image is not present lo‐
1109 cally.
1110 --quiet, -q
1114 Suppress output information when pulling images
1115 --read-only=true|false
1118 Mount the container's root filesystem as read only.
1119 By default a container will have its root filesystem writable allowing
1122 processes to write files anywhere. By specifying the --read-only flag,
1123 the container will have its root filesystem mounted as read only pro‐
1124 hibiting any writes.
1125 --read-only-tmpfs=true|false
1128 If container is running in --read-only mode, then mount a read-write
1129 tmpfs on /run, /tmp, and /var/tmp. The default is true.
1130 --replace=true|false
1133 If another container with the same name already exists, replace and re‐
1134 move it. The default is false.
1135 --requires=container
1138 Specify one or more requirements. A requirement is a dependency con‐
1139 tainer that will be started before this container. Containers can be
1140 specified by name or ID, with multiple containers being separated by
1141 commas.
1142 --restart=policy
1145 Restart policy to follow when containers exit. Restart policy will not
1146 take effect if a container is stopped via the podman kill or podman
1147 stop commands.
1148 Valid policy values are:
1151 • no : Do not restart containers on exit
1154 • on-failure[:max_retries] : Restart containers when they exit
1156 with a non-zero exit code, retrying indefinitely or until the
1157 optional max_retries count is hit
1158 • always : Restart containers when they exit,
1160 regardless of status, retrying indefinitely
1161 • unless-stopped : Identical to always
1163 Please note that restart will not restart containers after a system re‐
1167 boot. If this functionality is required in your environment, you can
1168 invoke Podman from a systemd.unit(5) file, or create an init script for
1169 whichever init system is in use. To generate systemd unit files,
1170 please see podman generate systemd.
1171 --rm=true|false
1174 Automatically remove the container when it exits. The default is false.
1175 --rmi=true|false
1178 After exit of the container, remove the image unless another container
1179 is using it. The default is false.
1180 --rootfs
1183 If specified, the first argument refers to an exploded container on the
1184 file system.
1185 This is useful to run a container without requiring any image manage‐
1188 ment, the rootfs of the container is assumed to be managed externally.
1189 Note: On SELinux systems, the rootfs needs the correct label, which is
1192 by default unconfined_u:object_r:container_file_t.
1193 --sdnotify=container|conmon|ignore
1196 Determines how to use the NOTIFY_SOCKET, as passed with systemd and
1197 Type=notify.
1198 Default is container, which means allow the OCI runtime to proxy the
1201 socket into the container to receive ready notification. Podman will
1202 set the MAINPID to conmon's pid. The conmon option sets MAINPID to
1203 conmon's pid, and sends READY when the container has started. The
1204 socket is never passed to the runtime or the container. The ignore op‐
1205 tion removes NOTIFY_SOCKET from the environment for itself and child
1206 processes, for the case where some other process above Podman uses NO‐
1207 TIFY_SOCKET and Podman should not use it.
1208 --seccomp-policy=policy
1211 Specify the policy to select the seccomp profile. If set to image, Pod‐
1212 man will look for a "io.containers.seccomp.profile" label in the con‐
1213 tainer-image config and use its value as a seccomp profile. Otherwise,
1214 Podman will follow the default policy by applying the default profile
1215 unless specified otherwise via --security-opt seccomp as described be‐
1216 low.
1217 Note that this feature is experimental and may change in the future.
1220 --secret=secret[,opt=opt ...]
1223 Give the container access to a secret. Can be specified multiple times.
1224 A secret is a blob of sensitive data which a container needs at runtime
1227 but should not be stored in the image or in source control, such as
1228 usernames and passwords, TLS certificates and keys, SSH keys or other
1229 important generic strings or binary content (up to 500 kb in size).
1230 When secrets are specified as type mount, the secrets are copied and
1233 mounted into the container when a container is created. When secrets
1234 are specified as type env, the secret will be set as an environment
1235 variable within the container. Secrets are written in the container at
1236 the time of container creation, and modifying the secret using podman
1237 secret commands after the container is created will not affect the se‐
1238 cret inside the container.
1239 Secrets and its storage are managed using the podman secret command.
1242 Secret Options
1245 • type=mount|env : How the secret will be exposed to the con‐
1248 tainer. Default mount.
1249 • target=target : Target of secret. Defauts to secret name.
1251 --security-opt=option
1255 Security Options
1256 • apparmor=unconfined : Turn off apparmor confinement for the
1259 container
1260 • apparmor=your-profile : Set the apparmor confinement profile
1262 for the container
1263 • label=user:USER: Set the label user for the container pro‐
1265 cesses
1266 • label=role:ROLE: Set the label role for the container pro‐
1268 cesses
1269 • label=type:TYPE: Set the label process type for the container
1271 processes
1272 • label=level:LEVEL: Set the label level for the container pro‐
1274 cesses
1275 • label=filetype:TYPE_: Set the label file type for the con‐
1277 tainer files
1278 • label=disable: Turn off label separation for the container
1280 Note: Labeling can be disabled for all containers by setting la‐
1284 bel=false in the containers.conf (/etc/containers/containers.conf or
1285 $HOME/.config/containers/containers.conf) file.
1286 • mask=/path/1:/path/2: The paths to mask separated by a colon.
1289 A masked path cannot be accessed inside the container.
1290 • no-new-privileges: Disable container processes from gaining
1292 additional privileges
1293 • seccomp=unconfined: Turn off seccomp confinement for the con‐
1295 tainer
1296 • seccomp=profile.json: Allowed syscall list seccomp JSON file
1298 to be used as a seccomp filter
1299 • proc-opts=OPTIONS : Comma-separated list of options to use for
1301 the /proc mount. More details for the possible mount options
1302 are specified in the proc(5) man page.
1303 • unmask=ALL or /path/1:/path/2, or shell expanded paths
1305 (/proc/*): Paths to unmask separated by a colon. If set to
1306 ALL, it will unmask all the paths that are masked or made read
1307 only by default. The default masked paths are /proc/acpi,
1308 /proc/kcore, /proc/keys, /proc/latency_stats, /proc/sched_de‐
1309 bug, /proc/scsi, /proc/timer_list, /proc/timer_stats,
1310 /sys/firmware, and /sys/fs/selinux.. The default paths that
1311 are read only are /proc/asound, /proc/bus, /proc/fs,
1312 /proc/irq, /proc/sys, /proc/sysrq-trigger, /sys/fs/cgroup.
1313 Note: Labeling can be disabled for all containers by setting la‐
1317 bel=false in the containers.conf(5) file.
1318 --shm-size=number[unit]
1321 Size of /dev/shm. A unit can be b (bytes), k (kilobytes), m
1322 (megabytes), or g (gigabytes). If you omit the unit, the system uses
1323 bytes. If you omit the size entirely, the default is 64m. When size is
1324 0, there is no limit on the amount of memory used for IPC by the con‐
1325 tainer.
1326 --sig-proxy=true|false
1329 Sets whether the signals sent to the podman run command are proxied to
1330 the container process. SIGCHLD, SIGSTOP, and SIGKILL are not proxied.
1331 The default is true.
1332 --stop-signal=signal
1335 Signal to stop a container. Default is SIGTERM.
1336 --stop-timeout=seconds
1339 Timeout to stop a container. Default is 10. Remote connections use lo‐
1340 cal containers.conf for defaults
1341 --subgidname=name
1344 Run the container in a new user namespace using the map with name in
1345 the /etc/subgid file. If calling podman run as an unprivileged user,
1346 the user needs to have the right to use the mapping. See subgid(5).
1347 This flag conflicts with --userns and --gidmap.
1348 --subuidname=name
1351 Run the container in a new user namespace using the map with name in
1352 the /etc/subuid file. If calling podman run as an unprivileged user,
1353 the user needs to have the right to use the mapping. See subuid(5).
1354 This flag conflicts with --userns and --uidmap.
1355 --sysctl=name=value
1358 Configure namespaced kernel parameters at runtime.
1359 For the IPC namespace, the following sysctls are allowed:
1362 • kernel.msgmax
1365 • kernel.msgmnb
1367 • kernel.msgmni
1369 • kernel.sem
1371 • kernel.shmall
1373 • kernel.shmmax
1375 • kernel.shmmni
1377 • kernel.shm_rmid_forced
1379 • Sysctls beginning with fs.mqueue.*
1381 Note: if you use the --ipc=host option, the above sysctls will not be
1385 allowed.
1386 For the network namespace, the following sysctls are allowed:
1389 • Sysctls beginning with net.*
1392 Note: if you use the --network=host option, these sysctls will not be
1396 allowed.
1397 --systemd=true|false|always
1400 Run container in systemd mode. The default is true.
1401 The value always enforces the systemd mode is enforced without looking
1404 at the executable name. Otherwise, if set to true and the command you
1405 are running inside the container is systemd, /usr/sbin/init, /sbin/init
1406 or /usr/local/sbin/init.
1407 If the command you are running inside of the container is systemd Pod‐
1410 man will setup tmpfs mount points in the following directories:
1411 • /run
1414 • /run/lock
1416 • /tmp
1418 • /sys/fs/cgroup/systemd
1420 • /var/lib/journal
1422 It will also set the default stop signal to SIGRTMIN+3.
1426 This allows systemd to run in a confined container without any modifi‐
1429 cations.
1430 Note that on SELinux systems, systemd attempts to write to the cgroup
1433 file system. Containers writing to the cgroup file system are denied by
1434 default. The container_manage_cgroup boolean must be enabled for this
1435 to be allowed on an SELinux separated system.
1436 setsebool -P container_manage_cgroup true
1439 --timeout=seconds
1443 Maximum time a container is allowed to run before conmon sends it the
1444 kill signal. By default containers will run until they exit or are
1445 stopped by podman stop.
1446 --tmpfs=fs
1449 Create a tmpfs mount.
1450 Mount a temporary filesystem (tmpfs) mount into a container, for exam‐
1453 ple:
1454 $ podman run -d --tmpfs /tmp:rw,size=787448k,mode=1777 my_image
1457 This command mounts a tmpfs at /tmp within the container. The supported
1461 mount options are the same as the Linux default mount flags. If you do
1462 not specify any options, the systems uses the following options:
1463 rw,noexec,nosuid,nodev.
1464 --tty, -t=true|false
1467 Allocate a pseudo-TTY. The default is false.
1468 When set to true, Podman will allocate a pseudo-tty and attach to the
1471 standard input of the container. This can be used, for example, to run
1472 a throwaway interactive shell. The default is false.
1473 NOTE: The --tty flag prevents redirection of standard output. It com‐
1476 bines STDOUT and STDERR, it can insert control characters, and it can
1477 hang pipes. This option should only be used when run interactively in a
1478 terminal. When feeding input to Podman, use -i only, not -it.
1479 echo "asdf" | podman run --rm -i someimage /bin/cat
1482 --tz=timezone
1486 Set timezone in container. This flag takes area-based timezones, GMT
1487 time, as well as local, which sets the timezone in the container to
1488 match the host machine. See /usr/share/zoneinfo/ for valid timezones.
1489 Remote connections use local containers.conf for defaults
1490 --umask=umask
1493 Set the umask inside the container. Defaults to 0022. Remote connec‐
1494 tions use local containers.conf for defaults
1495 --uidmap=container_uid:from_uid:amount
1498 Run the container in a new user namespace using the supplied mapping.
1499 This option conflicts with the --userns and --subuidname options. This
1500 option provides a way to map host UIDs to container UIDs. It can be
1501 passed several times to map different ranges.
1502 The _fromuid value is based upon the user running the command, either
1505 rootfull or rootless users. * rootfull user: con‐
1506 tainer_uid:host_uid:amount * rootless user: container_uid:intermedi‐
1507 ate_uid:amount
1508 When podman run is called by a privileged user, the option --uidmap
1511 works as a direct mapping between host UIDs and container UIDs.
1512 host UID -> container UID
1515 The amount specifies the number of consecutive UIDs that will be
1518 mapped. If for example amount is 4 the mapping would look like:
1519 | host UID | container UID | | - | -
1522 | | _fromuid | _containeruid | | _fromuid + 1 | _containeruid +
1523 1 | | _fromuid + 2 | _containeruid + 2 | | _fromuid + 3 | _containeruid
1524 + 3 |
1525 When podman run is called by an unprivileged user (i.e. running root‐
1528 less), the value _fromuid is interpreted as an "intermediate UID". In
1529 the rootless case, host UIDs are not mapped directly to container UIDs.
1530 Instead the mapping happens over two mapping steps:
1531 host UID -> intermediate UID -> container UID
1534 The --uidmap option only influences the second mapping step.
1537 The first mapping step is derived by Podman from the contents of the
1540 file /etc/subuid and the UID of the user calling Podman.
1541 First mapping step:
1544 | host UID | intermediate UID |
1547 | - | - |
1548 | UID for the user starting Podman | 0 |
1549 | 1st subordinate UID for the user starting Podman | 1 |
1550 | 2nd subordinate UID for the user starting Podman | 2 |
1551 | 3rd subordinate UID for the user starting Podman | 3 |
1552 | nth subordinate UID for the user starting Podman | n |
1553 To be able to use intermediate UIDs greater than zero, the user needs
1556 to have subordinate UIDs configured in /etc/subuid. See subuid(5).
1557 The second mapping step is configured with --uidmap.
1560 If for example amount is 5 the second mapping step would look like:
1563 | intermediate UID | container UID | | - |
1566 - | | _fromuid | _containeruid | |
1567 _fromuid + 1 | _containeruid + 1 | | _fromuid + 2 | _con‐
1568 taineruid + 2 | | _fromuid + 3 | _containeruid + 3 | | _fromuid +
1569 4 | _containeruid + 4 |
1570 Even if a user does not have any subordinate UIDs in /etc/subuid,
1573 --uidmap could still be used to map the normal UID of the user to a
1574 container UID by running podman run --uidmap $container_uid:0:1 --user
1575 $container_uid ....
1576 --ulimit=option
1579 Ulimit options. You can use host to copy the current configuration from
1580 the host.
1581 --user, -u=[user | user:group | uid | uid:gid | user:gid | uid:group ]
1584 Sets the username or UID used and optionally the groupname or GID for
1585 the specified command.
1586 Without this argument, the command will run as the user specified in
1589 the container image. Unless overridden by a USER command in the Con‐
1590 tainerfile or by a value passed to this option, this user generally de‐
1591 faults to root.
1592 When a user namespace is not in use, the UID and GID used within the
1595 container and on the host will match. When user namespaces are in use,
1596 however, the UID and GID in the container may correspond to another UID
1597 and GID on the host. In rootless containers, for example, a user name‐
1598 space is always used, and root in the container will by default corre‐
1599 spond to the UID and GID of the user invoking Podman.
1600 --userns=mode
1603 Set the user namespace mode for the container. It defaults to the POD‐
1604 MAN_USERNS environment variable. An empty value ("") means user name‐
1605 spaces are disabled unless an explicit mapping is set with the --uidmap
1606 and --gidmap options.
1607 Valid mode values are:
1610 • auto[:OPTIONS,...]: automatically create a namespace. It is
1613 possible to specify these options to auto:
1614 • gidmapping=_HOST_GID:CONTAINERGID:SIZE: to force a GID map‐
1616 ping to be present in the user namespace.
1617 • size=SIZE: to specify an explicit size for the automatic
1619 user namespace. e.g. --userns=auto:size=8192. If size is not
1620 specified, auto will estimate a size for the user namespace.
1621 • uidmapping=_HOST_UID:CONTAINERUID:SIZE: to force a UID map‐
1623 ping to be present in the user namespace.
1624 • container:id: join the user namespace of the specified con‐
1628 tainer.
1629 • host: run in the user namespace of the caller. The processes
1631 running in the container will have the same privileges on the
1632 host as any other process launched by the calling user (de‐
1633 fault).
1634 • keep-id: creates a user namespace where the current rootless
1636 user's UID:GID are mapped to the same values in the container.
1637 This option is ignored for containers created by the root
1638 user.
1639 • ns:namespace: run the container in the given existing user
1641 namespace.
1642 • private: create a new namespace for the container.
1644 This option is incompatible with --gidmap, --uidmap, --subuidname and
1648 --subgidname.
1649 --uts=mode
1652 Set the UTS namespace mode for the container. The following values are
1653 supported:
1654 • host: use the host's UTS namespace inside the container.
1657 • private: create a new namespace for the container (default).
1659 • ns:[path]: run the container in the given existing UTS name‐
1661 space.
1662 • container:[container]: join the UTS namespace of the specified
1664 container.
1665 --variant=VARIANT
1669 Use VARIANT instead of the default architecture variant of the con‐
1670 tainer image. Some images can use multiple variants of the arm archi‐
1671 tectures, such as arm/v5 and arm/v7.
1672 --volume, -v[=[[SOURCE-VOLUME|HOST-DIR:]CONTAINER-DIR[:OPTIONS]]]
1675 Create a bind mount. If you specify /HOST-DIR:/CONTAINER-DIR, Podman
1676 bind mounts host-dir in the host to CONTAINER-DIR in the Podman con‐
1677 tainer. Similarly, SOURCE-VOLUME:/CONTAINER-DIR will mount the volume
1678 in the host to the container. If no such named volume exists, Podman
1679 will create one. (Note when using the remote client, the volumes will
1680 be mounted from the remote server, not necessarily the client machine.)
1681 The options is a comma-separated list and can be: [1] ⟨#Footnote1⟩
1684 • rw|ro
1687 • z|Z
1689 • [r]shared|[r]slave|[r]private[r]unbindable
1691 • [r]bind
1693 • [no]exec
1695 • [no]dev
1697 • [no]suid
1699 • [O]
1701 • [U]
1703 The CONTAINER-DIR must be an absolute path such as /src/docs. The vol‐
1707 ume will be mounted into the container at this directory.
1708 Volumes may specify a source as well, as either a directory on the host
1711 or the name of a named volume. If no source is given, the volume will
1712 be created as an anonymously named volume with a randomly generated
1713 name, and will be removed when the container is removed via the --rm
1714 flag or podman rm --volumes.
1715 If a volume source is specified, it must be a path on the host or the
1718 name of a named volume. Host paths are allowed to be absolute or rela‐
1719 tive; relative paths are resolved relative to the directory Podman is
1720 run in. If the source does not exist, Podman will return an error.
1721 Users must pre-create the source files or directories.
1722 Any source that does not begin with a . or / will be treated as the
1725 name of a named volume. If a volume with that name does not exist, it
1726 will be created. Volumes created with names are not anonymous, and
1727 they are not removed by the --rm option and the podman rm --volumes
1728 command.
1729 You can specify multiple -v options to mount one or more volumes into a
1732 container.
1733 Write Protected Volume Mounts
1736 You can add :ro or :rw option to mount a volume in read-only or
1739 read-write mode, respectively. By default, the volumes are mounted
1740 read-write.
1741 Chowning Volume Mounts
1744 By default, Podman does not change the owner and group of source volume
1747 directories mounted into containers. If a container is created in a new
1748 user namespace, the UID and GID in the container may correspond to an‐
1749 other UID and GID on the host.
1750 The :U suffix tells Podman to use the correct host UID and GID based on
1753 the UID and GID within the container, to change recursively the owner
1754 and group of the source volume.
1755 Warning use with caution since this will modify the host filesystem.
1758 Labeling Volume Mounts
1761 Labeling systems like SELinux require that proper labels are placed on
1764 volume content mounted into a container. Without a label, the security
1765 system might prevent the processes running inside the container from
1766 using the content. By default, Podman does not change the labels set by
1767 the OS.
1768 To change a label in the container context, you can add either of two
1771 suffixes :z or :Z to the volume mount. These suffixes tell Podman to
1772 relabel file objects on the shared volumes. The z option tells Podman
1773 that two containers share the volume content. As a result, Podman la‐
1774 bels the content with a shared content label. Shared volume labels al‐
1775 low all containers to read/write content. The Z option tells Podman to
1776 label the content with a private unshared label.
1777 Overlay Volume Mounts
1780 The :O flag tells Podman to mount the directory from the host as a tem‐
1783 porary storage using the overlay file system. The container processes
1784 can modify content within the mountpoint which is stored in the con‐
1785 tainer storage in a separate directory. In overlay terms, the source
1786 directory will be the lower, and the container storage directory will
1787 be the upper. Modifications to the mount point are destroyed when the
1788 container finishes executing, similar to a tmpfs mount point being un‐
1789 mounted.
1790 Subsequent executions of the container will see the original source di‐
1793 rectory content, any changes from previous container executions no
1794 longer exist.
1795 One use case of the overlay mount is sharing the package cache from the
1798 host into the container to allow speeding up builds.
1799 Note:
1802 - The `O` flag conflicts with other options listed above.
1805 Content mounted into the container is labeled with the private label.
1809 On SELinux systems, labels in the source directory must be read‐
1810 able by the container label. Usually containers can read/execute con‐
1811 tainer_share_t and can read/write container_file_t. If you cannot
1812 change the labels on a source volume, SELinux container separation must
1813 be disabled for the container to work.
1814 - The source directory mounted into the container with an overlay
1815 mount should not be modified, it can cause unexpected failures. It is
1816 recommended that you do not modify the directory until the container
1817 finishes running.
1818 Only the current container can use a private volume.
1821 Mounts propagation
1824 By default bind mounted volumes are private. That means any mounts done
1827 inside container will not be visible on host and vice versa. One can
1828 change this behavior by specifying a volume mount propagation property.
1829 Making a volume shared mounts done under that volume inside container
1830 will be visible on host and vice versa. Making a volume slave enables
1831 only one way mount propagation and that is mounts done on host under
1832 that volume will be visible inside container but not the other way
1833 around. [1] ⟨#Footnote1⟩
1834 To control mount propagation property of volume one can use [r]shared,
1837 [r]slave, [r]private or [r]unbindable propagation flag. Propagation
1838 property can be specified only for bind mounted volumes and not for in‐
1839 ternal volumes or named volumes. For mount propagation to work source
1840 mount point (mount point where source dir is mounted on) has to have
1841 right propagation properties. For shared volumes, source mount point
1842 has to be shared. And for slave volumes, source mount has to be either
1843 shared or slave. [1] ⟨#Footnote1⟩
1844 If you want to recursively mount a volume and all of its submounts into
1847 a container, then you can use the rbind option. By default the bind op‐
1848 tion is used, and submounts of the source directory will not be mounted
1849 into the container.
1850 Mounting the volume with the nosuid options means that SUID applica‐
1853 tions on the volume will not be able to change their privilege. By de‐
1854 fault volumes are mounted with nosuid.
1855 Mounting the volume with the noexec option means that no executables on
1858 the volume will be able to executed within the container.
1859 Mounting the volume with the nodev option means that no devices on the
1862 volume will be able to be used by processes within the container. By
1863 default volumes are mounted with nodev.
1864 If the host-dir is a mount point, then dev, suid, and exec options are
1867 ignored by the kernel.
1868 Use df $hostdir to figure out the source mount, and then use findmnt -o
1871 TARGET,PROPAGATION source-mount-dir to figure out propagation proper‐
1872 ties of source mount. If findmnt[4m(1) utility is not available, then one
1873 can look at mount entry for source mount point in /proc/self/mountinfo.
1874 Look at the "optional fields" and see if any propagation properties are
1875 specified. In there, shared:N means the mount is shared, master:N
1876 means mount is slave, and if nothing is there, the mount is private.
1877 [1] ⟨#Footnote1⟩
1878 To change propagation properties of a mount point, use mount(8) com‐
1881 mand. For example, if one wants to bind mount source directory /foo,
1882 one can do mount --bind /foo /foo and mount --make-private
1883 --make-shared /foo. This will convert /foo into a shared mount point.
1884 Alternatively, one can directly change propagation properties of source
1885 mount. Say, if / is source mount for /foo, then use mount --make-shared
1886 / to convert / into a shared mount.
1887 Note: if the user only has access rights via a group, accessing the
1890 volume from inside a rootless container will fail. Use the --group-add
1891 keep-groups flag to pass the user's supplementary group access into the
1892 container.
1893 --volumes-from[=CONTAINER[:OPTIONS]]
1896 Mount volumes from the specified container(s). Used to share volumes
1897 between containers. The options is a comma-separated list with the fol‐
1898 lowing available elements:
1899 • rw|ro
1902 • z
1904 Mounts already mounted volumes from a source container onto another
1908 container. You must supply the source's container-id or container-name.
1909 To share a volume, use the --volumes-from option when running the tar‐
1910 get container. You can share volumes even if the source container is
1911 not running.
1912 By default, Podman mounts the volumes in the same mode (read-write or
1915 read-only) as it is mounted in the source container. You can change
1916 this by adding a ro or rw option.
1917 Labeling systems like SELinux require that proper labels are placed on
1920 volume content mounted into a container. Without a label, the security
1921 system might prevent the processes running inside the container from
1922 using the content. By default, Podman does not change the labels set by
1923 the OS.
1924 To change a label in the container context, you can add z to the volume
1927 mount. This suffix tells Podman to relabel file objects on the shared
1928 volumes. The z option tells Podman that two containers share the volume
1929 content. As a result, Podman labels the content with a shared content
1930 label. Shared volume labels allow all containers to read/write content.
1931 If the location of the volume from the source container overlaps with
1934 data residing on a target container, then the volume hides that data on
1935 the target.
1936 --workdir, -w=dir
1939 Working directory inside the container.
1940 The default working directory for running binaries within a container
1943 is the root directory (/). The image developer can set a different de‐
1944 fault with the WORKDIR instruction. The operator can override the work‐
1945 ing directory by using the -w option.
1946 --pidfile=path
1949 When the pidfile location is specified, the container process' PID will
1950 be written to the pidfile. (This option is not available with the re‐
1951 mote Podman client) If the pidfile option is not specified, the con‐
1952 tainer process' PID will be written to /run/containers/storage/${stor‐
1953 age-driver}-containers/$CID/userdata/pidfile.
1954 After the container is started, the location for the pidfile can be
1957 discovered with the following podman inspect command:
1958 $ podman inspect --format '{{ .PidFile }}' $CID
1961 /run/containers/storage/${storage-driver}-containers/$CID/userdata/pidfile
1962
1966 The exit code from podman run gives information about why the container
1967 failed to run or why it exited. When podman run exits with a non-zero
1968 code, the exit codes follow the chroot(1) standard, see below:
1969 125 The error is with Podman itself
1972 $ podman run --foo busybox; echo $?
1975 Error: unknown flag: --foo
1976 125
1977 126 The contained command cannot be invoked
1981 $ podman run busybox /etc; echo $?
1984 Error: container_linux.go:346: starting container process caused "exec: \"/etc\": permission denied": OCI runtime error
1985 126
1986 127 The contained command cannot be found
1990 $ podman run busybox foo; echo $?
1993 Error: container_linux.go:346: starting container process caused "exec: \"foo\": executable file not found in $PATH": OCI runtime error
1994 127
1995 Exit code contained command exit code
1999 $ podman run busybox /bin/sh -c 'exit 3'; echo $?
2002 3
2003
2007 Running container in read-only mode
2008 During container image development, containers often need to write to
2009 the image content. Installing packages into /usr, for example. In pro‐
2010 duction, applications seldom need to write to the image. Container ap‐
2011 plications write to volumes if they need to write to file systems at
2012 all. Applications can be made more secure by running them in read-only
2013 mode using the --read-only switch. This protects the containers image
2014 from modification. Read only containers may still need to write tempo‐
2015 rary data. The best way to handle this is to mount tmpfs directories on
2016 /run and /tmp.
2017 $ podman run --read-only -i -t fedora /bin/bash
2020 $ podman run --read-only --read-only-tmpfs=false --tmpfs /run -i -t fedora /bin/bash
2022 Exposing log messages from the container to the host's log
2026 If you want messages that are logged in your container to show up in
2027 the host's syslog/journal then you should bind mount the /dev/log di‐
2028 rectory as follows.
2029 $ podman run -v /dev/log:/dev/log -i -t fedora /bin/bash
2032 From inside the container you can test this by sending a message to the
2036 log.
2037 (bash)# logger "Hello from my container"
2040 Then exit and check the journal.
2044 (bash)# exit
2047 $ journalctl -b | grep Hello
2049 This should list the message sent to logger.
2053 Attaching to one or more from STDIN, STDOUT, STDERR
2056 If you do not specify -a, Podman will attach everything (stdin, stdout,
2057 stderr). You can specify to which of the three standard streams
2058 (stdin, stdout, stderr) you'd like to connect instead, as in:
2059 $ podman run -a stdin -a stdout -i -t fedora /bin/bash
2062 Sharing IPC between containers
2066 Using shm_server.c available here:
2067 https://www.cs.cf.ac.uk/Dave/C/node27.html
2068 Testing --ipc=host mode:
2071 Host shows a shared memory segment with 7 pids attached, happens to be
2074 from httpd:
2075 $ sudo ipcs -m
2078 ------ Shared Memory Segments --------
2080 key shmid owner perms bytes nattch status
2081 0x01128e25 0 root 600 1000 7
2082 Now run a regular container, and it correctly does NOT see the shared
2086 memory segment from the host:
2087 $ podman run -it shm ipcs -m
2090 ------ Shared Memory Segments --------
2092 key shmid owner perms bytes nattch status
2093 Run a container with the new --ipc=host option, and it now sees the
2097 shared memory segment from the host httpd:
2098 $ podman run -it --ipc=host shm ipcs -m
2101 ------ Shared Memory Segments --------
2103 key shmid owner perms bytes nattch status
2104 0x01128e25 0 root 600 1000 7
2105 Testing --ipc=container:id mode:
2109 Start a container with a program to create a shared memory segment:
2112 $ podman run -it shm bash
2115 $ sudo shm/shm_server
2116 $ sudo ipcs -m
2117 ------ Shared Memory Segments --------
2119 key shmid owner perms bytes nattch status
2120 0x0000162e 0 root 666 27 1
2121 Create a 2nd container correctly shows no shared memory segment from
2125 1st container:
2126 $ podman run shm ipcs -m
2129 ------ Shared Memory Segments --------
2131 key shmid owner perms bytes nattch status
2132 Create a 3rd container using the --ipc=container:id option, now it
2136 shows the shared memory segment from the first:
2137 $ podman run -it --ipc=container:ed735b2264ac shm ipcs -m
2140 $ sudo ipcs -m
2141 ------ Shared Memory Segments --------
2143 key shmid owner perms bytes nattch status
2144 0x0000162e 0 root 666 27 1
2145 Mapping Ports for External Usage
2149 The exposed port of an application can be mapped to a host port using
2150 the -p flag. For example, an httpd port 80 can be mapped to the host
2151 port 8080 using the following:
2152 $ podman run -p 8080:80 -d -i -t fedora/httpd
2155 Mounting External Volumes
2159 To mount a host directory as a container volume, specify the absolute
2160 path to the directory and the absolute path for the container directory
2161 separated by a colon. If the source is a named volume maintained by
2162 Podman, it is recommended to use its name rather than the path to the
2163 volume. Otherwise the volume will be considered as an orphan and wiped
2164 if you execute podman volume prune:
2165 $ podman run -v /var/db:/data1 -i -t fedora bash
2168 $ podman run -v data:/data2 -i -t fedora bash
2170 $ podman run -v /var/cache/dnf:/var/cache/dnf:O -ti fedora dnf -y update
2172 $ podman run -d -e MYSQL_ROOT_PASSWORD=root --user mysql --userns=keep-id -v ~/data:/var/lib/mysql:z,U mariadb
2174 Using --mount flags to mount a host directory as a container folder,
2178 specify the absolute path to the directory or the volume name, and the
2179 absolute path within the container directory:
2180 $ podman run --mount type=bind,src=/var/db,target=/data1 busybox sh
2183 $ podman run --mount type=bind,src=volume-name,target=/data1 busybox sh
2185 When using SELinux, be aware that the host has no knowledge of con‐
2189 tainer SELinux policy. Therefore, in the above example, if SELinux pol‐
2190 icy is enforced, the /var/db directory is not writable to the con‐
2191 tainer. A "Permission Denied" message will occur and an avc: message in
2192 the host's syslog.
2193 To work around this, at time of writing this man page, the following
2196 command needs to be run in order for the proper SELinux policy type la‐
2197 bel to be attached to the host directory:
2198 $ chcon -Rt svirt_sandbox_file_t /var/db
2201 Now, writing to the /data1 volume in the container will be allowed and
2205 the changes will also be reflected on the host in /var/db.
2206 Using alternative security labeling
2209 You can override the default labeling scheme for each container by
2210 specifying the --security-opt flag. For example, you can specify the
2211 MCS/MLS level, a requirement for MLS systems. Specifying the level in
2212 the following command allows you to share the same content between con‐
2213 tainers.
2214 podman run --security-opt label=level:s0:c100,c200 -i -t fedora bash
2217 An MLS example might be:
2221 $ podman run --security-opt label=level:TopSecret -i -t rhel7 bash
2224 To disable the security labeling for this container versus running with
2228 the
2229 --permissive flag, use the following command:
2232 $ podman run --security-opt label=disable -i -t fedora bash
2233 If you want a tighter security policy on the processes within a con‐
2237 tainer, you can specify an alternate type for the container. You could
2238 run a container that is only allowed to listen on Apache ports by exe‐
2239 cuting the following command:
2240 $ podman run --security-opt label=type:svirt_apache_t -i -t centos bash
2243 Note you would have to write policy defining a svirt_apache_t type.
2247 To mask additional specific paths in the container, specify the paths
2250 separated by a colon using the mask option with the --security-opt
2251 flag.
2252 $ podman run --security-opt mask=/foo/bar:/second/path fedora bash
2255 To unmask all the paths that are masked by default, set the unmask op‐
2259 tion to ALL. Or to only unmask specific paths, specify the paths as
2260 shown above with the mask option.
2261 $ podman run --security-opt unmask=ALL fedora bash
2264 To unmask all the paths that start with /proc, set the unmask option to
2268 /proc/*.
2269 $ podman run --security-opt unmask=/proc/* fedora bash
2272 $ podman run --security-opt unmask=/foo/bar:/sys/firmware fedora bash
2276 Setting device weight
2280 If you want to set /dev/sda device weight to 200, you can specify the
2281 device weight by --blkio-weight-device flag. Use the following command:
2282 $ podman run -it --blkio-weight-device "/dev/sda:200" ubuntu
2285 Using a podman container with input from a pipe
2289 $ echo "asdf" | podman run --rm -i --entrypoint /bin/cat someimage
2290 asdf
2291 Setting Namespaced Kernel Parameters (Sysctls)
2295 The --sysctl sets namespaced kernel parameters (sysctls) in the con‐
2296 tainer. For example, to turn on IP forwarding in the containers network
2297 namespace, run this command:
2298 $ podman run --sysctl net.ipv4.ip_forward=1 someimage
2301 Note that not all sysctls are namespaced. Podman does not support
2305 changing sysctls inside of a container that also modify the host sys‐
2306 tem. As the kernel evolves we expect to see more sysctls become names‐
2307 paced.
2308 See the definition of the --sysctl option above for the current list of
2311 supported sysctls.
2312 Set UID/GID mapping in a new user namespace
2315 Running a container in a new user namespace requires a mapping of the
2316 uids and gids from the host.
2317 $ podman run --uidmap 0:30000:7000 --gidmap 0:30000:7000 fedora echo hello
2320 Configuring Storage Options from the command line
2324 Podman allows for the configuration of storage by changing the values
2325 in the /etc/container/storage.conf or by using global options. This
2326 shows how to setup and use fuse-overlayfs for a one time run of busybox
2327 using global options.
2328 podman --log-level=debug --storage-driver overlay --storage-opt "overlay.mount_program=/usr/bin/fuse-overlayfs" run busybox /bin/sh
2331 Configure timezone in a container
2335 $ podman run --tz=local alpine date
2336 $ podman run --tz=Asia/Shanghai alpine date
2337 $ podman run --tz=US/Eastern alpine date
2338 Adding dependency containers
2342 The first container, container1, is not started initially, but must be
2343 running before container2 will start. The podman run command will
2344 start the container automatically before starting container2.
2345 $ podman create --name container1 -t -i fedora bash
2348 $ podman run --name container2 --requires container1 -t -i fedora bash
2349 Multiple containers can be required.
2353 $ podman create --name container1 -t -i fedora bash
2356 $ podman create --name container2 -t -i fedora bash
2357 $ podman run --name container3 --requires container1,container2 -t -i fedora bash
2358 Configure keep supplemental groups for access to volume
2362 $ podman run -v /var/lib/design:/var/lib/design --group-add keep-groups ubi8
2363 Rootless Containers
2367 Podman runs as a non root user on most systems. This feature requires
2368 that a new enough version of shadow-utils be installed. The
2369 shadow-utils package must include the newuidmap(1) and newgidmap(1) ex‐
2370 ecutables.
2371 Note: RHEL7 and Centos 7 will not have this feature until RHEL7.7 is
2374 released.
2375 In order for users to run rootless, there must be an entry for their
2378 username in /etc/subuid and /etc/subgid which lists the UIDs for their
2379 user namespace.
2380 Rootless Podman works better if the fuse-overlayfs and slirp4netns
2383 packages are installed. The fuse-overlay package provides a userspace
2384 overlay storage driver, otherwise users need to use the vfs storage
2385 driver, which is diskspace expensive and does not perform well.
2386 slirp4netns is required for VPN, without it containers need to be run
2387 with the --network=host flag.
2388
2391 Environment variables within containers can be set using multiple dif‐
2392 ferent options, in the following order of precedence (later entries
2393 override earlier entries):
2394 • Container image: Any environment variables specified in the
2397 container image.
2398 • --http-proxy: By default, several environment variables will
2400 be passed in from the host, such as http_proxy and no_proxy.
2401 See --http-proxy for details.
2402 • --env-host: Host environment of the process executing Podman
2404 is added.
2405 • --env-file: Any environment variables specified via env-files.
2407 If multiple files specified, then they override each other in
2408 order of entry.
2409 • --env: Any environment variables specified will override pre‐
2411 vious settings.
2412 Run containers and set the environment ending with a * and a *****:
2416 $ export ENV1=a
2419 $ podman run --env ENV* alpine printenv ENV1
2420 a
2421 $ podman run --env ENV*****=b alpine printenv ENV*****
2423 b
2424
2428 When Podman starts a container it actually executes the conmon program,
2429 which then executes the OCI Runtime. Conmon is the container monitor.
2430 It is a small program whose job is to watch the primary process of the
2431 container, and if the container dies, save the exit code. It also
2432 holds open the tty of the container, so that it can be attached to
2433 later. This is what allows Podman to run in detached mode (back‐
2434 grounded), so Podman can exit but conmon continues to run. Each con‐
2435 tainer has their own instance of conmon. Conmon waits for the container
2436 to exit, gathers and saves the exit code, and then launches a Podman
2437 process to complete the container cleanup, by shutting down the network
2438 and storage. For more information on conmon, please reference the
2439 conmon(8) man page.
2440
2443 /etc/subuid
2444 /etc/subgid
2447 NOTE: Use the environment variable TMPDIR to change the temporary stor‐
2450 age location of downloaded container images. Podman defaults to use
2451 /var/tmp.
2452
2455 podman(1), podman-save(1), podman-ps(1), podman-attach(1), pod‐
2456 man-pod-create(1), podman-port(1), podman-start(1), podman-kill(1),
2457 podman-stop(1), podman-generate-systemd(1) podman-rm(1), subgid(5),
2458 subuid(5), containers.conf(5), systemd.unit(5), setsebool(8),
2459 slirp4netns(1), fuse-overlayfs(1), proc(5), conmon(8).
2460
2463 September 2018, updated by Kunal Kushwaha <kushwaha_ku‐
2464 nal_v7@lab.ntt.co.jp>
2465 October 2017, converted from Docker documentation to Podman by Dan
2468 Walsh for Podman <dwalsh@redhat.com>
2469 November 2015, updated by Sally O'Malley <somalley@redhat.com>
2472 June 2014, updated by Sven Dowideit <SvenDowideit@home.org.au>
2475 April 2014, Originally compiled by William Henry <whenry@redhat.com>
2478 based on docker.com source material and internal work.
2479
2482 1: The Podman project is committed to inclusivity, a core value of open
2483 source. The master and slave mount propagation terminology used here is
2484 problematic and divisive, and should be changed. However, these terms
2485 are currently used within the Linux kernel and must be used as-is at
2486 this time. When the kernel maintainers rectify this usage, Podman will
2487 follow suit immediately.
2488 podman-run(1)()