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
37 indicate to programs they are running in a container. This file is
38 located at /run/.containerenv. When using the --privileged flag the
39 .containerenv 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
51 remote 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
98 internal 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.
225 --cpu-period=limit
228 Set the CPU period for the Completely Fair Scheduler (CFS), which is a
229 duration in microseconds. Once the container's CPU quota is used up, it
230 will not be scheduled to run until the current period ends. Defaults to
231 100000 microseconds.
232 On some systems, changing the CPU limits may not be allowed for
235 non-root users. For more details, see https://github.com/contain‐
236 ers/podman/blob/master/troubleshooting.md#26-running-contain‐
237 ers-with-cpu-limits-fails-with-a-permissions-error
238 --cpu-quota=limit
241 Limit the CPU Completely Fair Scheduler (CFS) quota.
242 Limit the container's CPU usage. By default, containers run with the
245 full CPU resource. The limit is a number in microseconds. If you pro‐
246 vide a number, the container will be allowed to use that much CPU time
247 until the CPU period ends (controllable via --cpu-period).
248 On some systems, changing the CPU limits may not be allowed for
251 non-root users. For more details, see https://github.com/contain‐
252 ers/podman/blob/master/troubleshooting.md#26-running-contain‐
253 ers-with-cpu-limits-fails-with-a-permissions-error
254 --cpu-rt-period=microseconds
257 Limit the CPU real-time period in microseconds.
258 Limit the container's Real Time CPU usage. This flag tell the kernel to
261 restrict the container's Real Time CPU usage to the period you specify.
262 --cpu-rt-runtime=microseconds
265 Limit the CPU real-time runtime in microseconds.
266 Limit the containers Real Time CPU usage. This flag tells the kernel to
269 limit the amount of time in a given CPU period Real Time tasks may con‐
270 sume. Ex: Period of 1,000,000us and Runtime of 950,000us means that
271 this container could consume 95% of available CPU and leave the remain‐
272 ing 5% to normal priority tasks.
273 The sum of all runtimes across containers cannot exceed the amount
276 allotted to the parent cgroup.
277 --cpu-shares=shares
280 CPU shares (relative weight).
281 By default, all containers get the same proportion of CPU cycles. This
284 proportion can be modified by changing the container's CPU share
285 weighting relative to the combined weight of all the running contain‐
286 ers. Default weight is 1024.
287 The proportion will only apply when CPU-intensive processes are run‐
290 ning. When tasks in one container are idle, other containers can use
291 the left-over CPU time. The actual amount of CPU time will vary depend‐
292 ing on the number of containers running on the system.
293 For example, consider three containers, one has a cpu-share of 1024 and
296 two others have a cpu-share setting of 512. When processes in all three
297 containers attempt to use 100% of CPU, the first container would
298 receive 50% of the total CPU time. If you add a fourth container with a
299 cpu-share of 1024, the first container only gets 33% of the CPU. The
300 remaining containers receive 16.5%, 16.5% and 33% of the CPU.
301 On a multi-core system, the shares of CPU time are distributed over all
304 CPU cores. Even if a container is limited to less than 100% of CPU
305 time, it can use 100% of each individual CPU core.
306 For example, consider a system with more than three cores. If you start
309 one container {C0} with --cpu-shares=512 running one process, and
310 another container {C1} with --cpu-shares=1024 running two processes,
311 this can result in the following division of CPU shares:
312 ┌────┬───────────┬─────┬──────────────┐
315 │PID │ container │ CPU │ CPU share │
316 ├────┼───────────┼─────┼──────────────┤
317 │100 │ {C0} │ 0 │ 100% of CPU0 │
318 ├────┼───────────┼─────┼──────────────┤
319 │101 │ {C1} │ 1 │ 100% of CPU1 │
320 ├────┼───────────┼─────┼──────────────┤
321 │102 │ {C1} │ 2 │ 100% of CPU2 │
322 ├────┼───────────┼─────┼──────────────┤
323 │ │ │ │ │
324 └────┴───────────┴─────┴──────────────┘
325 --cpus=number
327 Number of CPUs. The default is 0.0 which means no limit. This is short‐
328 hand for --cpu-period and --cpu-quota, so you may only set either
329 --cpus or --cpu-period and --cpu-quota.
332 On some systems, changing the CPU limits may not be allowed for
333 non-root users. For more details, see https://github.com/contain‐
334 ers/podman/blob/master/troubleshooting.md#26-running-contain‐
335 ers-with-cpu-limits-fails-with-a-permissions-error
336 --cpuset-cpus=number
339 CPUs in which to allow execution. Can be specified as a comma-separated
340 list (e.g. 0,1), as a range (e.g. 0-3), or any combination thereof
341 (e.g. 0-3,7,11-15).
342 --cpuset-mems=nodes
345 Memory nodes (MEMs) in which to allow execution. Only effective on NUMA
346 systems.
347 For example, if you have four memory nodes (0-3) on your system, use
350 --cpuset-mems=0,1 to only use memory from the first two memory nodes.
351 --detach, -d=true|false
354 Detached mode: run the container in the background and print the new
355 container ID. The default is false.
356 At any time you can run podman ps in the other shell to view a list of
359 the running containers. You can reattach to a detached container with
360 podman attach.
361 When attached in the tty mode, you can detach from the container (and
364 leave it running) using a configurable key sequence. The default
365 sequence is ctrl-p,ctrl-q. Configure the keys sequence using the
366 --detach-keys option, or specifying it in the containers.conf file: see
367 containers.conf(5) for more information.
368 --detach-keys=sequence
371 Specify the key sequence for detaching a container. Format is a single
372 character [a-Z] or one or more ctrl-<value> characters where <value> is
373 one of: a-z, @, ^, [, , or _. Specifying "" will disable this feature.
374 The default is ctrl-p,ctrl-q.
375 This option can also be set in containers.conf(5) file.
378 Specifying "" will disable this feature. The default is ctrl-p,ctrl-q.
381 --device=host-device[:container-device][:permissions]
384 Add a host device to the container. Optional permissions parameter can
385 be used to specify device permissions, it is combination of r for read,
386 w for write, and m for mknod(2).
387 Example: --device=/dev/sdc:/dev/xvdc:rwm.
390 Note: if _hostdevice is a symbolic link then it will be resolved first.
393 The container will only store the major and minor numbers of the host
394 device.
395 Note: if the user only has access rights via a group, accessing the
398 device from inside a rootless container will fail. The crun(1) runtime
399 offers a workaround for this by adding the option --annotation
400 run.oci.keep_original_groups=1.
401 Podman may load kernel modules required for using the specified device.
404 The devices that Podman will load modules when necessary are:
405 /dev/fuse.
406 --device-cgroup-rule=rule
409 Add a rule to the cgroup allowed devices list
410 --device-read-bps=path:rate
413 Limit read rate (in bytes per second) from a device (e.g.
414 --device-read-bps=/dev/sda:1mb).
415 --device-read-iops=path:rate
418 Limit read rate (in IO operations per second) from a device (e.g.
419 --device-read-iops=/dev/sda:1000).
420 --device-write-bps=path:rate
423 Limit write rate (in bytes per second) to a device (e.g.
424 --device-write-bps=/dev/sda:1mb).
425 --device-write-iops=path:rate
428 Limit write rate (in IO operations per second) to a device (e.g.
429 --device-write-iops=/dev/sda:1000).
430 --disable-content-trust
433 This is a Docker specific option to disable image verification to a
434 Docker registry and is not supported by Podman. This flag is a NOOP and
435 provided solely for scripting compatibility.
436 --dns=ipaddr
439 Set custom DNS servers. Invalid if using --dns with --network that is
440 set to none or container:id.
441 This option can be used to override the DNS configuration passed to the
444 container. Typically this is necessary when the host DNS configuration
445 is invalid for the container (e.g., 127.0.0.1). When this is the case
446 the --dns flags is necessary for every run.
447 The special value none can be specified to disable creation of
450 /etc/resolv.conf in the container by Podman. The /etc/resolv.conf file
451 in the image will be used without changes.
452 --dns-opt=option
455 Set custom DNS options. Invalid if using --dns-opt with --network that
456 is set to none or container:id.
457 --dns-search=domain
460 Set custom DNS search domains. Invalid if using --dns-search and --net‐
461 work that is set to none or container:id. Use --dns-search=. if you
462 don't wish to set the search domain.
463 --entrypoint="command" | '["command", arg1 , ...]'
466 Overwrite the default ENTRYPOINT of the image.
467 This option allows you to overwrite the default entrypoint of the
470 image.
471 The ENTRYPOINT of an image is similar to a COMMAND because it specifies
474 what executable to run when the container starts, but it is (purposely)
475 more difficult to override. The ENTRYPOINT gives a container its
476 default nature or behavior, so that when you set an ENTRYPOINT you can
477 run the container as if it were that binary, complete with default
478 options, and you can pass in more options via the COMMAND. But, some‐
479 times an operator may want to run something else inside the container,
480 so you can override the default ENTRYPOINT at runtime by using a
481 --entrypoint and a string to specify the new ENTRYPOINT.
482 You need to specify multi option commands in the form of a json string.
485 --env, -e=env
488 Set environment variables.
489 This option allows arbitrary environment variables that are available
492 for the process to be launched inside of the container. If an environ‐
493 ment variable is specified without a value, Podman will check the host
494 environment for a value and set the variable only if it is set on the
495 host. If an environment variable ending in * is specified, Podman will
496 search the host environment for variables starting with the prefix and
497 will add those variables to the container. If an environment variable
498 with a trailing ***** is specified, then a value must be supplied.
499 See Environment ⟨#environment⟩ note below for precedence and examples.
502 --env-host=true|false
505 Use host environment inside of the container. See Environment note
506 below for precedence. (Not available for remote commands)
507 --env-file=file
510 Read in a line delimited file of environment variables. See Environment
511 note below for precedence.
512 --expose=port
515 Expose a port, or a range of ports (e.g. --expose=3300-3310) to set up
516 port redirection on the host system.
517 --gidmap=container_gid:host_gid:amount
520 Run the container in a new user namespace using the supplied mapping.
521 This option conflicts with the --userns and --subgidname flags. This
522 option can be passed several times to map different ranges. If calling
523 podman run as an unprivileged user, the user needs to have the right to
524 use the mapping. See subuid(5). The example maps gids 0-1999 in the
525 container to the gids 30000-31999 on the host: --gidmap=0:30000:2000.
526 Important note: The new user namespace mapping based on --gidmap is
529 based on the initial mapping made in the /etc/subgid file. Assuming
530 there is a /etc/subgid mapping groupname:100000:65536, then groupname
531 is initially mapped to a namespace starting with gid 100000 for 65536
532 ids. From here the --gidmap mapping to the new namespace starts from 0
533 again, but is based on the initial mapping. Meaning groupname is ini‐
534 tially mapped to gid 100000 which is referenced as 0 in the following
535 --gidmap mapping. In terms of the example above: The group groupname is
536 mapped to group 100000 of the initial namespace then the 30000st id of
537 this namespace (which is gid 130000 in this namespace) is mapped to
538 container namespace group id 0. (groupname -> 100000 / 30000 -> 0)
539 --group-add=group
542 Add additional groups to run as
543 --health-cmd="command" | '["command", arg1 , ...]'
546 Set or alter a healthcheck command for a container. The command is a
547 command to be executed inside your container that determines your con‐
548 tainer health. The command is required for other healthcheck options to
549 be applied. A value of none disables existing healthchecks.
550 Multiple options can be passed in the form of a JSON array; otherwise,
553 the command will be interpreted as an argument to /bin/sh -c.
554 --health-interval=interval
557 Set an interval for the healthchecks. An interval of disable results in
558 no automatic timer setup. The default is 30s.
559 --health-retries=retries
562 The number of retries allowed before a healthcheck is considered to be
563 unhealthy. The default value is 3.
564 --health-start-period=period
567 The initialization time needed for a container to bootstrap. The value
568 can be expressed in time format like 2m3s. The default value is 0s.
569 --health-timeout=timeout
572 The maximum time allowed to complete the healthcheck before an interval
573 is considered failed. Like start-period, the value can be expressed in
574 a time format such as 1m22s. The default value is 30s.
575 --help
578 Print usage statement
579 --hostname=name, -h
582 Container host name
583 Sets the container host name that is available inside the container.
586 Can only be used with a private UTS namespace --uts=private (default).
587 If --pod is specified and the pod shares the UTS namespace (default)
588 the pod's hostname will be used.
589 --http-proxy=true|false
592 By default proxy environment variables are passed into the container if
593 set for the Podman process. This can be disabled by setting the value
594 to false. The environment variables passed in include http_proxy,
595 https_proxy, ftp_proxy, no_proxy, and also the upper case versions of
596 those. This option is only needed when the host system must use a proxy
597 but the container should not use any proxy. Proxy environment variables
598 specified for the container in any other way will override the values
599 that would have been passed through from the host. (Other ways to spec‐
600 ify the proxy for the container include passing the values with the
601 --env flag, or hard coding the proxy environment at container build
602 time.) (Not available for remote commands)
603 Defaults to true.
606 --image-volume, builtin-volume=bind|tmpfs|ignore
609 Tells Podman how to handle the builtin image volumes. Default is bind.
610 · bind: An anonymous named volume will be created and mounted
613 into the container.
614 · tmpfs: The volume is mounted onto the container as a tmpfs,
616 which allows the users to create content that disappears when
617 the container is stopped.
618 · ignore: All volumes are just ignored and no action is taken.
620 --init
624 Run an init inside the container that forwards signals and reaps pro‐
625 cesses.
626 --init-path=path
629 Path to the container-init binary.
630 --interactive, -i=true|false
633 When set to true, keep stdin open even if not attached. The default is
634 false.
635 --ip6=ip
638 Not implemented.
639 --ip=ip
642 Specify a static IP address for the container, for example
643 10.88.64.128. This option can only be used if the container is joined
644 to only a single network - i.e., --network=_network-name_ is used at
645 most once and if the container is not joining another container's net‐
646 work namespace via --network=container:_id_. The address must be
647 within the CNI network's IP address pool (default 10.88.0.0/16).
648 --ipc=mode
651 Set the IPC namespace mode for a container. The default is to create a
652 private IPC namespace.
653 · container:id: reuses another container shared memory, sema‐
656 phores and message queues
657 · host: use the host shared memory,semaphores and message queues
659 inside the container. Note: the host mode gives the container
660 full access to local shared memory and is therefore considered
661 insecure.
662 · ns:path: path to an IPC namespace to join.
664 --kernel-memory=number[unit]
668 Kernel memory limit. A unit can be b (bytes), k (kilobytes), m
669 (megabytes), or g (gigabytes).
670 Constrains the kernel memory available to a container. If a limit of 0
673 is specified (not using --kernel-memory), the container's kernel memory
674 is not limited. If you specify a limit, it may be rounded up to a mul‐
675 tiple of the operating system's page size and the value can be very
676 large, millions of trillions.
677 --label, -l=key=value
680 Add metadata to a container.
681 --label-file=file
684 Read in a line-delimited file of labels.
685 --link-local-ip=ip
688 Not implemented.
689 --log-driver="driver"
692 Logging driver for the container. Currently available options are
693 k8s-file, journald, and none, with json-file aliased to k8s-file for
694 scripting compatibility.
695 --log-opt=name=value
698 Logging driver specific options.
699 Set custom logging configuration. The following *name*s are supported:
702 path: specify a path to the log file
705 (e.g. --log-opt path=/var/log/container/mycontainer.json);
706 max-size: specify a max size of the log file
709 (e.g. --log-opt max-size=10mb);
710 tag: specify a custom log tag for the container
713 (e.g. --log-opt tag="{{.ImageName}}".
714 This option is currently supported only by the journald log driver.
717 --mac-address=address
720 Container MAC address (e.g. 92:d0:c6:0a:29:33).
721 Remember that the MAC address in an Ethernet network must be unique.
724 The IPv6 link-local address will be based on the device's MAC address
725 according to RFC4862.
726 --memory, -m=number[unit]
729 Memory limit. A unit can be b (bytes), k (kilobytes), m (megabytes), or
730 g (gigabytes).
731 Allows you to constrain the memory available to a container. If the
734 host supports swap memory, then the -m memory setting can be larger
735 than physical RAM. If a limit of 0 is specified (not using -m), the
736 container's memory is not limited. The actual limit may be rounded up
737 to a multiple of the operating system's page size (the value would be
738 very large, that's millions of trillions).
739 --memory-reservation=number[unit]
742 Memory soft limit. A unit can be b (bytes), k (kilobytes), m
743 (megabytes), or g (gigabytes).
744 After setting memory reservation, when the system detects memory con‐
747 tention or low memory, containers are forced to restrict their consump‐
748 tion to their reservation. So you should always set the value below
749 --memory, otherwise the hard limit will take precedence. By default,
750 memory reservation will be the same as memory limit.
751 --memory-swap=number[unit]
754 A limit value equal to memory plus swap. A unit can be b (bytes), k
755 (kilobytes), m (megabytes), or g (gigabytes).
756 Must be used with the -m (--memory) flag. The argument value should
759 always be larger than that of
760 -m (--memory) By default, it is set to double the value of --memory.
761 Set number to -1 to enable unlimited swap.
764 --memory-swappiness=number
767 Tune a container's memory swappiness behavior. Accepts an integer
768 between 0 and 100.
769 --mount=type=TYPE,TYPE-SPECIFIC-OPTION[,...]
772 Attach a filesystem mount to the container
773 Current supported mount TYPEs are bind, volume, image, tmpfs and
776 devpts. [1] ⟨#Footnote1⟩
777 e.g.
780 type=bind,source=/path/on/host,destination=/path/in/container
782 type=bind,src=/path/on/host,dst=/path/in/container,relabel=shared
784 type=volume,source=vol1,destination=/path/in/container,ro=true
786 type=tmpfs,tmpfs-size=512M,destination=/path/in/container
788 type=image,source=fedora,destination=/fedora-image,rw=true
790 type=devpts,destination=/dev/pts
792 Common Options:
794 · src, source: mount source spec for bind and volume. Mandatory for bind.
796 · dst, destination, target: mount destination spec.
798 Options specific to volume:
800 · ro, readonly: true or false (default).
802 Options specific to image:
804 · rw, readwrite: true or false (default).
806 Options specific to bind:
808 · ro, readonly: true or false (default).
810 · bind-propagation: shared, slave, private, unbindable, rshared, rslave, runbindable, or rprivate(default). See also mount(2).
812 . bind-nonrecursive: do not setup a recursive bind mount. By default it is recursive.
814 . relabel: shared, private.
816 Options specific to tmpfs:
818 · ro, readonly: true or false (default).
820 · tmpfs-size: Size of the tmpfs mount in bytes. Unlimited by default in Linux.
822 · tmpfs-mode: File mode of the tmpfs in octal. (e.g. 700 or 0700.) Defaults to 1777 in Linux.
824 · tmpcopyup: Enable copyup from the image directory at the same location to the tmpfs. Used by default.
826 · notmpcopyup: Disable copying files from the image to the tmpfs.
828 --name=name
832 Assign a name to the container.
833 The operator can identify a container in three ways:
836 · UUID long identifier
839 (“f78375b1c487e03c9438c729345e54db9d20cfa2ac1fc3494b6eb60872e74778”);
840 · UUID short identifier (“f78375b1c487”);
842 · Name (“jonah”).
844 Podman generates a UUID for each container, and if a name is not
848 assigned to the container with --name then it will generate a random
849 string name. The name is useful any place you need to identify a con‐
850 tainer. This works for both background and foreground containers.
851 --network=mode, --net
854 Set the network mode for the container. Invalid if using --dns,
855 --dns-opt, or --dns-search with --network that is set to none or con‐
856 tainer:id. If used together with --pod, the container will not join the
857 pods network namespace.
858 Valid mode values are:
861 · bridge: create a network stack on the default bridge;
864 · none: no networking;
866 · container:id: reuse another container's network stack;
868 · host: use the Podman host network stack. Note: the host mode
870 gives the container full access to local system services such
871 as D-bus and is therefore considered insecure;
872 · network-id: connect to a user-defined network, multiple net‐
874 works should be comma separated;
875 · ns:path: path to a network namespace to join;
877 · private: create a new namespace for the container (default)
879 · slirp4netns[:OPTIONS,...]: use slirp4netns(1) to create a user
881 network stack. This is the default for rootless containers. It
882 is possible to specify these additional options:
883 · allow_host_loopback=true|false: Allow the slirp4netns to
885 reach the host loopback IP (10.0.2.2). Default is false.
886 · mtu=MTU: Specify the MTU to use for this network. (Default
888 is 65520).
889 · cidr=CIDR: Specify ip range to use for this network.
891 (Default is 10.0.2.0/24).
892 · enable_ipv6=true|false: Enable IPv6. Default is false.
894 (Required for outbound_addr6).
895 · outbound_addr=INTERFACE: Specify the outbound interface
897 slirp should bind to (ipv4 traffic only).
898 · outbound_addr=IPv4: Specify the outbound ipv4 address slirp
900 should bind to.
901 · outbound_addr6=INTERFACE: Specify the outbound interface
903 slirp should bind to (ipv6 traffic only).
904 · outbound_addr6=IPv6: Specify the outbound ipv6 address slirp
906 should bind to.
907 · port_handler=rootlesskit: Use rootlesskit for port forward‐
909 ing. Default.
910 · port_handler=slirp4netns: Use the slirp4netns port forward‐
912 ing.
913 --network-alias=alias
919 Add network-scoped alias for the container
920 --no-healthcheck=true|false
923 Disable any defined healthchecks for container.
924 --no-hosts=true|false
927 Do not create /etc/hosts for the container.
928 By default, Podman will manage /etc/hosts, adding the container's own
931 IP address and any hosts from --add-host.
932 --no-hosts disables this, and the image's /etc/hosts will be preserved
935 unmodified.
936 This option conflicts with --add-host.
937 --oom-kill-disable=true|false
940 Whether to disable OOM Killer for the container or not.
941 --oom-score-adj=num
944 Tune the host's OOM preferences for containers (accepts values from
945 -1000 to 1000).
946 --os=OS
949 Override the OS, defaults to hosts, of the image to be pulled. For
950 example, windows.
951 --pid=mode
954 Set the PID namespace mode for the container. The efault is to create
955 a private PID namespace for the container.
956 · container:id: join another container's PID namespace;
959 · host: use the host's PID namespace for the container. Note the
961 host mode gives the container full access to local PID and is
962 therefore considered insecure;
963 · private: create a new namespace for the container (default)
965 · ns:path: join the specified PID namespace.
967 --pids-limit=limit
971 Tune the container's pids limit. Set to 0 to have unlimited pids for
972 the container. The default is 4096 on systems that support "pids"
973 cgroup controller.
974 --platform=OS/ARCH
977 Specify the platform for selecting the image. (Conflicts with --arch
978 and --os) The --platform option can be used to override the current
979 architecture and operating system.
980 --pod=name
983 Run container in an existing pod. If you want Podman to make the pod
984 for you, prefix the pod name with new:. To make a pod with more granu‐
985 lar options, use the podman pod create command before creating a con‐
986 tainer. If a container is run with a pod, and the pod has an
987 infra-container, the infra-container will be started before the con‐
988 tainer is.
989 --pod-id-file=path
992 Run container in an existing pod and read the pod's ID from the speci‐
993 fied file. If a container is run within a pod, and the pod has an
994 infra-container, the infra-container will be started before the con‐
995 tainer is.
996 --preserve-fds=N
999 Pass down to the process N additional file descriptors (in addition to
1000 0, 1, 2). The total FDs will be 3+N.
1001 --privileged=true|false
1004 Give extended privileges to this container. The default is false.
1005 By default, Podman containers are unprivileged (=false) and cannot, for
1008 example, modify parts of the operating system. This is because by
1009 default a container is only allowed limited access to devices. A "priv‐
1010 ileged" container is given the same access to devices as the user
1011 launching the container.
1012 A privileged container turns off the security features that isolate the
1015 container from the host. Dropped Capabilities, limited devices,
1016 read-only mount points, Apparmor/SELinux separation, and Seccomp fil‐
1017 ters are all disabled.
1018 Rootless containers cannot have more privileges than the account that
1021 launched them.
1022 --publish, -p=ip:hostPort:containerPort | ip::containerPort | hostPort:con‐
1025 tainerPort | containerPort
1026 Publish a container's port, or range of ports, to the host.
1027 Both hostPort and containerPort can be specified as a range of ports.
1030 When specifying ranges for both, the number of container ports in the
1033 range must match the number of host ports in the range.
1034 If host IP is set to 0.0.0.0 or not set at all, the port will be bound
1037 on all IPs on the host.
1038 Host port does not have to be specified (e.g. podman run -p
1041 127.0.0.1::80). If it is not, the container port will be randomly
1042 assigned a port on the host.
1043 Use podman port to see the actual mapping: podman port $CONTAINER $CON‐
1046 TAINERPORT.
1047 Note: if a container will be run within a pod, it is not necessary to
1050 publish the port for the containers in the pod. The port must only be
1051 published by the pod itself. Pod network stacks act like the network
1052 stack on the host - you have a variety of containers in the pod, and
1053 programs in the container, all sharing a single interface and IP
1054 address, and associated ports. If one container binds to a port, no
1055 other container can use that port within the pod while it is in use.
1056 Containers in the pod can also communicate over localhost by having one
1057 container bind to localhost in the pod, and another connect to that
1058 port.
1059 --publish-all, -P=true|false
1062 Publish all exposed ports to random ports on the host interfaces. The
1063 default is false.
1064 When set to true, publish all exposed ports to the host interfaces. The
1067 default is false. If the operator uses -P (or -p) then Podman will make
1068 the exposed port accessible on the host and the ports will be available
1069 to any client that can reach the host.
1070 When using this option, Podman will bind any exposed port to a random
1073 port on the host within an ephemeral port range defined by
1074 /proc/sys/net/ipv4/ip_local_port_range. To find the mapping between
1075 the host ports and the exposed ports, use podman port.
1076 --pull=always|missing|never
1079 Pull image before running. The default is missing.
1080 · missing: attempt to pull the latest image from the registries
1083 listed in registries.conf if a local image does not
1084 exist.Raise an error if the image is not in any listed reg‐
1085 istry and is not present locally.
1086 · always: Pull the image from the first registry it is found in
1088 as listed in registries.conf. Raise an error if not found in
1089 the registries, even if the image is present locally.
1090 · never: do not pull the image from the registry, use only the
1092 local version. Raise an error if the image is not present
1093 locally.
1094 --quiet, -q
1098 Suppress output information when pulling images
1099 --read-only=true|false
1102 Mount the container's root filesystem as read only.
1103 By default a container will have its root filesystem writable allowing
1106 processes to write files anywhere. By specifying the --read-only flag,
1107 the container will have its root filesystem mounted as read only pro‐
1108 hibiting any writes.
1109 --read-only-tmpfs=true|false
1112 If container is running in --read-only mode, then mount a read-write
1113 tmpfs on /run, /tmp, and /var/tmp. The default is true.
1114 --replace=true|false
1117 If another container with the same name already exists, replace and
1118 remove it. The default is false.
1119 --restart=policy
1122 Restart policy to follow when containers exit. Restart policy will not
1123 take effect if a container is stopped via the podman kill or podman
1124 stop commands.
1125 Valid policy values are:
1128 · no : Do not restart containers on exit
1131 · on-failure[:max_retries] : Restart containers when they exit
1133 with a non-zero exit code, retrying indefinitely or until the
1134 optional max_retries count is hit
1135 · always : Restart containers when they exit,
1137 regardless of status, retrying indefinitely
1138 · unless-stopped : Identical to always
1140 Please note that restart will not restart containers after a system
1144 reboot. If this functionality is required in your environment, you can
1145 invoke Podman from a systemd.unit(5) file, or create an init script for
1146 whichever init system is in use. To generate systemd unit files,
1147 please see podman generate systemd.
1148 --rm=true|false
1151 Automatically remove the container when it exits. The default is false.
1152 --rmi=true|false
1155 After exit of the container, remove the image unless another container
1156 is using it. The default is false.
1157 --rootfs
1160 If specified, the first argument refers to an exploded container on the
1161 file system.
1162 This is useful to run a container without requiring any image manage‐
1165 ment, the rootfs of the container is assumed to be managed externally.
1166 Note: On SELinux systems, the rootfs needs the correct label, which is
1169 by default unconfined_u:object_r:container_file_t.
1170 --sdnotify=container|conmon|ignore
1173 Determines how to use the NOTIFY_SOCKET, as passed with systemd and
1174 Type=notify.
1175 Default is container, which means allow the OCI runtime to proxy the
1178 socket into the container to receive ready notification. Podman will
1179 set the MAINPID to conmon's pid. The conmon option sets MAINPID to
1180 conmon's pid, and sends READY when the container has started. The
1181 socket is never passed to the runtime or the container. The ignore
1182 option removes NOTIFY_SOCKET from the environment for itself and child
1183 processes, for the case where some other process above Podman uses
1184 NOTIFY_SOCKET and Podman should not use it.
1185 --seccomp-policy=policy
1188 Specify the policy to select the seccomp profile. If set to image, Pod‐
1189 man will look for a "io.podman.seccomp.profile" label in the con‐
1190 tainer-image config and use its value as a seccomp profile. Otherwise,
1191 Podman will follow the default policy by applying the default profile
1192 unless specified otherwise via --security-opt seccomp as described
1193 below.
1194 Note that this feature is experimental and may change in the future.
1197 --security-opt=option
1200 Security Options
1201 · apparmor=unconfined : Turn off apparmor confinement for the
1204 container
1205 · apparmor=your-profile : Set the apparmor confinement profile
1207 for the container
1208 · label=user:USER: Set the label user for the container pro‐
1210 cesses
1211 · label=role:ROLE: Set the label role for the container pro‐
1213 cesses
1214 · label=type:TYPE: Set the label process type for the container
1216 processes
1217 · label=level:LEVEL: Set the label level for the container pro‐
1219 cesses
1220 · label=filetype:TYPE_: Set the label file type for the con‐
1222 tainer files
1223 · label=disable: Turn off label separation for the container
1225 · mask=/path/1:/path/2: The paths to mask separated by a colon.
1227 A masked path cannot be accessed inside the container.
1228 · no-new-privileges: Disable container processes from gaining
1230 additional privileges
1231 · seccomp=unconfined: Turn off seccomp confinement for the con‐
1233 tainer
1234 · seccomp=profile.json: Allowed syscall list seccomp JSON file
1236 to be used as a seccomp filter
1237 · proc-opts=OPTIONS : Comma separated list of options to use for
1239 the /proc mount. More details for the possible mount options
1240 are specified at proc(5) man page.
1241 · unmask=ALL or /path/1:/path/2: Paths to unmask separated by a
1243 colon. If set to ALL, it will unmask all the paths that are
1244 masked or made read only by default. The default masked paths
1245 are /proc/acpi, /proc/kcore, /proc/keys, /proc/latency_stats,
1246 /proc/sched_debug, /proc/scsi, /proc/timer_list,
1247 /proc/timer_stats, /sys/firmware, and /sys/fs/selinux.. The
1248 default paths that are read only are /proc/asound, /proc/bus,
1249 /proc/fs, /proc/irq, /proc/sys, /proc/sysrq-trigger.
1250 Note: Labeling can be disabled for all containers by setting
1254 label=false in the containers.conf(5) file.
1255 --shm-size=number[unit]
1258 Size of /dev/shm. A unit can be b (bytes), k (kilobytes), m
1259 (megabytes), or g (gigabytes). If you omit the unit, the system uses
1260 bytes. If you omit the size entirely, the default is 64m. When size is
1261 0, there is no limit on the amount of memory used for IPC by the con‐
1262 tainer.
1263 --sig-proxy=true|false
1266 Sets whether the signals sent to the podman run command are proxied to
1267 the container process. SIGCHLD, SIGSTOP, and SIGKILL are not proxied.
1268 The default is true.
1269 --stop-signal=signal
1272 Signal to stop a container. Default is SIGTERM.
1273 --stop-timeout=seconds
1276 Timeout to stop a container. Default is 10. Remote connections use
1277 local containers.conf for defaults
1278 --subgidname=name
1281 Run the container in a new user namespace using the map with name in
1282 the /etc/subgid file. If calling podman run as an unprivileged user,
1283 the user needs to have the right to use the mapping. See subgid(5).
1284 This flag conflicts with --userns and --gidmap.
1285 --subuidname=name
1288 Run the container in a new user namespace using the map with name in
1289 the /etc/subuid file. If calling podman run as an unprivileged user,
1290 the user needs to have the right to use the mapping. See subuid(5).
1291 This flag conflicts with --userns and --uidmap.
1292 --sysctl=name=value
1295 Configure namespaced kernel parameters at runtime.
1296 For the IPC namespace, the following sysctls are allowed:
1299 · kernel.msgmax
1302 · kernel.msgmnb
1304 · kernel.msgmni
1306 · kernel.sem
1308 · kernel.shmall
1310 · kernel.shmmax
1312 · kernel.shmmni
1314 · kernel.shm_rmid_forced
1316 · Sysctls beginning with fs.mqueue.*
1318 Note: if you use the --ipc=host option, the above sysctls will not be
1322 allowed.
1323 For the network namespace, the following sysctls are allowed:
1326 · Sysctls beginning with net.*
1329 Note: if you use the --network=host option, these sysctls will not be
1333 allowed.
1334 --systemd=true|false|always
1337 Run container in systemd mode. The default is true.
1338 The value always enforces the systemd mode is enforced without looking
1341 at the executable name. Otherwise, if set to true and the command you
1342 are running inside the container is systemd, /usr/sbin/init, /sbin/init
1343 or /usr/local/sbin/init.
1344 If the command you are running inside of the container is systemd Pod‐
1347 man will setup tmpfs mount points in the following directories:
1348 · /run
1351 · /run/lock
1353 · /tmp
1355 · /sys/fs/cgroup/systemd
1357 · /var/lib/journal
1359 It will also set the default stop signal to SIGRTMIN+3.
1363 This allows systemd to run in a confined container without any modifi‐
1366 cations.
1367 Note that on SELinux systems, systemd attempts to write to the cgroup
1370 file system. Containers writing to the cgroup file system are denied by
1371 default. The container_manage_cgroup boolean must be enabled for this
1372 to be allowed on an SELinux separated system.
1373 setsebool -P container_manage_cgroup true
1376 --tmpfs=fs
1380 Create a tmpfs mount.
1381 Mount a temporary filesystem (tmpfs) mount into a container, for exam‐
1384 ple:
1385 $ podman run -d --tmpfs /tmp:rw,size=787448k,mode=1777 my_image
1388 This command mounts a tmpfs at /tmp within the container. The supported
1392 mount options are the same as the Linux default mount flags. If you do
1393 not specify any options, the systems uses the following options:
1394 rw,noexec,nosuid,nodev.
1395 --tty, -t=true|false
1398 Allocate a pseudo-TTY. The default is false.
1399 When set to true, Podman will allocate a pseudo-tty and attach to the
1402 standard input of the container. This can be used, for example, to run
1403 a throwaway interactive shell. The default is false.
1404 NOTE: The --tty flag prevents redirection of standard output. It com‐
1407 bines STDOUT and STDERR, it can insert control characters, and it can
1408 hang pipes. This option should only be used when run interactively in a
1409 terminal. When feeding input to Podman, use -i only, not -it.
1410 echo "asdf" | podman run --rm -i someimage /bin/cat
1413 --tz=timezone
1417 Set timezone in container. This flag takes area-based timezones, GMT
1418 time, as well as local, which sets the timezone in the container to
1419 match the host machine. See /usr/share/zoneinfo/ for valid timezones.
1420 Remote connections use local containers.conf for defaults
1421 --umask=umask
1424 Set the umask inside the container. Defaults to 0022. Remote connec‐
1425 tions use local containers.conf for defaults
1426 --uidmap=container_uid:host_uid:amount
1429 Run the container in a new user namespace using the supplied mapping.
1430 This option conflicts with the --userns and --subuidname flags. This
1431 option can be passed several times to map different ranges. If calling
1432 podman run as an unprivileged user, the user needs to have the right to
1433 use the mapping. See subuid(5).
1434 The following example maps uids 0-1999 in the container to the uids
1437 30000-31999 on the host: --uidmap=0:30000:2000.
1438 Important note: The new user namespace mapping based on --uidmap is
1441 based on the initial mapping made in the /etc/subuid file. Assuming
1442 there is a /etc/subuid mapping username:100000:65536, then username
1443 is initially mapped to a namespace starting with uid 100000 for 65536
1444 ids. From here the --uidmap mapping to the new namespace starts from 0
1445 again, but is based on the initial mapping. Meaning username is ini‐
1446 tially mapped to uid 100000 which is referenced as 0 in the following
1447 --uidmap mapping. In terms of the example above: The user username is
1448 mapped to user 100000 of the initial namespace then the 30000st id of
1449 this namespace (which is uid 130000 in this namespace) is mapped to
1450 container namespace user id 0. (username -> 100000 / 30000 -> 0)
1451 Note: A minimal mapping has to have at least container uid 0 mapped to
1454 the parent user namespace.
1455 --ulimit=option
1458 Ulimit options. You can use host to copy the current configuration from
1459 the host.
1460 --user, -u=[user | user:group | uid | uid:gid | user:gid | uid:group ]
1463 Sets the username or UID used and optionally the groupname or GID for
1464 the specified command.
1465 Without this argument, the command will run as the user specified in
1468 the container image. Unless overridden by a USER command in the Con‐
1469 tainerfile or by a value passed to this option, this user generally
1470 defaults to root.
1471 When a user namespace is not in use, the UID and GID used within the
1474 container and on the host will match. When user namespaces are in use,
1475 however, the UID and GID in the container may correspond to another UID
1476 and GID on the host. In rootless containers, for example, a user names‐
1477 pace is always used, and root in the container will by default corre‐
1478 spond to the UID and GID of the user invoking Podman.
1479 --userns=auto|host|keep-id|container:id|ns:namespace
1482 Set the user namespace mode for the container. It defaults to the POD‐
1483 MAN_USERNS environment variable. An empty value ("") means user names‐
1484 paces are disabled unless an explicit mapping is set with they
1485 --uidmapping and --gidmapping options.
1486 · auto: automatically create a namespace. It is possible to
1489 specify other options to auto. The supported options are
1490 size=SIZE to specify an explicit size for the automatic user
1491 namespace. e.g. --userns=auto:size=8192. If size is not speci‐
1492 fied, auto will guess a size for the user namespace. uidmap‐
1493 ping=HOST_UID:CONTAINER_UID:SIZE to force a UID mapping to be
1494 present in the user namespace. gidmapping=HOST_UID:CON‐
1495 TAINER_UID:SIZE to force a GID mapping to be present in the
1496 user namespace.
1497 · host: run in the user namespace of the caller. The processes
1499 running in the container will have the same privileges on the
1500 host as any other process launched by the calling user
1501 (default).
1502 · keep-id: creates a user namespace where the current rootless
1504 user's UID:GID are mapped to the same values in the container.
1505 This option is ignored for containers created by the root
1506 user.
1507 · ns: run the container in the given existing user namespace.
1509 · private: create a new namespace for the container.
1511 · container: join the user namespace of the specified container.
1513 This option is incompatible with --gidmap, --uidmap, --subuidname and
1517 --subgidname.
1518 --uts=mode
1521 Set the UTS namespace mode for the container. The following values are
1522 supported:
1523 · host: use the host's UTS namespace inside the container.
1526 · private: create a new namespace for the container (default).
1528 · ns:[path]: run the container in the given existing UTS names‐
1530 pace.
1531 · container:[container]: join the UTS namespace of the specified
1533 container.
1534 --variant=VARIANT
1538 Use VARIANT instead of the default architecture variant of the con‐
1539 tainer image. Some images can use multiple variants of the arm archi‐
1540 tectures, such as arm/v5 and arm/v7.
1541 --volume, -v[=[[SOURCE-VOLUME|HOST-DIR:]CONTAINER-DIR[:OPTIONS]]]
1544 Create a bind mount. If you specify /HOST-DIR:/CONTAINER-DIR, Podman
1545 bind mounts host-dir in the host to CONTAINER-DIR in the Podman con‐
1546 tainer. Similarly, SOURCE-VOLUME:/CONTAINER-DIR will mount the volume
1547 in the host to the container. If no such named volume exists, Podman
1548 will create one.
1549 The options is a comma delimited list and can be: [1] ⟨#Footnote1⟩
1552 · rw|ro
1555 · z|Z
1557 · [r]shared|[r]slave|[r]private[r]unbindable
1559 · [r]bind
1561 · [no]exec
1563 · [no]dev
1565 · [no]suid
1567 · [O]
1569 The CONTAINER-DIR must be an absolute path such as /src/docs. The vol‐
1573 ume will be mounted into the container at this directory.
1574 Volumes may specify a source as well, as either a directory on the host
1577 or the name of a named volume. If no source is given, the volume will
1578 be created as an anonymously named volume with a randomly generated
1579 name, and will be removed when the container is removed via the --rm
1580 flag or podman rm --volumes.
1581 If a volume source is specified, it must be a path on the host or the
1584 name of a named volume. Host paths are allowed to be absolute or rela‐
1585 tive; relative paths are resolved relative to the directory Podman is
1586 run in. If the source does not exist, Podman will return an error.
1587 Users must pre-create the source files or directories.
1588 Any source that does not begin with a . or / will be treated as the
1591 name of a named volume. If a volume with that name does not exist, it
1592 will be created. Volumes created with names are not anonymous, and
1593 they are not removed by the --rm option and the podman rm --volumes
1594 command.
1595 You can specify multiple -v options to mount one or more volumes into a
1598 container.
1599 Write Protected Volume Mounts
1602 You can add :ro or :rw option to mount a volume in read-only or
1605 read-write mode, respectively. By default, the volumes are mounted
1606 read-write.
1607 Labeling Volume Mounts
1610 Labeling systems like SELinux require that proper labels are placed on
1613 volume content mounted into a container. Without a label, the security
1614 system might prevent the processes running inside the container from
1615 using the content. By default, Podman does not change the labels set by
1616 the OS.
1617 To change a label in the container context, you can add either of two
1620 suffixes :z or :Z to the volume mount. These suffixes tell Podman to
1621 relabel file objects on the shared volumes. The z option tells Podman
1622 that two containers share the volume content. As a result, Podman
1623 labels the content with a shared content label. Shared volume labels
1624 allow all containers to read/write content. The Z option tells Podman
1625 to label the content with a private unshared label.
1626 Overlay Volume Mounts
1629 The :O flag tells Podman to mount the directory from the host as a tem‐
1632 porary storage using the overlay file system. The container processes
1633 can modify content within the mountpoint which is stored in the con‐
1634 tainer storage in a separate directory. In overlay terms, the source
1635 directory will be the lower, and the container storage directory will
1636 be the upper. Modifications to the mount point are destroyed when the
1637 container finishes executing, similar to a tmpfs mount point being
1638 unmounted.
1639 Subsequent executions of the container will see the original source
1642 directory content, any changes from previous container executions no
1643 longer exist.
1644 One use case of the overlay mount is sharing the package cache from the
1647 host into the container to allow speeding up builds.
1648 Note:
1651 - The `O` flag conflicts with other options listed above.
1654 Content mounted into the container is labeled with the private label.
1658 On SELinux systems, labels in the source directory must be read‐
1659 able by the container label. Usually containers can read/execute con‐
1660 tainer_share_t and can read/write container_file_t. If you can not
1661 change the labels on a source volume, SELinux container separation must
1662 be disabled for the container to work.
1663 - The source directory mounted into the container with an overlay
1664 mount should not be modified, it can cause unexpected failures. It is
1665 recommended that you do not modify the directory until the container
1666 finishes running.
1667 Only the current container can use a private volume.
1670 Mounts propagation
1673 By default bind mounted volumes are private. That means any mounts done
1676 inside container will not be visible on host and vice versa. One can
1677 change this behavior by specifying a volume mount propagation property.
1678 Making a volume shared mounts done under that volume inside container
1679 will be visible on host and vice versa. Making a volume slave enables
1680 only one way mount propagation and that is mounts done on host under
1681 that volume will be visible inside container but not the other way
1682 around. [1] ⟨#Footnote1⟩
1683 To control mount propagation property of volume one can use [r]shared,
1686 [r]slave, [r]private or [r]unbindable propagation flag. Propagation
1687 property can be specified only for bind mounted volumes and not for
1688 internal volumes or named volumes. For mount propagation to work source
1689 mount point (mount point where source dir is mounted on) has to have
1690 right propagation properties. For shared volumes, source mount point
1691 has to be shared. And for slave volumes, source mount has to be either
1692 shared or slave. [1] ⟨#Footnote1⟩
1693 If you want to recursively mount a volume and all of its submounts into
1696 a container, then you can use the rbind option. By default the bind
1697 option is used, and submounts of the source directory will not be
1698 mounted into the container.
1699 Mounting the volume with the nosuid options means that SUID applica‐
1702 tions on the volume will not be able to change their privilege. By
1703 default volumes are mounted with nosuid.
1704 Mounting the volume with the noexec option means that no executables on
1707 the volume will be able to executed within the container.
1708 Mounting the volume with the nodev option means that no devices on the
1711 volume will be able to be used by processes within the container. By
1712 default volumes are mounted with nodev.
1713 If the host-dir is a mount point, then dev, suid, and exec options are
1716 ignored by the kernel.
1717 Use df $hostdir to figure out the source mount, and then use findmnt -o
1720 TARGET,PROPAGATION source-mount-dir to figure out propagation proper‐
1721 ties of source mount. If findmnt[4m(1) utility is not available, then one
1722 can look at mount entry for source mount point in /proc/self/mountinfo.
1723 Look at the "optional fields" and see if any propagation properties are
1724 specified. In there, shared:N means the mount is shared, master:N
1725 means mount is slave, and if nothing is there, the mount is private.
1726 [1] ⟨#Footnote1⟩
1727 To change propagation properties of a mount point, use mount(8) com‐
1730 mand. For example, if one wants to bind mount source directory /foo,
1731 one can do mount --bind /foo /foo and mount --make-private
1732 --make-shared /foo. This will convert /foo into a shared mount point.
1733 Alternatively, one can directly change propagation properties of source
1734 mount. Say, if / is source mount for /foo, then use mount --make-shared
1735 / to convert / into a shared mount.
1736 --volumes-from[=CONTAINER[:OPTIONS]]
1739 Mount volumes from the specified container(s). Used to share volumes
1740 between containers. The options is a comma delimited list with the fol‐
1741 lowing available elements:
1742 · rw|ro
1745 · z
1747 Mounts already mounted volumes from a source container onto another
1751 container. You must supply the source's container-id or container-name.
1752 To share a volume, use the --volumes-from option when running the tar‐
1753 get container. You can share volumes even if the source container is
1754 not running.
1755 By default, Podman mounts the volumes in the same mode (read-write or
1758 read-only) as it is mounted in the source container. You can change
1759 this by adding a ro or rw option.
1760 Labeling systems like SELinux require that proper labels are placed on
1763 volume content mounted into a container. Without a label, the security
1764 system might prevent the processes running inside the container from
1765 using the content. By default, Podman does not change the labels set by
1766 the OS.
1767 To change a label in the container context, you can add z to the volume
1770 mount. This suffix tells Podman to relabel file objects on the shared
1771 volumes. The z option tells Podman that two containers share the volume
1772 content. As a result, Podman labels the content with a shared content
1773 label. Shared volume labels allow all containers to read/write content.
1774 If the location of the volume from the source container overlaps with
1777 data residing on a target container, then the volume hides that data on
1778 the target.
1779 --workdir, -w=dir
1782 Working directory inside the container.
1783 The default working directory for running binaries within a container
1786 is the root directory (/). The image developer can set a different
1787 default with the WORKDIR instruction. The operator can override the
1788 working directory by using the -w option.
1789
1792 The exit code from podman run gives information about why the container
1793 failed to run or why it exited. When podman run exits with a non-zero
1794 code, the exit codes follow the chroot(1) standard, see below:
1795 125 The error is with Podman itself
1798 $ podman run --foo busybox; echo $?
1801 Error: unknown flag: --foo
1802 125
1803 126 The contained command cannot be invoked
1807 $ podman run busybox /etc; echo $?
1810 Error: container_linux.go:346: starting container process caused "exec: \"/etc\": permission denied": OCI runtime error
1811 126
1812 127 The contained command cannot be found
1816 $ podman run busybox foo; echo $?
1819 Error: container_linux.go:346: starting container process caused "exec: \"foo\": executable file not found in $PATH": OCI runtime error
1820 127
1821 Exit code contained command exit code
1825 $ podman run busybox /bin/sh -c 'exit 3'; echo $?
1828 3
1829
1833 Running container in read-only mode
1834 During container image development, containers often need to write to
1835 the image content. Installing packages into /usr, for example. In pro‐
1836 duction, applications seldom need to write to the image. Container
1837 applications write to volumes if they need to write to file systems at
1838 all. Applications can be made more secure by running them in read-only
1839 mode using the --read-only switch. This protects the containers image
1840 from modification. Read only containers may still need to write tempo‐
1841 rary data. The best way to handle this is to mount tmpfs directories on
1842 /run and /tmp.
1843 $ podman run --read-only -i -t fedora /bin/bash
1846 $ podman run --read-only --read-only-tmpfs=false --tmpfs /run -i -t fedora /bin/bash
1848 Exposing log messages from the container to the host's log
1852 If you want messages that are logged in your container to show up in
1853 the host's syslog/journal then you should bind mount the /dev/log
1854 directory as follows.
1855 $ podman run -v /dev/log:/dev/log -i -t fedora /bin/bash
1858 From inside the container you can test this by sending a message to the
1862 log.
1863 (bash)# logger "Hello from my container"
1866 Then exit and check the journal.
1870 (bash)# exit
1873 $ journalctl -b | grep Hello
1875 This should list the message sent to logger.
1879 Attaching to one or more from STDIN, STDOUT, STDERR
1882 If you do not specify -a, Podman will attach everything (stdin, stdout,
1883 stderr). You can specify to which of the three standard streams
1884 (stdin, stdout, stderr) you'd like to connect instead, as in:
1885 $ podman run -a stdin -a stdout -i -t fedora /bin/bash
1888 Sharing IPC between containers
1892 Using shm_server.c available here:
1893 https://www.cs.cf.ac.uk/Dave/C/node27.html
1894 Testing --ipc=host mode:
1897 Host shows a shared memory segment with 7 pids attached, happens to be
1900 from httpd:
1901 $ sudo ipcs -m
1904 ------ Shared Memory Segments --------
1906 key shmid owner perms bytes nattch status
1907 0x01128e25 0 root 600 1000 7
1908 Now run a regular container, and it correctly does NOT see the shared
1912 memory segment from the host:
1913 $ podman run -it shm ipcs -m
1916 ------ Shared Memory Segments --------
1918 key shmid owner perms bytes nattch status
1919 Run a container with the new --ipc=host option, and it now sees the
1923 shared memory segment from the host httpd:
1924 $ podman run -it --ipc=host shm ipcs -m
1927 ------ Shared Memory Segments --------
1929 key shmid owner perms bytes nattch status
1930 0x01128e25 0 root 600 1000 7
1931 Testing --ipc=container:id mode:
1935 Start a container with a program to create a shared memory segment:
1938 $ podman run -it shm bash
1941 $ sudo shm/shm_server
1942 $ sudo ipcs -m
1943 ------ Shared Memory Segments --------
1945 key shmid owner perms bytes nattch status
1946 0x0000162e 0 root 666 27 1
1947 Create a 2nd container correctly shows no shared memory segment from
1951 1st container:
1952 $ podman run shm ipcs -m
1955 ------ Shared Memory Segments --------
1957 key shmid owner perms bytes nattch status
1958 Create a 3rd container using the --ipc=container:id option, now it
1962 shows the shared memory segment from the first:
1963 $ podman run -it --ipc=container:ed735b2264ac shm ipcs -m
1966 $ sudo ipcs -m
1967 ------ Shared Memory Segments --------
1969 key shmid owner perms bytes nattch status
1970 0x0000162e 0 root 666 27 1
1971 Mapping Ports for External Usage
1975 The exposed port of an application can be mapped to a host port using
1976 the -p flag. For example, an httpd port 80 can be mapped to the host
1977 port 8080 using the following:
1978 $ podman run -p 8080:80 -d -i -t fedora/httpd
1981 Mounting External Volumes
1985 To mount a host directory as a container volume, specify the absolute
1986 path to the directory and the absolute path for the container directory
1987 separated by a colon. If the source is a named volume maintained by
1988 Podman, it is recommended to use its name rather than the path to the
1989 volume. Otherwise the volume will be considered as an orphan and wiped
1990 if you execute podman volume prune:
1991 $ podman run -v /var/db:/data1 -i -t fedora bash
1994 $ podman run -v data:/data2 -i -t fedora bash
1996 $ podman run -v /var/cache/dnf:/var/cache/dnf:O -ti fedora dnf -y update
1998 Using --mount flags to mount a host directory as a container folder,
2002 specify the absolute path to the directory or the volume name, and the
2003 absolute path within the container directory:
2004 $ podman run --mount type=bind,src=/var/db,target=/data1 busybox sh
2007 $ podman run --mount type=bind,src=volume-name,target=/data1 busybox sh
2009 When using SELinux, be aware that the host has no knowledge of con‐
2013 tainer SELinux policy. Therefore, in the above example, if SELinux pol‐
2014 icy is enforced, the /var/db directory is not writable to the con‐
2015 tainer. A "Permission Denied" message will occur and an avc: message in
2016 the host's syslog.
2017 To work around this, at time of writing this man page, the following
2020 command needs to be run in order for the proper SELinux policy type
2021 label to be attached to the host directory:
2022 $ chcon -Rt svirt_sandbox_file_t /var/db
2025 Now, writing to the /data1 volume in the container will be allowed and
2029 the changes will also be reflected on the host in /var/db.
2030 Using alternative security labeling
2033 You can override the default labeling scheme for each container by
2034 specifying the --security-opt flag. For example, you can specify the
2035 MCS/MLS level, a requirement for MLS systems. Specifying the level in
2036 the following command allows you to share the same content between con‐
2037 tainers.
2038 podman run --security-opt label=level:s0:c100,c200 -i -t fedora bash
2041 An MLS example might be:
2045 $ podman run --security-opt label=level:TopSecret -i -t rhel7 bash
2048 To disable the security labeling for this container versus running with
2052 the
2053 --permissive flag, use the following command:
2056 $ podman run --security-opt label=disable -i -t fedora bash
2057 If you want a tighter security policy on the processes within a con‐
2061 tainer, you can specify an alternate type for the container. You could
2062 run a container that is only allowed to listen on Apache ports by exe‐
2063 cuting the following command:
2064 $ podman run --security-opt label=type:svirt_apache_t -i -t centos bash
2067 Note you would have to write policy defining a svirt_apache_t type.
2071 To mask additional specific paths in the container, specify the paths
2074 separated by a colon using the mask option with the --security-opt
2075 flag.
2076 $ podman run --security-opt mask=/foo/bar:/second/path fedora bash
2079 To unmask all the paths that are masked by default, set the unmask
2083 option to ALL. Or to only unmask specific paths, specify the paths as
2084 shown above with the mask option.
2085 $ podman run --security-opt unmask=ALL fedora bash
2088 $ podman run --security-opt unmask=/foo/bar:/sys/firmware fedora bash
2092 Setting device weight
2096 If you want to set /dev/sda device weight to 200, you can specify the
2097 device weight by --blkio-weight-device flag. Use the following command:
2098 $ podman run -it --blkio-weight-device "/dev/sda:200" ubuntu
2101 Using a podman container with input from a pipe
2105 $ echo "asdf" | podman run --rm -i --entrypoint /bin/cat someimage
2106 asdf
2107 Setting Namespaced Kernel Parameters (Sysctls)
2111 The --sysctl sets namespaced kernel parameters (sysctls) in the con‐
2112 tainer. For example, to turn on IP forwarding in the containers network
2113 namespace, run this command:
2114 $ podman run --sysctl net.ipv4.ip_forward=1 someimage
2117 Note that not all sysctls are namespaced. Podman does not support
2121 changing sysctls inside of a container that also modify the host sys‐
2122 tem. As the kernel evolves we expect to see more sysctls become names‐
2123 paced.
2124 See the definition of the --sysctl option above for the current list of
2127 supported sysctls.
2128 Set UID/GID mapping in a new user namespace
2131 Running a container in a new user namespace requires a mapping of the
2132 uids and gids from the host.
2133 $ podman run --uidmap 0:30000:7000 --gidmap 0:30000:7000 fedora echo hello
2136 Configuring Storage Options from the command line
2140 Podman allows for the configuration of storage by changing the values
2141 in the /etc/container/storage.conf or by using global options. This
2142 shows how to setup and use fuse-overlayfs for a one time run of busybox
2143 using global options.
2144 podman --log-level=debug --storage-driver overlay --storage-opt "overlay.mount_program=/usr/bin/fuse-overlayfs" run busybox /bin/sh
2147 Configure timezone in a container
2151 $ podman run --tz=local alpine date
2152 $ podman run --tz=Asia/Shanghai alpine date
2153 $ podman run --tz=US/Eastern alpine date
2154 Rootless Containers
2158 Podman runs as a non root user on most systems. This feature requires
2159 that a new enough version of shadow-utils be installed. The
2160 shadow-utils package must include the newuidmap(1) and newgidmap(1)
2161 executables.
2162 Note: RHEL7 and Centos 7 will not have this feature until RHEL7.7 is
2165 released.
2166 In order for users to run rootless, there must be an entry for their
2169 username in /etc/subuid and /etc/subgid which lists the UIDs for their
2170 user namespace.
2171 Rootless Podman works better if the fuse-overlayfs and slirp4netns
2174 packages are installed. The fuse-overlay package provides a userspace
2175 overlay storage driver, otherwise users need to use the vfs storage
2176 driver, which is diskspace expensive and does not perform well.
2177 slirp4netns is required for VPN, without it containers need to be run
2178 with the --network=host flag.
2179
2182 Environment variables within containers can be set using multiple dif‐
2183 ferent options, in the following order of precedence (later entries
2184 override earlier entries):
2185 · Container image: Any environment variables specified in the
2188 container image.
2189 · --http-proxy: By default, several environment variables will
2191 be passed in from the host, such as http_proxy and no_proxy.
2192 See --http-proxy for details.
2193 · --env-host: Host environment of the process executing Podman
2195 is added.
2196 · --env-file: Any environment variables specified via env-files.
2198 If multiple files specified, then they override each other in
2199 order of entry.
2200 · --env: Any environment variables specified will override pre‐
2202 vious settings.
2203 Run containers and set the environment ending with a * and a *****:
2207 $ export ENV1=a
2210 $ podman run --env ENV* alpine printenv ENV1
2211 a
2212 $ podman run --env ENV*****=b alpine printenv ENV*****
2214 b
2215
2219 /etc/subuid
2220 /etc/subgid
2223 NOTE: Use the environment variable TMPDIR to change the temporary stor‐
2226 age location of downloaded container images. Podman defaults to use
2227 /var/tmp.
2228
2231 podman(1), podman-save(1), podman-ps(1), podman-attach(1), pod‐
2232 man-pod-create(1), podman-port(1), podman-kill(1), podman-stop(1), pod‐
2233 man-generate-systemd(1) podman-rm(1), subgid(5), subuid(5), contain‐
2234 ers.conf(5), systemd.unit(5), setsebool(8), slirp4netns(1), fuse-over‐
2235 layfs(1), proc(5)**.
2236
2239 September 2018, updated by Kunal Kushwaha
2240 October 2017, converted from Docker documentation to Podman by Dan
2243 Walsh for Podman dwalsh@redhat.com ⟨mailto:dwalsh@redhat.com⟩
2244 November 2015, updated by Sally O'Malley somalley@redhat.com
2247 ⟨mailto:somalley@redhat.com⟩
2248 June 2014, updated by Sven Dowideit SvenDowideit@home.org.au
2251 ⟨mailto:SvenDowideit@home.org.au⟩
2252 April 2014, Originally compiled by William Henry whenry@redhat.com
2255 ⟨mailto:whenry@redhat.com⟩ based on docker.com source material and
2256 internal work.
2257
2260 1: The Podman project is committed to inclusivity, a core value of open
2261 source. The master and slave mount propagation terminology used here is
2262 problematic and divisive, and should be changed. However, these terms
2263 are currently used within the Linux kernel and must be used as-is at
2264 this time. When the kernel maintainers rectify this usage, Podman will
2265 follow suit immediately.
2266 podman-run(1)()