5 include::attributes.txt[]
11 pct - Tool to manage Linux Containers (LXC) on Proxmox VE
17 include::pct.1-synopsis.adoc[]
24 Proxmox Container Toolkit
25 =========================
26 include::attributes.txt[]
30 :title: Linux Container
33 Containers are a lightweight alternative to fully virtualized
34 VMs. Instead of emulating a complete Operating System (OS), containers
35 simply use the OS of the host they run on. This implies that all
36 containers use the same kernel, and that they can access resources
37 from the host directly.
39 This is great because containers do not waste CPU power nor memory due
40 to kernel emulation. Container run-time costs are close to zero and
41 usually negligible. But there are also some drawbacks you need to
44 * You can only run Linux based OS inside containers, i.e. it is not
45 possible to run FreeBSD or MS Windows inside.
47 * For security reasons, access to host resources needs to be
48 restricted. This is done with AppArmor, SecComp filters and other
49 kernel features. Be prepared that some syscalls are not allowed
52 {pve} uses https://linuxcontainers.org/[LXC] as underlying container
53 technology. We consider LXC as low-level library, which provides
54 countless options. It would be too difficult to use those tools
55 directly. Instead, we provide a small wrapper called `pct`, the
56 "Proxmox Container Toolkit".
58 The toolkit is tightly coupled with {pve}. That means that it is aware
59 of the cluster setup, and it can use the same network and storage
60 resources as fully virtualized VMs. You can even use the {pve}
61 firewall, or manage containers using the HA framework.
63 Our primary goal is to offer an environment as one would get from a
64 VM, but without the additional overhead. We call this "System
67 NOTE: If you want to run micro-containers (with docker, rkt, ...), it
68 is best to run them inside a VM.
71 Security Considerations
72 -----------------------
74 Containers use the same kernel as the host, so there is a big attack
75 surface for malicious users. You should consider this fact if you
76 provide containers to totally untrusted people. In general, fully
77 virtualized VMs provide better isolation.
79 The good news is that LXC uses many kernel security features like
80 AppArmor, CGroups and PID and user namespaces, which makes containers
81 usage quite secure. We distinguish two types of containers:
87 Security is done by dropping capabilities, using mandatory access
88 control (AppArmor), SecComp filters and namespaces. The LXC team
89 considers this kind of container as unsafe, and they will not consider
90 new container escape exploits to be security issues worthy of a CVE
91 and quick fix. So you should use this kind of containers only inside a
92 trusted environment, or when no untrusted task is running as root in
96 Unprivileged Containers
97 ~~~~~~~~~~~~~~~~~~~~~~~
99 This kind of containers use a new kernel feature called user
100 namespaces. The root UID 0 inside the container is mapped to an
101 unprivileged user outside the container. This means that most security
102 issues (container escape, resource abuse, ...) in those containers
103 will affect a random unprivileged user, and so would be a generic
104 kernel security bug rather than an LXC issue. The LXC team thinks
105 unprivileged containers are safe by design.
107 [[pct_configuration]]
111 The `/etc/pve/lxc/<CTID>.conf` file stores container configuration,
112 where `<CTID>` is the numeric ID of the given container. Like all
113 other files stored inside `/etc/pve/`, they get automatically
114 replicated to all other cluster nodes.
116 NOTE: CTIDs < 100 are reserved for internal purposes, and CTIDs need to be
119 .Example Container Configuration
126 net0: bridge=vmbr0,hwaddr=66:64:66:64:64:36,ip=dhcp,name=eth0,type=veth
127 rootfs: local:107/vm-107-disk-1.raw,size=7G
130 Those configuration files are simple text files, and you can edit them
131 using a normal text editor (`vi`, `nano`, ...). This is sometimes
132 useful to do small corrections, but keep in mind that you need to
133 restart the container to apply such changes.
135 For that reason, it is usually better to use the `pct` command to
136 generate and modify those files, or do the whole thing using the GUI.
137 Our toolkit is smart enough to instantaneously apply most changes to
138 running containers. This feature is called "hot plug", and there is no
139 need to restart the container in that case.
145 Container configuration files use a simple colon separated key/value
146 format. Each line has the following format:
153 Blank lines in those files are ignored, and lines starting with a `#`
154 character are treated as comments and are also ignored.
156 It is possible to add low-level, LXC style configuration directly, for
159 lxc.init_cmd: /sbin/my_own_init
163 lxc.init_cmd = /sbin/my_own_init
165 Those settings are directly passed to the LXC low-level tools.
172 When you create a snapshot, `pct` stores the configuration at snapshot
173 time into a separate snapshot section within the same configuration
174 file. For example, after creating a snapshot called ``testsnapshot'',
175 your configuration file will look like this:
177 .Container configuration with snapshot
191 There are a few snapshot related properties like `parent` and
192 `snaptime`. The `parent` property is used to store the parent/child
193 relationship between snapshots. `snaptime` is the snapshot creation
194 time stamp (Unix epoch).
197 Guest Operating System Configuration
198 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
200 We normally try to detect the operating system type inside the
201 container, and then modify some files inside the container to make
202 them work as expected. Here is a short list of things we do at
205 set /etc/hostname:: to set the container name
207 modify /etc/hosts:: to allow lookup of the local hostname
209 network setup:: pass the complete network setup to the container
211 configure DNS:: pass information about DNS servers
213 adapt the init system:: for example, fix the number of spawned getty processes
215 set the root password:: when creating a new container
217 rewrite ssh_host_keys:: so that each container has unique keys
219 randomize crontab:: so that cron does not start at the same time on all containers
221 Changes made by {PVE} are enclosed by comment markers:
229 Those markers will be inserted at a reasonable location in the
230 file. If such a section already exists, it will be updated in place
231 and will not be moved.
233 Modification of a file can be prevented by adding a `.pve-ignore.`
234 file for it. For instance, if the file `/etc/.pve-ignore.hosts`
235 exists then the `/etc/hosts` file will not be touched. This can be a
236 simple empty file creatd via:
238 # touch /etc/.pve-ignore.hosts
240 Most modifications are OS dependent, so they differ between different
241 distributions and versions. You can completely disable modifications
242 by manually setting the `ostype` to `unmanaged`.
244 OS type detection is done by testing for certain files inside the
247 Ubuntu:: inspect /etc/lsb-release (`DISTRIB_ID=Ubuntu`)
249 Debian:: test /etc/debian_version
251 Fedora:: test /etc/fedora-release
253 RedHat or CentOS:: test /etc/redhat-release
255 ArchLinux:: test /etc/arch-release
257 Alpine:: test /etc/alpine-release
259 Gentoo:: test /etc/gentoo-release
261 NOTE: Container start fails if the configured `ostype` differs from the auto
269 include::pct.conf.5-opts.adoc[]
272 [[pct_container_images]]
276 Container images, sometimes also referred to as ``templates'' or
277 ``appliances'', are `tar` archives which contain everything to run a
278 container. You can think of it as a tidy container backup. Like most
279 modern container toolkits, `pct` uses those images when you create a
280 new container, for example:
282 pct create 999 local:vztmpl/debian-8.0-standard_8.0-1_amd64.tar.gz
284 {pve} itself ships a set of basic templates for most common
285 operating systems, and you can download them using the `pveam` (short
286 for {pve} Appliance Manager) command line utility. You can also
287 download https://www.turnkeylinux.org/[TurnKey Linux] containers using
288 that tool (or the graphical user interface).
290 Our image repositories contain a list of available images, and there
291 is a cron job run each day to download that list. You can trigger that
292 update manually with:
296 After that you can view the list of available images using:
300 You can restrict this large list by specifying the `section` you are
301 interested in, for example basic `system` images:
303 .List available system images
305 # pveam available --section system
306 system archlinux-base_2015-24-29-1_x86_64.tar.gz
307 system centos-7-default_20160205_amd64.tar.xz
308 system debian-6.0-standard_6.0-7_amd64.tar.gz
309 system debian-7.0-standard_7.0-3_amd64.tar.gz
310 system debian-8.0-standard_8.0-1_amd64.tar.gz
311 system ubuntu-12.04-standard_12.04-1_amd64.tar.gz
312 system ubuntu-14.04-standard_14.04-1_amd64.tar.gz
313 system ubuntu-15.04-standard_15.04-1_amd64.tar.gz
314 system ubuntu-15.10-standard_15.10-1_amd64.tar.gz
317 Before you can use such a template, you need to download them into one
318 of your storages. You can simply use storage `local` for that
319 purpose. For clustered installations, it is preferred to use a shared
320 storage so that all nodes can access those images.
322 pveam download local debian-8.0-standard_8.0-1_amd64.tar.gz
324 You are now ready to create containers using that image, and you can
325 list all downloaded images on storage `local` with:
329 local:vztmpl/debian-8.0-standard_8.0-1_amd64.tar.gz 190.20MB
332 The above command shows you the full {pve} volume identifiers. They include
333 the storage name, and most other {pve} commands can use them. For
334 example you can delete that image later with:
336 pveam remove local:vztmpl/debian-8.0-standard_8.0-1_amd64.tar.gz
339 [[pct_container_storage]]
343 Traditional containers use a very simple storage model, only allowing
344 a single mount point, the root file system. This was further
345 restricted to specific file system types like `ext4` and `nfs`.
346 Additional mounts are often done by user provided scripts. This turned
347 out to be complex and error prone, so we try to avoid that now.
349 Our new LXC based container model is more flexible regarding
350 storage. First, you can have more than a single mount point. This
351 allows you to choose a suitable storage for each application. For
352 example, you can use a relatively slow (and thus cheap) storage for
353 the container root file system. Then you can use a second mount point
354 to mount a very fast, distributed storage for your database
357 The second big improvement is that you can use any storage type
358 supported by the {pve} storage library. That means that you can store
359 your containers on local `lvmthin` or `zfs`, shared `iSCSI` storage,
360 or even on distributed storage systems like `ceph`. It also enables us
361 to use advanced storage features like snapshots and clones. `vzdump`
362 can also use the snapshot feature to provide consistent container
365 Last but not least, you can also mount local devices directly, or
366 mount local directories using bind mounts. That way you can access
367 local storage inside containers with zero overhead. Such bind mounts
368 also provide an easy way to share data between different containers.
374 The root mount point is configured with the `rootfs` property, and you can
375 configure up to 10 additional mount points. The corresponding options
376 are called `mp0` to `mp9`, and they can contain the following setting:
378 include::pct-mountpoint-opts.adoc[]
380 Currently there are basically three types of mount points: storage backed
381 mount points, bind mounts and device mounts.
383 .Typical container `rootfs` configuration
385 rootfs: thin1:base-100-disk-1,size=8G
389 Storage Backed Mount Points
390 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
392 Storage backed mount points are managed by the {pve} storage subsystem and come
393 in three different flavors:
395 - Image based: these are raw images containing a single ext4 formatted file
397 - ZFS subvolumes: these are technically bind mounts, but with managed storage,
398 and thus allow resizing and snapshotting.
399 - Directories: passing `size=0` triggers a special case where instead of a raw
400 image a directory is created.
406 Bind mounts allow you to access arbitrary directories from your Proxmox VE host
407 inside a container. Some potential use cases are:
409 - Accessing your home directory in the guest
410 - Accessing an USB device directory in the guest
411 - Accessing an NFS mount from the host in the guest
413 Bind mounts are considered to not be managed by the storage subsystem, so you
414 cannot make snapshots or deal with quotas from inside the container. With
415 unprivileged containers you might run into permission problems caused by the
416 user mapping and cannot use ACLs.
418 NOTE: The contents of bind mount points are not backed up when using `vzdump`.
420 WARNING: For security reasons, bind mounts should only be established
421 using source directories especially reserved for this purpose, e.g., a
422 directory hierarchy under `/mnt/bindmounts`. Never bind mount system
423 directories like `/`, `/var` or `/etc` into a container - this poses a
426 NOTE: The bind mount source path must not contain any symlinks.
428 For example, to make the directory `/mnt/bindmounts/shared` accessible in the
429 container with ID `100` under the path `/shared`, use a configuration line like
430 `mp0: /mnt/bindmounts/shared,mp=/shared` in `/etc/pve/lxc/100.conf`.
431 Alternatively, use `pct set 100 -mp0 /mnt/bindmounts/shared,mp=/shared` to
432 achieve the same result.
438 Device mount points allow to mount block devices of the host directly into the
439 container. Similar to bind mounts, device mounts are not managed by {PVE}'s
440 storage subsystem, but the `quota` and `acl` options will be honored.
442 NOTE: Device mount points should only be used under special circumstances. In
443 most cases a storage backed mount point offers the same performance and a lot
446 NOTE: The contents of device mount points are not backed up when using `vzdump`.
452 WARNING: Because of existing issues in the Linux kernel's freezer
453 subsystem the usage of FUSE mounts inside a container is strongly
454 advised against, as containers need to be frozen for suspend or
455 snapshot mode backups.
457 If FUSE mounts cannot be replaced by other mounting mechanisms or storage
458 technologies, it is possible to establish the FUSE mount on the Proxmox host
459 and use a bind mount point to make it accessible inside the container.
462 Using Quotas Inside Containers
463 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
465 Quotas allow to set limits inside a container for the amount of disk
466 space that each user can use. This only works on ext4 image based
467 storage types and currently does not work with unprivileged
470 Activating the `quota` option causes the following mount options to be
471 used for a mount point:
472 `usrjquota=aquota.user,grpjquota=aquota.group,jqfmt=vfsv0`
474 This allows quotas to be used like you would on any other system. You
475 can initialize the `/aquota.user` and `/aquota.group` files by running
482 and edit the quotas via the `edquota` command. Refer to the documentation
483 of the distribution running inside the container for details.
485 NOTE: You need to run the above commands for every mount point by passing
486 the mount point's path instead of just `/`.
489 Using ACLs Inside Containers
490 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
492 The standard Posix **A**ccess **C**ontrol **L**ists are also available inside containers.
493 ACLs allow you to set more detailed file ownership than the traditional user/
497 [[pct_container_network]]
501 You can configure up to 10 network interfaces for a single
502 container. The corresponding options are called `net0` to `net9`, and
503 they can contain the following setting:
505 include::pct-network-opts.adoc[]
515 It is possible to use the `vzdump` tool for container backup. Please
516 refer to the `vzdump` manual page for details.
519 Restoring Container Backups
520 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
522 Restoring container backups made with `vzdump` is possible using the
523 `pct restore` command. By default, `pct restore` will attempt to restore as much
524 of the backed up container configuration as possible. It is possible to override
525 the backed up configuration by manually setting container options on the command
526 line (see the `pct` manual page for details).
528 NOTE: `pvesm extractconfig` can be used to view the backed up configuration
529 contained in a vzdump archive.
531 There are two basic restore modes, only differing by their handling of mount
535 ``Simple'' Restore Mode
536 ^^^^^^^^^^^^^^^^^^^^^^^
538 If neither the `rootfs` parameter nor any of the optional `mpX` parameters
539 are explicitly set, the mount point configuration from the backed up
540 configuration file is restored using the following steps:
542 . Extract mount points and their options from backup
543 . Create volumes for storage backed mount points (on storage provided with the
544 `storage` parameter, or default local storage if unset)
545 . Extract files from backup archive
546 . Add bind and device mount points to restored configuration (limited to root user)
548 NOTE: Since bind and device mount points are never backed up, no files are
549 restored in the last step, but only the configuration options. The assumption
550 is that such mount points are either backed up with another mechanism (e.g.,
551 NFS space that is bind mounted into many containers), or not intended to be
554 This simple mode is also used by the container restore operations in the web
558 ``Advanced'' Restore Mode
559 ^^^^^^^^^^^^^^^^^^^^^^^^^
561 By setting the `rootfs` parameter (and optionally, any combination of `mpX`
562 parameters), the `pct restore` command is automatically switched into an
563 advanced mode. This advanced mode completely ignores the `rootfs` and `mpX`
564 configuration options contained in the backup archive, and instead only
565 uses the options explicitly provided as parameters.
567 This mode allows flexible configuration of mount point settings at restore time,
570 * Set target storages, volume sizes and other options for each mount point
572 * Redistribute backed up files according to new mount point scheme
573 * Restore to device and/or bind mount points (limited to root user)
576 Managing Containers with `pct`
577 ------------------------------
579 `pct` is the tool to manage Linux Containers on {pve}. You can create
580 and destroy containers, and control execution (start, stop, migrate,
581 ...). You can use pct to set parameters in the associated config file,
582 like network configuration or memory limits.
588 Create a container based on a Debian template (provided you have
589 already downloaded the template via the web interface)
591 pct create 100 /var/lib/vz/template/cache/debian-8.0-standard_8.0-1_amd64.tar.gz
597 Start a login session via getty
601 Enter the LXC namespace and run a shell as root user
605 Display the configuration
609 Add a network interface called `eth0`, bridged to the host bridge `vmbr0`,
610 set the address and gateway, while it's running
612 pct set 100 -net0 name=eth0,bridge=vmbr0,ip=192.168.15.147/24,gw=192.168.15.1
614 Reduce the memory of the container to 512MB
616 pct set 100 -memory 512
619 Obtaining Debugging Logs
620 ~~~~~~~~~~~~~~~~~~~~~~~~
622 In case `pct start` is unable to start a specific container, it might be
623 helpful to collect debugging output by running `lxc-start` (replace `ID` with
626 lxc-start -n ID -F -l DEBUG -o /tmp/lxc-ID.log
628 This command will attempt to start the container in foreground mode, to stop the container run `pct shutdown ID` or `pct stop ID` in a second terminal.
630 The collected debug log is written to `/tmp/lxc-ID.log`.
632 NOTE: If you have changed the container's configuration since the last start
633 attempt with `pct start`, you need to run `pct start` at least once to also
634 update the configuration used by `lxc-start`.
640 `/etc/pve/lxc/<CTID>.conf`::
642 Configuration file for the container '<CTID>'.
648 * Simple, and fully integrated into {pve}. Setup looks similar to a normal
651 ** Storage (ZFS, LVM, NFS, Ceph, ...)
659 * Fast: minimal overhead, as fast as bare metal
661 * High density (perfect for idle workloads)
665 * Direct hardware access
671 * Integrated into {pve} graphical user interface (GUI)
673 * LXC (https://linuxcontainers.org/)
675 * lxcfs to provide containerized /proc file system
679 * CRIU: for live migration (planned)
681 * We use latest available kernels (4.4.X)
683 * Image based deployment (templates)
685 * Container setup from host (network, DNS, storage, ...)
689 include::pve-copyright.adoc[]