runc v1.5.0-rc.1 -- "憎しみを束ねてもそれは脆い！"
This is the first release candidate of the runc 1.5.0 release. It
contains a couple of new features, but is mostly made up of various
cleanups (such as the removal of many deprecated APIs) and improvements.
runc v1.5.0-rc.1 includes all of the patches backported to runc v1.4.1.
Users are strongly encouraged to test our release candidates over the
next two months so we can fix issues before the general release. You
should expect runc 1.5.0 to be released at the end of April 2026 (at
which point, runc 1.3.z will only receive high-severity security fixes
for 6 months and runc 1.2.z will become unmaintained -- users are thus
very strongly encouraged to migrate to a newer version).
libcontainer API:
- The following deprecated Go APIs have been removed:
- "CleanPath", "StripRoot", and "WithProcfd" from
"libcontainer/utils". Note that "WithProcfdFile" has not been
removed (due to import cycle issues) but is instead marked as
internal in its godoc comment. (#5051)
- All of the cgroup-related types and functions from
"libcontainer/configs" which are now maintained in
"github.com/opencontainers/cgroups" (#5141):
- "libcontainer/configs.Cgroup"
- "libcontainer/configs.Resources"
- "libcontainer/configs.FreezerState"
- "libcontainer/configs.LinuxRdma"
- "libcontainer/configs.BlockIODevice"
- "libcontainer/configs.WeightDevice"
- "libcontainer/configs.ThrottleDevice"
- "libcontainer/configs.HugepageLimit"
- "libcontainer/configs.IfPrioMap"
- "libcontainer/configs.Undefined"
- "libcontainer/configs.Frozen"
- "libcontainer/configs.Thawed"
- "libcontainer/configs.NewWeightDevice"
- "libcontainer/configs.NewThrottleDevice"
- "libcontainer/configs.HookList.RunHooks". (#5141)
- "libcontainer/configs.MPOL_*" (#5141)
- All of the types in "libcontainer/devices" which are now maintained
in "github.com/opencontainers/cgroups/devices/config" (#5141):
- "libcontainer/devices.Wildcard"
- "libcontainer/devices.WildcardDevice"
- "libcontainer/devices.BlockDevice"
- "libcontainer/devices.CharDevice"
- "libcontainer/devices.FifoDevice"
- "libcontainer/devices.Device"
- "libcontainer/devices.Permissions"
- "libcontainer/devices.Type"
- "libcontainer/devices.Rule"
- "libcontainer.Process" methods ("Wait", "Pid", "Signal") and
"libcontainer/configs.Config" methods ("HostUID", "HostRootUID",
"HostGID", "HostRootGID") now use pointer receivers. (#5088)
- The example code for libcontainer has been moved out of a README and
into a proper Example* test file that will be compile-tested by our
CI. As mentioned elsewhere, we still *do not* recommend users make use
of the libcontainer API directly. (#5127)
Deprecated:
- The "libcontainer/configs.Mount.Relabel" configuration field (used to
relabel mounts with the "z" and "Z" "pseudo" mount options) was never
accessible outside of the libcontainer API, and in practice the
relabel logic has always lived in higher level runtimes. It has been
made into a no-op and the field will be removed entirely in runc 1.7.
(#5152, #5160)
Removed:
- The memfd-bind helper binary has been removed, as it has never been
particularly useful and was completely obsoleted by the changes to
/proc/self/exe sealing we introduced in runc 1.2.0. (#5141)
Added:
- User-namespaced containers can now configure user.* sysctls. (#4889)
- Preliminary loong64 support. (#4938)
- Intel RDT: the RDT subdirectory is now only removed if runc created
it, matching the updated runtime-spec guidance. (#3832, #5155)
Changed:
- Our release binaries and default build configuration now use libpathrs
by default, providing better hardening against certain kinds of
attacks. Users of runc should not see any changes as a result of this,
but packagers will need to adjust their packaging accordingly. runc can
still be built without libpathrs (by building without the libpathrs
build tag), but we currently plan to make runc 1.6 *require*
libpathrs. (#5103)
- "runc exec" will now request systemd to move the "exec" process into
the container cgroup, making the procedure more rootless-friendly.
(#4822)
- seccomp: minor documentation updates. (#4902)
- Errors from "runc init" have historically been quite painful to
understand and debug, we have made several improvements to make them
more comprehensive and thus useful when debugging issues. (#4951,
#4928)
- Update spec conformance documentation for OCI runtime-spec v1.3.0.
(#4948, #5150)
- Our release archives now have the name "runc-$version.tar.xz" to make
distro packaging a little easier by matching the filename to the
top-level directory name in the archive. (#5052)
Thanks to the following contributors for making this release possible:
* Akihiro Suda <akihiro.suda.cz@hco.ntt.co.jp>
* Aleksa Sarai <aleksa@amutable.com>
* Antti Kervinen <antti.kervinen@intel.com>
* Ariel Otilibili <otilibil@eurecom.fr>
* Arina Cherednik <arinacherednik034@gmail.com>
* Curd Becker <me@curd-becker.de>
* Dimitri John Ledkov <dimitri.ledkov@surgut.co.uk>
* Donet Tom <donettom@linux.ibm.com>
* Efim Verzakov <efimverzakov@gmail.com>
* Ismo Puustinen <ismo.puustinen@intel.com>
* Joshua Rogers <MegaManSec@users.noreply.github.com>
* Kir Kolyshkin <kolyshkin@gmail.com>
* Lei Wang <ssst0n3@gmail.com>
* Li Fubang <lifubang@acmcoder.com>
* Luke Hinds <luke@stacklok.com>
* Markus Lehtonen <markus.lehtonen@intel.com>
* Osama Abdelkader <osama.abdelkader@gmail.com>
* Phil Estes <estesp@gmail.com>
* Ricardo Branco <rbranco@suse.de>
* Rodrigo Campos Catelin <rodrigo@amutable.com>
* Tianon Gravi <admwiggin@gmail.com>
* Tycho Andersen <tycho@tycho.pizza>
* Tõnis Tiigi <tonistiigi@gmail.com>
* Vishal Chourasia <vishalc@linux.ibm.com>
* zhaixiaojuan <zhaixiaojuan@loongson.cn>
Signed-off-by: Aleksa Sarai <cyphar@cyphar.com>
VERSION: release v1.5.0-rc.1 Signed-off-by: Aleksa Sarai <cyphar@cyphar.com>
runc is a CLI tool for spawning and running containers on Linux according to the OCI specification.
You can find official releases of runc on the release page.
All releases are signed by one of the keys listed in the runc.keyring file in the root of this repository.
The reporting process and disclosure communications are outlined here.
A third party security audit was performed by Cure53, you can see the full report here.
runc only supports Linux. See the header of go.mod for the minimally required Go version.
In addition to Go, building runc requires multiple utilities and libraries to be installed on your system.
On Ubuntu/Debian, you can install the required dependencies with:
apt update && apt install -y make gcc linux-libc-dev libseccomp-dev pkg-config git
On CentOS/Fedora, you can install the required dependencies with:
yum install -y make gcc kernel-headers libseccomp-devel pkg-config git
On Alpine Linux, you can install the required dependencies with:
apk --update add bash make gcc libseccomp-dev musl-dev linux-headers git
The following dependencies are optional:
libseccomp - only required if you enable seccomp support; to disable, see Build Tags# create a 'github.com/opencontainers' in your GOPATH/src cd github.com/opencontainers git clone https://github.com/opencontainers/runc cd runc make sudo make install
You can also use go get to install to your GOPATH, assuming that you have a github.com parent folder already created under src:
go get github.com/opencontainers/runc cd $GOPATH/src/github.com/opencontainers/runc make sudo make install
runc will be installed to /usr/local/sbin/runc on your system.
You can see the runc version by running runc --version. You can append a custom string to the version using the EXTRA_VERSION make variable when building, e.g.:
make EXTRA_VERSION="+build-1"
Bear in mind to include some separator for readability.
runc supports optional build tags for compiling support of various features, with some of them enabled by default (see BUILDTAGS in top-level Makefile).
To change build tags from the default, set the BUILDTAGS variable for make, e.g. to disable seccomp:
make BUILDTAGS=""
To add some more build tags to the default set, use the EXTRA_BUILDTAGS make variable, e.g. to disable checkpoint/restore:
make EXTRA_BUILDTAGS="runc_nocriu"
| Build Tag | Feature | Enabled by Default | Dependencies |
|---|---|---|---|
seccomp | Syscall filtering using libseccomp. | yes | libseccomp |
libpathrs | Use libpathrs for path safety. | yes | libpathrs |
runc_nocriu | Disables runc checkpoint/restore. | no | criu |
The following build tags were used earlier, but are now obsoleted:
runc currently supports running its test suite via Docker. To run the suite just type make test.
make test
There are additional make targets for running the tests outside of a container but this is not recommended as the tests are written with the expectation that they can write and remove anywhere.
You can run a specific test case by setting the TESTFLAGS variable.
# make test TESTFLAGS="-run=SomeTestFunction"
You can run a specific integration test by setting the TESTPATH variable.
# make test TESTPATH="/checkpoint.bats"
You can run a specific rootless integration test by setting the ROOTLESS_TESTPATH variable.
# make test ROOTLESS_TESTPATH="/checkpoint.bats"
You can run a test using your container engine's flags by setting CONTAINER_ENGINE_BUILD_FLAGS and CONTAINER_ENGINE_RUN_FLAGS variables.
# make test CONTAINER_ENGINE_BUILD_FLAGS="--build-arg http_proxy=http://yourproxy/" CONTAINER_ENGINE_RUN_FLAGS="-e http_proxy=http://yourproxy/"
runc uses Go Modules for dependencies management. Please refer to Go Modules for how to add or update new dependencies.
# Update vendored dependencies make vendor # Verify all dependencies make verify-dependencies
Please note that runc is a low level tool not designed with an end user in mind. It is mostly employed by other higher level container software.
Therefore, unless there is some specific use case that prevents the use of tools like Docker or Podman, it is not recommended to use runc directly.
If you still want to use runc, here's how.
In order to use runc you must have your container in the format of an OCI bundle. If you have Docker installed you can use its export method to acquire a root filesystem from an existing Docker container.
# create the top most bundle directory mkdir /mycontainer cd /mycontainer # create the rootfs directory mkdir rootfs # export busybox via Docker into the rootfs directory docker export $(docker create busybox) | tar -C rootfs -xvf -
After a root filesystem is populated you just generate a spec in the format of a config.json file inside your bundle. runc provides a spec command to generate a base template spec that you are then able to edit. To find features and documentation for fields in the spec please refer to the specs repository.
runc spec
Assuming you have an OCI bundle from the previous step you can execute the container in two different ways.
The first way is to use the convenience command run that will handle creating, starting, and deleting the container after it exits.
# run as root cd /mycontainer runc run mycontainerid
If you used the unmodified runc spec template this should give you a sh session inside the container.
The second way to start a container is using the specs lifecycle operations. This gives you more power over how the container is created and managed while it is running. This will also launch the container in the background so you will have to edit the config.json to remove the terminal setting for the simple examples below (see more details about runc terminal handling). Your process field in the config.json should look like this below with "terminal": false and "args": ["sleep", "5"].
"process": { "terminal": false, "user": { "uid": 0, "gid": 0 }, "args": [ "sleep", "5" ], "env": [ "PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin", "TERM=xterm" ], "cwd": "/", "capabilities": { "bounding": [ "CAP_AUDIT_WRITE", "CAP_KILL", "CAP_NET_BIND_SERVICE" ], "effective": [ "CAP_AUDIT_WRITE", "CAP_KILL", "CAP_NET_BIND_SERVICE" ], "inheritable": [ "CAP_AUDIT_WRITE", "CAP_KILL", "CAP_NET_BIND_SERVICE" ], "permitted": [ "CAP_AUDIT_WRITE", "CAP_KILL", "CAP_NET_BIND_SERVICE" ], "ambient": [ "CAP_AUDIT_WRITE", "CAP_KILL", "CAP_NET_BIND_SERVICE" ] }, "rlimits": [ { "type": "RLIMIT_NOFILE", "hard": 1024, "soft": 1024 } ], "noNewPrivileges": true },
Now we can go through the lifecycle operations in your shell.
# run as root cd /mycontainer runc create mycontainerid # view the container is created and in the "created" state runc list # start the process inside the container runc start mycontainerid # after 5 seconds view that the container has exited and is now in the stopped state runc list # now delete the container runc delete mycontainerid
This allows higher level systems to augment the containers creation logic with setup of various settings after the container is created and/or before it is deleted. For example, the container's network stack is commonly set up after create but before start.
runc has the ability to run containers without root privileges. This is called rootless. You need to pass some parameters to runc in order to run rootless containers. See below and compare with the previous version.
Note: In order to use this feature, “User Namespaces” must be compiled and enabled in your kernel. There are various ways to do this depending on your distribution:
CONFIG_USER_NS=y is set in your kernel configuration (normally found in /proc/config.gz)echo 1 > /proc/sys/kernel/unprivileged_userns_cloneecho 28633 > /proc/sys/user/max_user_namespacesRun the following commands as an ordinary user:
# Same as the first example mkdir ~/mycontainer cd ~/mycontainer mkdir rootfs docker export $(docker create busybox) | tar -C rootfs -xvf - # The --rootless parameter instructs runc spec to generate a configuration for a rootless container, which will allow you to run the container as a non-root user. runc spec --rootless # The --root parameter tells runc where to store the container state. It must be writable by the user. runc --root /tmp/runc run mycontainerid
runc can be used with process supervisors and init systems to ensure that containers are restarted when they exit. An example systemd unit file looks something like this.
[Unit] Description=Start My Container [Service] Type=forking ExecStart=/usr/local/sbin/runc run -d --pid-file /run/mycontainerid.pid mycontainerid ExecStopPost=/usr/local/sbin/runc delete mycontainerid WorkingDirectory=/mycontainer PIDFile=/run/mycontainerid.pid [Install] WantedBy=multi-user.target
The code and docs are released under the Apache 2.0 license.