VERSION: release 1.0.0~rc93

[cyphar]

Release Notes

v1.0.0~rc93

This is the last feature-rich RC release and we are in a feature-freeze until
1.0. 1.0.0~rc94 will be released in a few weeks with minimal bug fixes only,
and 1.0.0 will be released soon afterwards.

• runc's cgroupv2 support is no longer considered experimental. It is now
  believed to be fully ready for production deployments. In addition, runc's
  cgroup code has been improved:

  • The systemd cgroup driver has been improved to be more resilient and
    handle more systemd properties correctly.
  • We now make use of openat2(2) when possible to improve the security of
    cgroup operations (in future runc will be wholesale ported to libpathrs to
    get this protection in all codepaths).

• runc's mountinfo parsing code has been reworked significantly, making
  container startup times significantly faster and less wasteful in general.

• runc now has special handling for seccomp profiles to avoid making new
  syscalls unusable for glibc. This is done by installing a custom prefix to
  all seccomp filters which returns -ENOSYS for syscalls that are newer than
  any syscall in the profile (meaning they have a larger syscall number).

  This should not cause any regressions (because previously users would simply
  get -EPERM rather than -ENOSYS, and the rule applied above is the most
  conservative rule possible) but please report any regressions you find as a
  result of this change -- in particular, programs which have special fallback
  code that is only run in the case of -EPERM.

• runc now supports the following new runtime-spec features:

  • The umask of a container can now be specified.
  • The new Linux 5.9 capabilities (CAP_PERFMON, CAP_BPF, and
    CAP_CHECKPOINT_RESTORE) are now supported.
  • The "unified" cgroup configuration option, which allows users to explicitly
    specify the limits based on the cgroup file names rather than abstracting
    them through OCI configuration. This is currently limited in scope to
    cgroupv2.

• Various rootless containers improvements:

  • runc will no longer cause conflicts if a user specifies a custom device
    which conflicts with a user-configured device -- the user device takes
    precedence.
  • runc no longer panics if /sys/fs/cgroup is missing in rootless mode.

• runc --root is now always treated as local to the current working directory.

• The --no-pivot-root hardening was improved to handle nested mounts properly
  (please note that we still strongly recommend that users do not use
  --no-pivot-root -- it is still an insecure option).

• A large number of code cleanliness and other various cleanups, including
  fairly large changes to our tests and CI to make them all run more
  efficiently.

For packagers the following changes have been made which will have impact on
your packaging of runc:

• The "selinux" and "apparmor" buildtags have been removed, and now all runc
  builds will have SELinux and AppArmor support enabled. Note that "seccomp"
  is still optional (though we very highly recommend you enable it).

• make install DESTDIR= now functions correctly.

Thanks to the following people who made this release possible:

• acetang [email protected]
• Adrian Reber [email protected]
• Akihiro Suda [email protected]
• Aleksa Sarai [email protected]
• Amim Knabben [email protected]
• An Long [email protected]
• Aos Dabbagh [email protected]
• Ashok Pon Kumar [email protected]
• Cesar Talledo [email protected]
• Chaitanya Bandi [email protected]
• Cory Bennett [email protected]
• Daniel J Walsh [email protected]
• Eduardo Vega [email protected]
• Feng Sun [email protected]
• Giuseppe Scrivano [email protected]
• Jeff Zvier [email protected]
• Kenta Tada [email protected]
• Kir Kolyshkin [email protected]
• Manabu Sugimoto [email protected]
• Mauricio Vásquez [email protected]
• Michael Crosby [email protected]
• Mrunal Patel [email protected]
• Paweł Szulik [email protected]
• Peter Hunt [email protected]
• Piotr Wagner [email protected]
• Sascha Grunert [email protected]
• SataQiu [email protected]
• Sebastiaan van Stijn [email protected]
• Shengjing Zhu [email protected]
• Shukui Yang [email protected]
• wangtianxia [email protected]
• Wei Fu [email protected]
• Xiaochen Shen [email protected]
• Xiaodong Liu [email protected]

Signed-off-by: Aleksa Sarai [email protected]

[cyphar]

https://groups.google.com/a/opencontainers.org/g/dev/c/iTHGf2eywXo is the ML thread for the vote.

[kolyshkin]

(I blatantly fixed the second-level item list markdown formatting in release notes).

The "unified" cgroup configuration option, which allows users to explicitly
specify the limits based on the cgroup file names rather than abstracting
them through OCI configuration.

Maybe mention this functionality is for cgroup v2 only.

Might also add

• improved support for systemd cgroup driver (libc/cg: convert r.CPU.Cpus/Mems to systemd props #2727, libct/cgroups/systemd: eliminate runc/systemd race #2614, libct/cg/fs/freezer: fix freezing race #2774)
• more secure cgroup drivers via use of openat2 (where available) (libcontainer/cgroups/fscommon: add openat2 support #2668) -- maybe not as it's an implementation detail
• rootless fixes and improvements (Remove runc default devices that overlap with spec devices. #2522, don't panic when /sys/fs/cgroup is missing for rootless #2634)
• notable changes for distro packagers:
  • removal of selinux/apparmor build tags (remove "apparmor" and "selinux" build tags #2709)
  • fix for make install DESTDIR= (Makefile: fix DESTDIR handling #2578)

[kolyshkin]

Maybe mention this functionality is for cgroup v2 only.

And provide a link to spec (https://github.com/opencontainers/runtime-spec/blob/master/config-linux.md#unified)

[kolyshkin]

And provide a link to spec (https://github.com/opencontainers/runtime-spec/blob/master/config-linux.md#unified)

Upon (re-)reading which I suddenly realized its scope is NOT limited to cgroup v2 like I had imagined (but nevertheless the runc implementation is).

[cyphar]

Yup, I'll add notes for those. I usually don't link directly to bugs because it's too many things to track (in future we're going to have a CHANGELOG.md so I don't have to write these by hand each time).

[kolyshkin]

LGTM (and yay! 🐶)

[cyphar]

Alright we have 4/7, but I'll wait until tomorrow before merging to give @crosbymichael @hqhq and @dqminh a chance to review as well.

[kolyshkin]

in future we're going to have a CHANGELOG.md so I don't have to write these by hand each time).

I'd say add each bug to the milestone, so we don't have to write anything at all! For example (this is obviously incomplete now): v1.0.0-rc93 issues.

[cyphar]

I still think it's better to have a changelog file which has a user-understandable description so that PR authors do the writing and I can just copy the section -- the goal of the release notes is to be more understandable than a list of issue and PR titles (which are usually not very helpful, even to me). This is how I do umoci releases.