### Bug or Regression

- Fix a bug where the kubelet ephemerally fails with `failed to initialize top level QOS containers: root container [kubepods] doesn't exist`, due to the cpuset cgroup being deleted on v2 with systemd cgroup manager. ([#125923](https://github.com/kubernetes/kubernetes/pull/125923), [@haircommander](https://github.com/haircommander)) [SIG Node and Testing]

*   Problems deleting local static pods have been resolved. ([#48339](https://github.com/kubernetes/kubernetes/pull/48339),[ @dixudx](https://github.com/dixudx))
*   CRI now only calls UpdateContainerResources when cpuset is set. ([#53122](https://github.com/kubernetes/kubernetes/pull/53122),[ @resouer](https://github.com/resouer))

## Changelog since v1.21.0-beta.1

## Urgent Upgrade Notes

### (No, really, you MUST read this before you upgrade)

 - Migrated pkg/kubelet/cm/cpuset/cpuset.go to structured logging. Exit code changed from 255 to 1. ([#100007](https://github.com/kubernetes/kubernetes/pull/100007), [@utsavoza](https://github.com/utsavoza)) [SIG Instrumentation and Node]
 
## Changes by Kind

- Output for `kubectl describe podsecuritypolicy` is now kind specific and cleaner ([#101436](https://github.com/kubernetes/kubernetes/pull/101436), [@KnVerey](https://github.com/KnVerey))
- Parsing of cpuset information now properly detects more invalid input such as `1--3` or `10-6`. ([#100565](https://github.com/kubernetes/kubernetes/pull/100565), [@lack](https://github.com/lack))