- Apiserver add a new flag --goaway-chance which is the fraction of requests that will be closed gracefully(GOAWAY) to prevent HTTP/2 clients from getting stuck on a single apiserver. 
  After the connection closed(received GOAWAY), the client's other in-flight requests won't be affected, and the client will reconnect. 

* Fixes a bug where dry-run is not honored for pod/eviction sub-resource. ([#76969](https://github.com/kubernetes/kubernetes/pull/76969), [@apelisse](https://github.com/apelisse))

- Fixes regression in logging spurious stack traces when proxied connections are closed by the backend ([#82588](https://github.com/kubernetes/kubernetes/pull/82588), [@liggitt](https://github.com/liggitt))

- github.com/bgentry/speakeasy: [v0.1.0](https://github.com/bgentry/speakeasy/tree/v0.1.0)
- github.com/bketelsen/crypt: [5cbc8cc](https://github.com/bketelsen/crypt/tree/5cbc8cc)
- github.com/fatih/color: [v1.7.0](https://github.com/fatih/color/tree/v1.7.0)
- github.com/go-gl/glfw: [e6da0ac](https://github.com/go-gl/glfw/tree/e6da0ac)
- github.com/google/martian: [v2.1.0+incompatible](https://github.com/google/martian/tree/v2.1.0)

- github.com/docker/distribution: [v2.8.2+incompatible](https://github.com/docker/distribution/tree/v2.8.2)
- github.com/fatih/color: [v1.7.0](https://github.com/fatih/color/tree/v1.7.0)
- github.com/frankban/quicktest: [v1.14.0](https://github.com/frankban/quicktest/tree/v1.14.0)
- github.com/go-gl/glfw: [e6da0ac](https://github.com/go-gl/glfw/tree/e6da0ac)

* GCE: provide an option to disable docker's live-restore on COS/ubuntu ([#55260](https://github.com/kubernetes/kubernetes/pull/55260), [@yujuhong](https://github.com/yujuhong))
* Fix kube-proxy hang due to panic when requesting a closed healthcheck. ([#46450](https://github.com/kubernetes/kubernetes/pull/46450), [@MrHohn](https://github.com/MrHohn))

- The `--leader-elect*` CLI args are now honored in scheduler. ([#105915](https://github.com/kubernetes/kubernetes/pull/105915), [@Huang-Wei](https://github.com/Huang-Wei))
- The `--leader-elect*` CLI args are now honored in the scheduler. ([#105712](https://github.com/kubernetes/kubernetes/pull/105712), [@Huang-Wei](https://github.com/Huang-Wei))

- Support more than 100 disk mounts on Windows (#105673, @andyzhangx) [SIG Storage and Windows]
- The --leader-elect* CLI args are now honored correctly in scheduler. (#106130, @Huang-Wei) [SIG Scheduling]
- The kube-proxy sync_proxy_rules_iptables_total metric now gives
  the correct number of rules, rather than being off by one.
  

- The PodDisruptionBudget `spec.unhealthyPodEvictionPolicy` field has graduated to beta and is enabled by default. On servers with the feature enabled, this field may be set to `AlwaysAllow` to always allow unhealthy pods covered by the PodDisruptionBudget to be evicted. ([#115363](https://github.com/kubernetes/kubernetes/pull/115363), [@ravisantoshgudimetla](https://github.com/ravisantoshgudimetla)) [SIG Apps, Auth and Node]

* You can now use `kubeadm token` specifying `--kubeconfig`.  In this case, kubeadm searches the current user home path and the environment variable KUBECONFIG for existing files. If provided, the `--kubeconfig` flag will be honored instead. ([#62850](https://github.com/kubernetes/kubernetes/pull/62850), [@neolit123](https://github.com/neolit123))