2. CacheSize type was changed from int32 to *int32. This allows defaulting logic to differentiate between cases where users explicitly supplied 0 vs. not supplied any value.
   3. KMS Provider's endpoint (path to Unix socket) is now validated when the EncryptionConfiguration files is loaded. This used to be handled by the GRPCService.

### Other notable changes

- Gives the right error message when using `kubectl delete` a wrong resource. ([#83825](https://github.com/kubernetes/kubernetes/pull/83825), [@zhouya0](https://github.com/zhouya0))

  - `v1alpha3` has split apart `MasterConfiguration` into separate components; `InitConfiguration`, `ClusterConfiguration`, `JoinConfiguration`, `KubeletConfiguration`, and `KubeProxyConfiguration`
  - Different configuration types can be supplied all in the same file separated by `---`.
- Improved CRI handling
  - crictl is no longer required in docker-only setups.
  - Better detection of installed CRI.
  - Better output for image pull errors.

[Auditing](https://kubernetes.io/docs/tasks/debug/debug-cluster/audit/) is now enabled behind a featureGate in kubeadm. A user can supply their own audit policy with configuration option as well as a place for the audit logs to live. If no policy is supplied a default policy will be provided. The default policy will log all Metadata level policy logs. It is the example provided in the documentation. ([#59067](https://github.com/kubernetes/kubernetes/pull/59067), [@chuckha](https://github.com/chuckha))...

* The podGC controller will now always run, irrespective of the value supplied to the "terminated-pod-gc-threshold" flag supplied to the controller manager.  ([#35476](https://github.com/kubernetes/kubernetes/pull/35476), [@foxish](https://github.com/foxish))

- Kubeadm now distinguishes between generated and user supplied component configs, regenerating the former ones if a config upgrade is required ([#86070](https://github.com/kubernetes/kubernetes/pull/86070), [@rosti](https://github.com/rosti)) [SIG Cluster Lifecycle]

- Fix a regression where kubeadm bails out with a fatal error when an optional version command line argument is supplied to the "kubeadm upgrade plan" command ([#94421](https://github.com/kubernetes/kubernetes/pull/94421), [@rosti](https://github.com/rosti)) [SIG Cluster Lifecycle]

- Enhanced cloud provider integrations to support optional, per-Node custom labels that can be supplied and applied to Nodes by the node controller. These extra labels will only be applied where the cloud provider integration supports this feature.