* apiserver_request_total
  * apiserver_request_duration_seconds  
New "APPLY" value for the "verb" metric label which indicates a PATCH with "Content-Type: apply-patch+yaml". This value is experimental and will only be present if the ServerSideApply alpha feature is enabled. ([#74997](https://github.com/kubernetes/kubernetes/pull/74997), [@jennybuckley](https://github.com/jennybuckley))

      name: Age
      type: date
    name: v1
    schema:
      openAPIV3Schema:
        properties:
          spec:
            description: 'Configuration affecting label/content routing, sni routing,
              etc. See more details at: https://istio.io/docs/reference/config/networking/virtual-service.html'
            properties:
              exportTo:

### API Change

	cache := store.lock()
	defer store.unlock()

	// Handle policy mapping removal.
	if policy == "" {
		if store.getUsersSysType() == LDAPUsersSysType {
			// Add a fallback removal towards previous content that may come back
			// as a ghost user due to lack of delete, this change occurred
			// introduced in PR #11840
			store.deleteMappedPolicy(ctx, name, regUser, false)
		}

- Resolves a spurious "Unknown discovery response content-type" error in client-go discovery requests by tolerating extra content-type parameters in API responses ([#117571](https://github.com/kubernetes/kubernetes/pull/117571), [@seans3](https://github.com/seans3)) [SIG API Machinery]

- Fixes a regression in v1beta1 PodDisruptionBudget handling of `strategic merge patch`-type API requests for the `selector` field. Prior to 1.21, these requests would merge `matchLabels` content and replace `matchExpressions` content. In 1.21, patch requests touching the `selector` field started replacing the entire selector. This is consistent with server-side apply and the v1 PodDisruptionBudget behavior, but should not have been changed for v1beta1. ([#...

	   annotation, you should delete any existing NetworkPolicy
	   objects. These would have had no effect before, but with v1
	   semantics they might cause some traffic to be blocked that you
	   didn't intend to be blocked.

### Other notable changes

* Added exponential backoff to Azure cloudprovider ([#46660](https://github.com/kubernetes/kubernetes/pull/46660), [@jackfrancis](https://github.com/jackfrancis))

### Other notable changes

* Fix panic create/update CRD when mutating/validating webhook configured. ([#61404](https://github.com/kubernetes/kubernetes/pull/61404), [@hzxuzhonghu](https://github.com/hzxuzhonghu))
* Add support to ingest log entries to Stackdriver against new "k8s_container" and "k8s_node" resources. ([#62076](https://github.com/kubernetes/kubernetes/pull/62076), [@qingling128](https://github.com/qingling128))

- Client-ca bundles for the all generic-apiserver based servers will dynamically reload from disk on content changes ([#83579](https://github.com/kubernetes/kubernetes/pull/83579), [@deads2k](https://github.com/deads2k))

* When performing a GET then PUT, the kube-apiserver must write the canonical representation of the object to etcd if the current value does not match. That allows external agents to migrate content in etcd from one API version to another, across different storage types, or across varying encryption levels. This fixes a bug introduced in 1.5 where we unintentionally stopped writing the newest data. ([#48394](https://github.com/kubernetes/kub...