- kube-apiserver: `--service-account-max-token-expiration` can now be used in combination with an external token signer `--service-account-signing-endpoint`, as long as the `--service-account-max-token-expiration` is not longer than the external token signer's max expiration. ([#129816](https://github.com/kubernetes/kubernetes/pull/129816), [@sambdavidson](https://github.com/sambdavidson)) [SIG API Machinery and Auth]

- Kubeadm: during execution of the "check expiration" command, treat the etcd CA as external if there is a missing etcd CA key file (etcd/ca.key) and perform the proper validation on certificates signed by the etcd CA. Additionally, make sure that the CA for all entries in the output table is included - for both certificates on disk and...

- Fixing issue in Winkernel Proxier - Unexpected active TCP connection drops while horizontally scaling the endpoints for a LoadBalancer Service with External Traffic Policy: Local ([#114038](https://github.com/kubernetes/kubernetes/pull/114038), [@princepereira](https://github.com/princepereira)) [SIG Network]

  - [alpha] Access Review APIs expose authorization engine to external inquiries for delegation, inspection, and debugging ([docs](http://kubernetes.io/docs/admin/authorization/)) ([kubernetes/features#37](https://github.com/kubernetes/enhancements/issues/37))
- **Cluster Lifecycle**

	parentUserToCredsMap := make(map[string][]auth.Credentials)
	// DN to ldap username mapping for each LDAP user
	actualDNToLDAPUsernameMap := make(map[string]string)
	// External (actual) LDAP DN to internal normalized representation
	actualDNToParentUserMap := make(map[string]string)
	for _, cred := range allCreds {
		// Expired credentials don't need parent user updates.
		if cred.IsExpired() {

- kubeadm: produce errors if they occur when resetting cluster status for a control-plane node ([#80573](https://github.com/kubernetes/kubernetes/pull/80573), [@bart0sh](https://github.com/bart0sh))

// Health - returns current status of the object layer health,
// provides if write access exists across sets, additionally
// can be used to query scenarios if health may be lost
// if this node is taken down by an external orchestrator.
func (z *erasureServerPools) Health(ctx context.Context, opts HealthOptions) HealthResult {
	reqInfo := (&logger.ReqInfo{}).AppendTags("maintenance", strconv.FormatBool(opts.Maintenance))

- The scheduler now clears the `nominatedNodeName` field for Pods upon scheduling or binding failure. External components, such as Cluster Autoscaler and Karpenter, should not overwrite this field. ([#135007](https://github.com/kubernetes/kubernetes/pull/135007), [@ania-borowiec](https://github.com/ania-borowiec)) [SIG Scheduling and Testing]

- Fixed a regression where `kubelet --hostname-override` no longer worked
  correctly with an external cloud provider. ([#124516](https://github.com/kubernetes/kubernetes/pull/124516), [@danwinship](https://github.com/danwinship))

		// effective policy as `consoleAdmin`.
		//
		// In the latter case, we let the UI render everything, but individual
		// actions would fail if not permitted by the external authZ service.
		for _, policy := range policy.DefaultPolicies {
			if policy.Name == "consoleAdmin" {
				effectivePolicy = policy.Definition
				break
			}
		}

	case roleArn != "":