- Fixes...

			},
			{
				Description: MetricDescription{
					Namespace: nodeMetricNamespace,
					Subsystem: scannerSubsystem,
					Name:      "bucket_scans_started",
					Help:      "Total number of bucket scans started since server uptime",
					Type:      counterMetric,
				},
				Value: float64(globalScannerMetrics.lifetime(scannerMetricScanBucketDrive) + uint64(globalScannerMetrics.activeDrives())),
			},
			{

- Performance improvements in `klog` ([#115277](https://github.com/kubernetes/kubernetes/pull/115277), [@pohly](https://github.com/pohly))
- Pod template `schedulingGates` are now mutable for Jobs that are suspended and have never been started ([#115940](https://github.com/kubernetes/kubernetes/pull/115940), [@ahg-g](https://github.com/ahg-g)) [SIG Apps]

	if tgt == nil {
		replLogIf(ctx, fmt.Errorf("replication resync failed for %s - target could not be created for arn %s", opts.bucket, opts.arn))
		return
	}
	// mark resync status as resync started
	if !heal {
		s.markStatus(ResyncStarted, opts, objectAPI)
	}

	// Walk through all object versions - Walk() is always in ascending order needed to ensure

* Allow access to ClusterIP from the host network namespace when kube-proxy is started in IPVS mode without either masqueradeAll or clusterCIDR flags ([#65388](https://github.com/kubernetes/kubernetes/pull/65388), [@lbernail](https://github.com/lbernail))

- Fix the bug which could result in Job status updates failing with the error:
  status.startTime: Required value: startTime cannot be removed for unsuspended job
  The error could be raised after a Job is resumed, if started and suspended previously. ([#135129](https://github.com/kubernetes/kubernetes/pull/135129), [@dejanzele](https://github.com/dejanzele)) [SIG Apps and Testing]

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. ([#108142](https://github.com/kubernetes/kubernetes/pull/108142), [@liggitt](https://github.com/liggitt))...

  Once core Kubernetes support for user namespaces is generally available and kubeadm has started to support running the control plane in userns pods, the kubeadm `RootlessControlPlane` feature gate will be removed entirely.

* rkt version 1.23.0+

    * known issues with the rkt runtime are [listed in the Getting Started Guide](https://kubernetes.io/docs/getting-started-guides/rkt/notes/)

* etcd version 3.0.17

* Go version: 1.8.3. [Link to announcement](https://groups.google.com/d/msg/kubernetes-dev/0XRRz6UhhTM/YODWVnuDBQAJ)

* Ensure cloudprovider.InstanceNotFound is reported when the VM is not found on Azure ([#61531](https://github.com/kubernetes/kubernetes/pull/61531), [@feiskyer](https://github.com/feiskyer))
* On AWS kubelet returns an error when started under conditions that do not allow it to work (AWS has not yet tagged the instance). ([#60125](https://github.com/kubernetes/kubernetes/pull/60125), [@vainu-arto](https://github.com/vainu-arto))