- Kubelet should now more reliably report the same primary node IP even if the set of node IPs reported by the CloudProvider changes. ([#79391](https://github.com/kubernetes/kubernetes/pull/79391), [@danwinship](https://github.com/danwinship))

- QueueingHint implementation for NodeAffinity is reverted because we found potential scenarios where events that make Pods schedulable could be missed. ([#122327](https://github.com/kubernetes/kubernetes/pull/122327), [@sanposhiho](https://github.com/sanposhiho)) [SIG Scheduling]

regression that could incorrectly reject pods with OutOfCpu errors if they were rapidly scheduled after other pods were reported as complete in the API. The Kubelet now waits to report the phase of a pod as terminal in the API until all running containers are guaranteed to have stopped and no new containers can be started.  Short-lived pods may take slightly longer (~1s) to report Succeeded or Failed after this change. ([#108723](https://github.com/kubernetes/kubernetes/pull/108723), [@bobbypage](...

regression that could incorrectly reject pods with OutOfCpu errors if they were rapidly scheduled after other pods were reported as complete in the API. The Kubelet now waits to report the phase of a pod as terminal in the API until all running containers are guaranteed to have stopped and no new containers can be started.  Short-lived pods may take slightly longer (~1s) to report Succeeded or Failed after this change. ([#108749](https://github.com/kubernetes/kubernetes/pull/108749), [@bobbypage](...

- Fix detach azure disk issue when vm not exist ([#95177](https://github.com/kubernetes/kubernetes/pull/95177), [@andyzhangx](https://github.com/andyzhangx)) [SIG Cloud Provider]
- Fix etcd_object_counts metric reported by kube-apiserver ([#94817](https://github.com/kubernetes/kubernetes/pull/94817), [@tkashem](https://github.com/tkashem)) [SIG API Machinery]

 and *_admission_latencies_milliseconds metrics due to reporting wrong unit (was labelled milliseconds, but reported seconds), and multiple naming guideline violations (units should be in base units and "duration" is the best practice labelling to measure the time a request takes). Please convert to use *_admission_duration_seconds and *_admission_duration_seconds_summary, these now report the unit as described, and follow the instrumentation best practices. ([#75279](https://github.com/kubernet...

* Fixed kube-proxy to work correctly with iptables 1.6.2 and later. ([#61734](https://github.com/kubernetes/kubernetes/pull/61734), [@danwinship](https://github.com/danwinship))
* 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))

- github.com/json-iterator/go: [v1.1.8 → v1.1.10](https://github.com/json-iterator/go/compare/v1.1.8...v1.1.10)
- github.com/jstemmer/go-junit-report: [af01ea7 → v0.9.1](https://github.com/jstemmer/go-junit-report/compare/af01ea7...v0.9.1)
- github.com/konsorten/go-windows-terminal-sequences: [v1.0.1 → v1.0.3](https://github.com/konsorten/go-windows-terminal-sequences/compare/v1.0.1...v1.0.3)

	// uncompressed expected content.
	// If content is very large, an alternative to setting Content or File
	// is to set Size, which will then be checked against the header-reported size
	// but will bypass the decompressing of the actual data.
	// This last option is used for testing very large (multi-GB) compressed files.
	ContentErr error
	Content    []byte
	File       string
	Size       uint64

* Kubelet will now set extended resource capacity to zero after it restarts. If the extended resource is exported by a device plugin, its capacity will change to a valid value after the device plugin re-connects with the Kubelet. If the extended resource is exported by an external component through direct node status capacity patching, the component should repatch the field after kubelet becomes ready again. During the time gap, pods...