* update gophercloud/gophercloud dependency for reauthentication fixes ([#45545](https://github.com/kubernetes/kubernetes/pull/45545), [@stuart-warren](https://github.com/stuart-warren))
* fix sync loop health check with separating runtime errors ([#47124](https://github.com/kubernetes/kubernetes/pull/47124), [@andyxning](https://github.com/andyxning))

* update gophercloud/gophercloud dependency for reauthentication fixes ([#45545](https://github.com/kubernetes/kubernetes/pull/45545), [@stuart-warren](https://github.com/stuart-warren))
* fix sync loop health check with separating runtime errors ([#47124](https://github.com/kubernetes/kubernetes/pull/47124), [@andyxning](https://github.com/andyxning))

- Fixed bug that kubelet resource metric `container_start_time_seconds` had timestamp equal to container start time. ([#120518](https://github.com/kubernetes/kubernetes/pull/120518), [@saschagrunert](https://github.com/saschagrunert)) [SIG Instrumentation, Node and Testing]

- `kubelet_pod_worker_latency_microseconds` -> `kubelet_pod_worker_duration_seconds`
- `kubelet_pod_start_latency_microseconds` -> `kubelet_pod_start_duration_seconds`
- `kubelet_cgroup_manager_latency_microseconds` -> `kubelet_cgroup_manager_duration_seconds`
- `kubelet_pod_worker_start_latency_microseconds` -> `kubelet_pod_worker_start_duration_seconds`
- `kubelet_pleg_relist_latency_microseconds` -> `kubelet_pleg_relist_duration_seconds`

  // If this value is in the future, no logs will be returned.
  // Only one of sinceSeconds or sinceTime may be specified.
  // +optional
  optional int64 sinceSeconds = 4;

  // An RFC3339 timestamp from which to show logs. If this value
  // precedes the time a pod was started, only logs since the pod start will be returned.

  [overwrite of manual edits to GCP Health Checks](https://github.com/kubernetes/ingress/issues/842)
  managed by the GLBC Ingress Controller. This can cause the health checks to start failing,
  requiring you to reapply the manual edits.
  * This issue does not affect clusters that start out with Kubernetes v1.6.4 or higher.
  * This issue does not affect Health Checks that were left in the configuration

- Failure to start a container cannot accidentally result in the pod being considered "Succeeded" in the presence of deletion. ([#108882](https://github.com/kubernetes/kubernetes/pull/108882), [@rphillips](https://github.com/rphillips)) [SIG Node]

- Expose `/debug/flags/v` to allow dynamically setting log level for kube-proxy. ([#98306](https://github.com/kubernetes/kubernetes/pull/98306), [@borgerli](https://github.com/borgerli)) [SIG Network]
- Expose container start time as `container_start_time_seconds` in the kubelet `/metrics/resource` endpoint. ([#102444](https://github.com/kubernetes/kubernetes/pull/102444), [@sanwishe](https://github.com/sanwishe))

- Sped up kubelet start time by executing an immediate runtime and node status update when the Kubelet sees that it has a CIDR. ([#67031](https://github.com/kubernetes/kubernetes/pull/67031), [@krzysztof-jastrzebski](https://github.com/krzysztof-jastrzebski))

persistent storage is available everywhere in the cluster and has infinite capacity. Topology constraints addressed the first point, but up to now pod scheduling was still done without considering that the remaining storage capacity may not be enough to start a new pod. [Storage capacity tracking](https://docs.k8s.io/concepts/storage/storage-capacity/) addresses that by adding an API for a CSI driver to report storage capacity and uses that information in the Kubernetes scheduler when choosing a node for...