Now, when unschedulable Pods are updated, the scheduling queue checks with QueueingHint(s) whether the update may make the pods schedulable, and requeues them to activeQ/backoffQ **only when** at least one QueueingHint(s) return Queue. 
  
  Action required for custom scheduler plugin developers:

    // Connection is pooled!
    assertThat(server.takeRequest().exchangeIndex).isEqualTo(1)
  }

  @Test
  fun streamDiscardingIsTimely() {
    // This response takes at least a full second to serve: 10,000 bytes served 100 bytes at a time.
    server.enqueue(
      MockResponse
        .Builder()
        .body(Buffer().write(ByteArray(10000)))

* Updates Fluentd to 1.3.2 & added filter_parser  ([#71180](https://github.com/kubernetes/kubernetes/pull/71180), [@monotek](https://github.com/monotek))
* The leaderelection package allows the lease holder to release its lease when the calling context is cancelled. This allows ([#71490](https://github.com/kubernetes/kubernetes/pull/71490), [@smarterclayton](https://github.com/smarterclayton))

`Initializers` admission plugin is enabled.([#51436](https://github.com/kubernetes/kubernetes/pull/51436))

* [Action required] The validation rule for metadata.initializers.pending[x].name is tightened. The initializer name must contain at least three segments, separated by dots. You can create objects with pending initializers and not rely on the API server to add pending initializers according to `initializerConfiguration`. If you do so, update the initializer name in the existing objects...

- In the `wait.PollUntilContextTimeout` function, if `immediate` is true, the condition will now be invoked before waiting, guaranteeing that the condition is invoked at least once and then wait a interval before executing again. ([#119762](https://github.com/kubernetes/kubernetes/pull/119762), [@AxeZhan](https://github.com/AxeZhan))

You can still access kubelet's `/healthz`, `/pods` and `/configz` by granting the caller `nodes/proxy` permission in RBAC but that also grants the caller permissions to exec, run and attach to containers on the nodes and doing so does not follow the least privilege principle. Granting callers more permissions than they need can give attackers an opportunity to escalate privileges. ([#126347](https://github.com/kubernetes/kubernetes/pull/126347), [@vinayakankugoyal](https://github.com/vinayakankugoyal))...

- `kubectl exec` now allows using the resource name to select a matching pod and `--pod-running-timeout` flag to wait till at least one pod is running. ([#73664](https://github.com/kubernetes/kubernetes/pull/73664), [@prksu](https://github.com/prksu))

- The `Should recreate evicted statefulset` conformance [test]( https://github.com/kubernetes/kubernetes/blob/master/test/e2e/apps/statefulset.go) fails because `Pod ss-0 expected to be re-created at least once`. This was caused by the `Predicate PodFitsHostPorts failed` scheduling error. The root cause was a host port conflict for port `21017`. This port was in-use as an ephemeral port by another application running on the node. This will be looked...

  - apiserver_flowcontrol_current_r: R(the time of the last change in state of the queues)
  - apiserver_flowcontrol_dispatch_r: R(the time of the latest request dispatch)
  - apiserver_flowcontrol_latest_s: S(the request last dispatched) = R(when that request starts executing in the virtual world)

   ([#121954](https://github.com/kubernetes/kubernetes/pull/121954), [@AxeZhan](https://github.com/AxeZhan))
- The watch cache now waits until it is at least as fresh as the given requestedWatchRV if sendInitialEvents was requested.
   ([#122830](https://github.com/kubernetes/kubernetes/pull/122830), [@p0lyn0mial](https://github.com/p0lyn0mial))