* `list`/`watch` API requests with a `fieldSelector` that specifies `metadata.name` can now be authorized as requests for an individual named resource ([#63469](https://github.com/kubernetes/kubernetes/pull/63469), [@wojtek-t](https://github.com/wojtek-t))

* Add `metrics-port` to kube-proxy cmd flags. ([#72682](https://github.com/kubernetes/kubernetes/pull/72682), [@whypro](https://github.com/whypro))

description: Provide supporting details for a feature in development
labels: kind/feature
body:
  - type: textarea
    id: feature
    attributes:
      label: What would you like to be added?
      description: |
        Feature requests are unlikely to make progress as issues. Please consider engaging with SIGs on slack and mailing lists, instead.
        A proposal that works through the design along with the implications of the change can be opened as a KEP.

## Changes by Kind

### API Change

  - `priority_level_request_count_samples` is replaced with `priority_level_request_utilization`, which samples every nanosecond rather than every millisecond; the old metric conveyed utilization despite its name.
  - `priority_level_request_count_watermarks` is removed.

* The behavior of the default handler for 404 requests fro the GCE Ingress load balancer is slightly modified in the sense that it now exports metrics using prometheus. The metrics exported include:  ([#79106](https://github.com/kubernetes/kubernetes/pull/79106), [@vbannai](https://github.com/vbannai))
    * - http_404_request_total  (the number of 404 requests handled)
    * - http_404_request_duration_ms (the amount of time the server took to respond in ms)

- If the user specifies an invalid timeout in the request URL, the request will be aborted with an HTTP 400.

  - 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)
  - apiserver_flowcontrol_next_s_bounds: min and max next S among non-empty queues

- Fixes a regression in v1beta1 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...

- Apiserver add a new flag --goaway-chance which is the fraction of requests that will be closed gracefully(GOAWAY) to prevent HTTP/2 clients from getting stuck on a single apiserver. 
  After the connection closed(received GOAWAY), the client's other in-flight requests won't be affected, and the client will reconnect. 
  The flag min value is 0 (off), max is .02 (1/50 requests); .001 (1/1000) is a recommended starting point.