- Apiserver repair controller is resilient to etcd errors during bootstrap and retries during 30 seconds before failing. ([#126671](https://github.com/kubernetes/kubernetes/pull/126671), [@fusida](https://github.com/fusida)) [SIG Network]

* GCE/GKE load balancer health check default interval changes from 2 seconds to 8 seconds, unhealthyThreshold to 3. ([#70099](https://github.com/kubernetes/kubernetes/pull/70099), [@grayluck](https://github.com/grayluck))
    * Health check parameters are configurable to be bigger than default values.

          },
        ).writeTimeout(Duration.ofMillis(500))
        .build()
    server.start()
    server.enqueue(
      MockResponse
        .Builder()
        .throttleBody(1, 1, TimeUnit.SECONDS)
        .build(),
    ) // Prevent the server from reading!
    val request =
      Request(
        url = server.url("/"),
        body =
          object : RequestBody() {

- Expose `rest_client_rate_limiter_duration_seconds` metric to component-base to track client side rate limiter latency in seconds. Broken down by verb and URL. ([#100311](https://github.com/kubernetes/kubernetes/pull/100311), [@IonutBajescu](https://github.com/IonutBajescu)) [SIG API Machinery, Cluster Lifecycle and Instrumentation]

* Caching added to the OIDC client auth plugin to fix races and reduce the time kubectl commands using this plugin take by several seconds. ([#38167](https://github.com/kubernetes/kubernetes/pull/38167), [@ericchiang](https://github.com/ericchiang))

* CSI detach timeout increased from 10 seconds to 2 minutes ([#84321](https://github.com/kubernetes/kubernetes/pull/84321), [@cduchesne](https://github.com/cduchesne))

    public static final String LABELS_search_result_status_over = "{labels.search_result_status_over}";

    /** The key of the message: ({0} seconds) */
    public static final String LABELS_search_result_time = "{labels.search_result_time}";

    /** The key of the message: Prev */
    public static final String LABELS_prev_page = "{labels.prev_page}";

- Kubelets configured to rotate client certificates now publish a `certificate_manager_server_ttl_seconds` gauge metric indicating the remaining seconds until certificate expiration. ([#91148](https://github.com/kubernetes/kubernetes/pull/91148), [@liggitt](https://github.com/liggitt)) [SIG Auth and Node]

- A new kubelet metric `image_pull_duration_seconds` was added. The metric tracks the duration (in seconds) it takes for an image to be pulled, including the time spent in the waiting queue of image puller. The metric is broken down by bucketed image size. ([#121719](https://github.com/kubernetes/kubernetes/pull/121719), [@ruiwen-zhao](https://github.com/ruiwen-zhao))...

- `PodLifecycleSleepAction` is now turned on by default allowing users to create containers with sleep lifecycle action with a duration of zero seconds ([#130621](https://github.com/kubernetes/kubernetes/pull/130621), [@sreeram-venkitesh](https://github.com/sreeram-venkitesh)) [SIG Node]