- Updated `kubectl scale` to return a consistent error message when a specified resource is not found. Previously, it returned: `error: no objects passed to scale <GroupResource> "<ResourceName>" not found`. It now matches the format used by other commands (e.g., `kubectl get`): `Error from server (NotFound): <GroupResource> "<ResourceName>"...

- Fixes an issue where Windows kube-proxy's ModifyLoadBalancer API updates did not match HNS state in version 15.4. ModifyLoadBalancer policy is supported from Kubernetes 1.31+. ([#131506](https://github.com/kubernetes/kubernetes/pull/131506), [@princepereira](https://github.com/princepereira)) [SIG Windows]

everywhere. Otherwise the two cache instances will stomp on each other, corrupt the response cache, and possibly crash your program.

Response caching uses HTTP headers for all configuration. You can add request headers like `Cache-Control: max-stale=3600` and OkHttp's cache will honor them. Your webserver configures how long responses are cached with its own response headers, like `Cache-Control: max-age=9600`. There are cache headers to force a cached response, force a network response, or force...

- Kubeadm will cleanup the stale data on best effort basis. Stale data will be removed when each reset phase are executed, default etcd data directory will be cleanup when the `remove-etcd-member` phase are executed. ([#110972](https://github.com/kubernetes/kubernetes/pull/110972), [@chendave](h...

     *   failure to reach any individual IP address doesn't fail the overall request. This can
     *   increase availability of multi-homed services.
     *
     * * **Stale pooled connections.** The [ConnectionPool] reuses sockets
     *   to decrease request latency, but these connections will occasionally time out.
     *
     * * **Unreachable proxy servers.** A [ProxySelector] can be used to

	// cycle until we can safely update the in-memory cache.
	//
	// An in-memory cache must be replaced only if we know for sure that the
	// values loaded from disk are not stale. They might be stale if the
	// cached.updatedAt is more recent than the refresh cycle began.
	if cache.updatedAt.Before(loadedAt) || firstTime {
		// No one has updated anything since the config was loaded,

Besides the above new features, support for Azure Virtual Machine Scale Sets (VMSS) and Cluster-Autoscaler is now stable and considered GA:

- Azure virtual machine scale sets (VMSS) allow you to create and manage identical load balanced VMs that automatically increase or decrease based on demand or a set schedule.

- Graduating `pending_pods`, `preemption_attempts_total`, `preemption_victims` and `schedule_attempts_total` metrics to stable. Also `e2e_scheduling_duration_seconds` is renamed to `scheduling_attempt_duration_seconds` and the latter is graduated to stable. (#105941, @rezakrimi) [SIG Instrumentation, Scheduling and Testing]

    * sync time.
 
    * The goal is for administrators to be able to alert on proxies that, for
    * whatever reason, are quite stale.
* Fix a string comparison bug in IPVS graceful termination where UDP real servers are not deleted. ([#78999](https://github.com/kubernetes/kubernetes/pull/78999), [@andrewsykim](https://github.com/andrewsykim))

- kubelet: Fix - the volume manager didn't check the device mount state in the actual state of the world before marking the volume as detached. It may cause a pod to be stuck in the Terminating state due to the above issue when it was deleted. ([#128219](https://github.com/kubernetes/kubernetes/pull/128219), [@carlory](https://github.com/carlory))