- kube-proxy: Fixed a potential memory leak that could occur in clusters with a high volume of UDP workflows. ([#130032](https://github.com/kubernetes/kubernetes/pull/130032), [@aroradaman](https://github.com/aroradaman))

**Fixed Versions**:
  - kubelet v1.28.4
  - kubelet v1.27.8
  - kubelet v1.26.11
  - kubelet v1.25.16

This vulnerability was reported by Tomer Peled @tomerpeled92"


**CVSS Rating:** High (7.2) [CVSS:3.1/AV:N/AC:L/PR:H/UI:N/S:U/C:H/I:H/A:H](https://www.first.org/cvss/calculator/3.1#CVSS:3.1/AV:N/AC:L/PR:H/UI:N/S:U/C:H/I:H/A:H)

## Changes by Kind

### API Change

   * and accesses that were performed on the map. The queue is drained on writes and when it exceeds
   * its capacity threshold.
   *
   * The Least Recently Used page replacement algorithm was chosen due to its simplicity, high hit
   * rate, and ability to be implemented with O(1) time complexity. The initial LRU implementation
   * operates per-segment rather than globally for increased implementation simplicity. We expect

    }
    LinkedHashSet<E> set = new LinkedHashSet<>();
    Iterables.addAll(set, elements);
    return set;
  }

  /**
   * Creates a {@code LinkedHashSet} instance, with a high enough "initial capacity" that it
   * <i>should</i> hold {@code expectedSize} elements without growth. This behavior cannot be
   * broadly guaranteed, but it is observed to be true for OpenJDK 1.7. It also can't be guaranteed

   * and accesses that were performed on the map. The queue is drained on writes and when it exceeds
   * its capacity threshold.
   *
   * The Least Recently Used page replacement algorithm was chosen due to its simplicity, high hit
   * rate, and ability to be implemented with O(1) time complexity. The initial LRU implementation
   * operates per-segment rather than globally for increased implementation simplicity. We expect

	bucketPeerMetricsGroups []*MetricsGroupV2
)

// v2MetricsMaxBuckets enforces a bucket count limit on metrics for v2 calls.
// If people hit this limit, they should move to v3, as certain calls explode with high bucket count.
const v2MetricsMaxBuckets = 100

func init() {
	clusterMetricsGroups := []*MetricsGroupV2{
		getNodeHealthMetrics(MetricsGroupOpts{dependGlobalNotificationSys: true}),

* fix smb mount issue ([#65751](https://github.com/kubernetes/kubernetes/pull/65751), [@andyzhangx](https://github.com/andyzhangx))
* Fix a scalability issue where high rates of event writes degraded etcd performance. ([#64539](https://github.com/kubernetes/kubernetes/pull/64539), [@ccding](https://github.com/ccding))

- The SchedulerAsyncAPICalls feature gate has been disabled to mitigate a bug where its interaction with asynchronous preemption in could degrade kube-scheduler performance, particularly under high kube-apiserver load. ([#134401](https://github.com/kubernetes/kubernetes/pull/134401), [@macsko](https://github.com/macsko)) [SIG Scheduling]

### Other (Cleanup or Flake)

**Fixed Versions**:
  - kubelet v1.28.4
  - kubelet v1.27.8
  - kubelet v1.26.11
  - kubelet v1.25.16

This vulnerability was reported by Tomer Peled @tomerpeled92"


**CVSS Rating:** High (7.2) [CVSS:3.1/AV:N/AC:L/PR:H/UI:N/S:U/C:H/I:H/A:H](https://www.first.org/cvss/calculator/3.1#CVSS:3.1/AV:N/AC:L/PR:H/UI:N/S:U/C:H/I:H/A:H)

## Changes by Kind

### API Change

		if err := p.loadResync(ctx, buckets, objAPI); err == nil {
			<-ctx.Done()
			return
		}
		duration := max(time.Duration(r.Float64()*float64(time.Minute)),
			// Make sure to sleep at least a second to avoid high CPU ticks.
			time.Second)
		time.Sleep(duration)
	}
}

// Loads bucket replication resync statuses into memory.