- Allowed specifying `ExternalTrafficPolicy` for `Services` with `ExternalIPs`. ([#119150](https://github.com/kubernetes/kubernetes/pull/119150), [@tnqn](https://github.com/tnqn))
- Changed kubelet logs from `error` to `info` for uncached partitions when using CRI stats provider. ([#100448](https://github.com/kubernetes/kubernetes/pull/100448), [@saschagrunert](https://github.com/saschagrunert))

        swap(array, min, from);
      }
      return;
    }

    // Let's play quickselect! We'll repeatedly partition the range [from, to] containing the
    // required element, as long as it has more than one element.
    while (to > from) {
      int partitionPoint = partition(array, from, to);
      if (partitionPoint >= required) {
        to = partitionPoint - 1;
      }

 */
internal fun reorderForHappyEyeballs(addresses: List<InetAddress>): List<InetAddress> {
  if (addresses.size < 2) {
    return addresses
  }

  val (ipv6, ipv4) = addresses.partition { it is Inet6Address }

  return if (ipv6.isEmpty() || ipv4.isEmpty()) {
    addresses
  } else {
    interleave(ipv6, ipv4)
  }

* Fixed data loss issue if using existing azure disk with partitions in disk mount  ([#63270](https://github.com/kubernetes/kubernetes/pull/63270), [@andyzhangx](https://github.com/andyzhangx))

- Fix: using NVMe AWS EBS volumes partitions. ([#100500](https://github.com/kubernetes/kubernetes/pull/100500), [@jsafrane](https://github.com/jsafrane))

        // Implementation would store registration details for recovery
    }
}
```

### 8.2 Network Partition Handling
```java
public class WitnessNetworkPartitionHandler {
    public void handleNetworkPartition() {
        log.warn("Network partition detected - witness notifications may be delayed");
        
        // Switch to more aggressive connection retry

   */
  public static <T extends @Nullable Object> List<List<T>> partition(List<T> list, int size) {
    checkNotNull(list);
    checkArgument(size > 0);
    return (list instanceof RandomAccess)
        ? new RandomAccessPartition<>(list, size)
        : new Partition<>(list, size);
  }

  private static class Partition<T extends @Nullable Object> extends AbstractList<List<T>> {
    final List<T> list;

* Add --no-negcache flag to kube-dns to prevent caching of NXDOMAIN responses. ([#66036](https://github.com/kubernetes/kubernetes/pull/66036), [@prameshj](https://github.com/prameshj))
* fix data loss issue if using existing azure disk with partitions in disk mount  ([#63270](https://github.com/kubernetes/kubernetes/pull/63270), [@andyzhangx](https://github.com/andyzhangx))

* Fix a bug that preempting a pod may block forever. ([#65987](https://github.com/kubernetes/kubernetes/pull/65987), [@Random-Liu](https://github.com/Random-Liu))
* fix data loss issue if using existing azure disk with partitions in disk mount  ([#63270](https://github.com/kubernetes/kubernetes/pull/63270), [@andyzhangx](https://github.com/andyzhangx))

* Pods that are terminating due to eviction by the nodecontroller (typically due to unresponsive kubelet, or network partition) now surface in `kubectl get` output  ([#36017](https://github.com/kubernetes/kubernetes/pull/36017), [@foxish](https://github.com/foxish))
    * as being in state "Unknown", along with a longer description in `kubectl describe` output.