- Fixed an issue in the Windows kube-proxy (winkernel) where IPv4 and IPv6 Service load balancers could be incorrectly shared, causing broken dual-stack Service behavior. The kube-proxy now tracks load balancers per IP family, enabling correct support for PreferDualStack and RequireDualStack Services on Windows nodes. ([#136373](https://github.com/kubernetes/kubernetes/pull/136373), [@princepereira](https://gith...

<details><summary> View **DEPRECATED** older policy association commands</summary>

Please **do not use** these as they may be removed or their behavior may change.

```sh
mc admin policy attach myminio mypolicy --user='uid=james,cn=accounts,dc=myldapserver,dc=com'
```


```sh

*   The default `auto-detect` behavior of the kubelet's `--cloud-provider` flag is removed.
    *   You can manually set `--cloud-provider=auto-detect`, but be aware that this behavior will be removed completely in a future version.

   * least one of the previous mean and the value is non-finite.
   */
  static double calculateNewMeanNonFinite(double previousMean, double value) {
    /*
     * Desired behaviour is to match the results of applying the naive mean formula. In particular,
     * the update formula can subtract infinities in cases where the naive formula would add them.
     *
     * Consequently:

  }

  /**
   * Associates {@code value} with the specified row and column indices. The logic {@code
   * put(rowKeyList().get(rowIndex), columnKeyList().get(columnIndex), value)} has the same
   * behavior, but this method runs more quickly.
   *
   * @param rowIndex position of the row key in {@link #rowKeyList()}
   * @param columnIndex position of the row key in {@link #columnKeyList()}

       * more information, see the comments in that class.
       *
       * We already know that that's how it behaves, and subclasses of Converter can't change that
       * behavior. So there's no sense in making all subclass authors exclude the method from any
       * NullPointerTester tests that they have.
       */
      ignoredMembers.add(Converter.class.getMethod("apply", Object.class));

      // without waiting for the user's executor to run our submitted Runnable. However, this can
      // interact poorly with the reentrancy-avoiding behavior of this executor - when the operation
      // before the cancelled future completes, it will synchronously complete both the newFuture
      // from the cancelled operation and its own. This can cause one runnable to queue two tasks,

   *       simply switch over to use the JDK implementation wholesale if probable hash flooding is
   *       detected, sacrificing the compactness guarantee in very rare cases in exchange for much
   *       more reliable worst-case behavior.
   *   <li>null, if no entries have yet been added to the map
   * </ul>
   */
  private transient @Nullable Object table;

  /**

	if rcfg.Config != nil && rcfg.Config.RoleArn != "" {
		// For backward compatibility of objects pending/failed replication.
		// Save replication related statuses in the new internal representation for
		// compatible behavior.
		if !oi.ReplicationStatus.Empty() {
			oi.ReplicationStatusInternal = fmt.Sprintf("%s=%s;", rcfg.Config.RoleArn, oi.ReplicationStatus)
		}
		if !oi.VersionPurgeStatus.Empty() {

  }

  /**
   * Associates {@code value} with the specified row and column indices. The logic {@code
   * put(rowKeyList().get(rowIndex), columnKeyList().get(columnIndex), value)} has the same
   * behavior, but this method runs more quickly.
   *
   * @param rowIndex position of the row key in {@link #rowKeyList()}
   * @param columnIndex position of the row key in {@link #columnKeyList()}