* A new volume manager was introduced in kubelet that synchronizes volume mount/unmount (and attach/detach, if attach/detach controller is not enabled). ([#26801](https://github.com/kubernetes/kubernetes/pull/26801), [@saad-ali](https://github.com/saad-ali))
    * This eliminates the race conditions between the pod creation loop and the orphaned volumes loops. It also removes the unmount/detach from the `syncPod()` path so volume clean up never blocks the `syncPod` loop.

- Fix: Mount point may become local without calling NodePublishVolume after node rebooting. ([#119923](https://github.com/kubernetes/kubernetes/pull/119923), [@cvvz](https://github.com/cvvz)) [SIG Node and Storage]

- On compatible systems, a mounter's Unmount implementation is changed to not return an error when the specified target can be detected as not a mount point. On Linux, the behavior of detecting a mount point depends on `umount` command is validated when the mounter is created. Additionally, mount point checks will be skipped in CleanupMountPoint/CleanupMountWithForce if the mounter's Unmount having the changed behavior of not returning error when target is not a...

* fix mount unmount failure for a Windows pod ([#63272](https://github.com/kubernetes/kubernetes/pull/63272), [@andyzhangx](https://github.com/andyzhangx))
* Fix concurrent map access panic ([#65327](https://github.com/kubernetes/kubernetes/pull/65327), [@dashpole](https://github.com/dashpole))
    * Don't watch .mount cgroups to reduce number of inotify watches

        // Process multiple elements
        int count = 1;
        while (transaction.hasMoreElements() && count < 10) {
            transaction.nextElement();
            count++;
        }

        assertTrue(count > 1, "Large data should require multiple elements");
    }

    @Test
    @DisplayName("Test transaction name handling")

   * becomes {@code count}.
   *
   * @return the count of {@code element} in the multiset before this call
   * @throws IllegalArgumentException if {@code count} is negative
   */
  @CanIgnoreReturnValue
  @Override
  public int setCount(E element, int count) {
    checkNotNull(element);
    checkNonnegative(count, "count");
    while (true) {

  static final int DEFAULT_SIZE = 3;

  /**
   * Minimum size of the hash table of a compact hash-based collection. Because small hash tables
   * use a byte[], any smaller size uses the same amount of memory due to object padding.
   */
  private static final int MIN_HASH_TABLE_SIZE = 4;

  private static final int BYTE_MAX_SIZE = 1 << Byte.SIZE; // 2^8 = 256

- On compatible systems, a mounter's Unmount implementation is changed to not return an error when the specified target can be detected as not a mount point. On Linux, the behavior of detecting a mount point depends on `umount` command is validated when the mounter is created. Additionally, mount point checks will be skipped in CleanupMountPoint/CleanupMountWithForce if the mounter's Unmount having the changed behavior of not returning error when target is not a...

   * returns the expected data.
   *
   * @throws Throwable if the listener isn't called or if it resulted in a throwable or if the
   *     result doesn't match the expected value.
   */
  public void assertSuccess(Object expectedData) throws Throwable {
    // Verify that the listener executed in a reasonable amount of time.

#### Storage

* [stable] Mount options

  * The ability to specify mount options for volumes is moved from beta to stable.

  * A new `MountOptions` field in the `PersistentVolume` spec is available to specify mount options. This field replaces an annotation.

  * A new `MountOptions` field in the `StorageClass` spec allows configuration of mount options for dynamically provisioned volumes.