- Kubernetes components that accepted X.509 client certificate authentication now read the user UID from a certificate subject name RDN with object ID `1.3.6.1.4.1.57683.2`. An RDN with this object ID had to contain a string value and appear no more than once in the certificate subject. Reading the user UID from this RDN could be disabled by setting the beta feature gate `AllowParsingUserUIDFromCertAuth` to `false`(until the feature gate graduated to...

     * so that we can pass non-null instances of the interface.
     *
     * (Currently it's safe to pass instances for which compareTo() always
     * returns 0, but if we had a SingletonImmutableSortedSet, this might no
     * longer be the case.)
     *
     * javax.naming.Name and java.util.concurrent.Delayed might work, but

### Bug or Regression

- Fix a scheduler preemption issue where the victim pod was not deleted due to incorrect status patching. This issue occurred when the preemptor and victim pods had different QoS classes in their status, causing the preemption to fail entirely. ([#126694](https://github.com/kubernetes/kubernetes/pull/126694), [@Huang-Wei](https://github.com/Huang-Wei)) [SIG Scheduling]

- The following roles have had `Watch` added to them (prefixed with `system:controller:`):
  
  - `cronjob-controller`
  - `endpoint-controller`
  - `endpointslice-controller`
  - `endpointslicemirroring-controller`
  - `horizontal-pod-autoscaler`

 *
 * <p>This version adds some verbosity around the {@code Guard} objects, but removes that same
 * verbosity, and more, from the {@code get} and {@code set} methods. {@code Monitor} implements the
 * same efficient signaling as we had to hand-code in the {@code ReentrantLock} version above.
 * Finally, the programmer no longer has to hand-code the wait loop, and therefore doesn't have to
 * remember to use {@code while} instead of {@code if}.
 *

- kube-scheduler implemented scheduling hints for the `NodeUnschedulable` plugin. The scheduling hints allowed the scheduler to only retry scheduling a Pod that had been previously rejected by the `NodeSchedulable` plugin if a new Node or a Node update had set `.spec.unschedulable` to false.
   ([#122334](https://github.com/kubernetes/kubernetes/pull/122334), [@carlory](https://github.com/carlory))

   *
   * However, for a variety of reasons, Google developers have written a ton of code over the past
   * decade that assumes that they can use checkNotNull for non-precondition checks. I had hoped to
   * take a principled stand on this, but the amount of such code is simply overwhelming. To avoid
   * creating a lot of compile errors that users would not find to be informative, we're giving in

        return whenClosed = new CountDownLatch(1);
      }
    }
  }

  /**
   * Returns an object that can be used to wait until this objects' deferred closeables have all had
   * {@link Runnable}s that close them submitted to each one's closing {@link Executor}.
   */
  @VisibleForTesting
  CountDownLatch whenClosedCountDown() {
    return closeables.whenClosedCountDown();
  }

    assertThat(response3.body.string()).isEqualTo("def")
    assertThat(server.takeRequest().exchangeIndex).isEqualTo(1) // New connection.
  }

  /**
   * We had a bug where we'd perform infinite retries of route that fail with connection shutdown
   * errors. The problem was that the logic that decided whether to reuse a route didn't track

### Bug or Regression

- Fix a scheduler preemption issue where the victim pod was not deleted due to incorrect status patching. This issue occurred when the preemptor and victim pods had different QoS classes in their status, causing the preemption to fail entirely. ([#126695](https://github.com/kubernetes/kubernetes/pull/126695), [@Huang-Wei](https://github.com/Huang-Wei)) [SIG Scheduling]