}

  /**
   * Add a shutdown hook to wait for thread completion in the given {@link ExecutorService service}.
   * This is useful if the given service uses daemon threads, and we want to keep the JVM from
   * exiting immediately on shutdown, instead giving these daemon threads a chance to terminate
   * normally.
   *
   * @param service ExecutorService which uses daemon threads

// is for performance optimization by reusing an exiting unused op rather than
// creating a new op every time. If `raw_device_name` is `NULL` or empty, it
// does not set the device name. If it's not `NULL`, then it attempts to parse
// and set the device name. It's effectively `TFE_OpSetDevice`, but it is faster
// than separately calling it because if the existing op has the same

        Executors.newSingleThreadScheduledExecutor(new ThreadFactoryImpl());
    // Add a listener to shut down the executor after the service is stopped. This ensures that the
    // JVM shutdown will not be prevented from exiting after this service has stopped or failed.
    // Technically this listener is added after start() was called so it is a little gross, but it

}
------------------------------------------------

in `build.gradle`.

=== Limitations

The following should be taken into consideration when writing new tests or adjusting existing ones:

==== TLS

`JKS` and `PKCS#12` keystores cannot be used in FIPS mode. If the test depends on being able to use

* Set an upper bound (5 minutes) on how long the Kubelet will wait before exiting when the client cert from disk is missing or invalid. This prevents the Kubelet from waiting forever without attempting to bootstrap a new client credentials. ([#59316](https://github.com/kubernetes/kubernetes/pull/59316), [@smarterclayton](https://github.com/smarterclayton))

      }
    }
    connection.writePing()
    connection.shutdown(ErrorCode.PROTOCOL_ERROR)
    assertThat(connection.openStreamCount()).isEqualTo(1)
    connection.awaitPong() // Prevent the peer from exiting prematurely.

    // Verify the peer received what was expected.
    val synStream1 = peer.takeFrame()
    assertThat(synStream1.type).isEqualTo(Http2.TYPE_HEADERS)
    val pingFrame = peer.takeFrame()

    - [Client Binaries](#client-binaries-8)
    - [Server Binaries](#server-binaries-8)
  - [Major Themes](#major-themes)
  - [Features](#features)
  - [Known Issues](#known-issues)
  - [Notable Changes to Existing Behavior](#notable-changes-to-existing-behavior)
  - [Deprecations](#deprecations)
  - [Action Required Before Upgrading](#action-required-before-upgrading)
  - [External Dependency Version Information](#external-dependency-version-information)

*   The admission webhook configurations API, part of the admissionregistration API, is now at v1beta1. Delete any existing webhook configurations before you upgrade, and update your configuration files to use the latest API. For this and the previous...

  - NetworkPolicy resources will no longer be served from `extensions/v1beta1` in v1.16. Migrate use to the `networking.k8s.io/v1` API, available since v1.8. Existing persisted data can be retrieved via the `networking.k8s.io/v1` API.

EndpointSlices with Pod referencing Nodes that don't exist couldn't be created or updated. The behavior on the EndpointSlice controller has been modified to update the EndpointSlice without the Pods that reference non-existing Nodes and keep retrying until all Pods reference existing Nodes. However, if `service.Spec.PublishNotReadyAddresses` is set, all the Pods are published without retrying. Fixed EndpointSlices metrics to reflect correctly the number of desired EndpointSlices when no endpoints...