([kubernetes/features#82](https://github.com/kubernetes/enhancements/issues/82))
  - [beta] `Federated Replica Sets` create and maintain matching
    `Replica Set`s in some or all clusters in a federation, with the
    desired replica count distributed equally or according to
    specified per-cluster weights.
	([docs](http://kubernetes.io/docs/user-guide/federation/federated-replicasets.md))

defines the number of nodes that if not Ready in time won’t cause kube-up
failure.
  * “kubectl rolling-update” only supports Replication Controllers (it doesn’t
support Replica Sets). It’s recommended to use Deployment 1.2 with “kubectl
rollout” commands instead, if you want to rolling update Replica Sets.
  * When live upgrading Kubelet to 1.2 without draining the pods running on the node,

  }
  auto* config = server_def.mutable_default_session_config();
  config->mutable_experimental()->set_collective_group_leader(
      absl::StrCat("/job:", job_name, "/replica:0/task:", 0));
  auto* rewrite_options =
      config->mutable_graph_options()->mutable_rewrite_options();
  rewrite_options->set_scoped_allocator_optimization(
      tensorflow::RewriterConfig::ON);

      CreateContext(serialized_server_def_1,
                    /*isolate_session_state=*/false, init_timeout_in_ms);

  // Remote device on `worker2`.
  const char remote_device[] = "/job:localhost/replica:0/task:2/device:CPU:0";
  // `ctx_0`, `ctx_1` contains `remote_device`.
  {
    const std::vector<std::string>& device_names = ListDeviceNames(ctx_0);
    ASSERT_TRUE(std::find(device_names.begin(), device_names.end(),

                              const char* prefix, TF_Status* status) {
  std::vector<std::unique_ptr<tensorflow::Device>> devices;

  if (prefix == nullptr || strlen(prefix) == 0)
    prefix = "/job:localhost/replica:0/task:0";

  tensorflow::SessionOptions sess_options;
  (*sess_options.config.mutable_device_count())["CPU"] = num_cpus;
  status->status =
      tensorflow::DeviceFactory::AddCpuDevices(sess_options, prefix, &devices);

    return nullptr;
  }
  std::vector<std::unique_ptr<tensorflow::Device>> devices;
  status->status = tensorflow::DeviceFactory::AddDevices(
      opts->session_options.options, "/job:localhost/replica:0/task:0",
      &devices);
  if (!status->status.ok()) return nullptr;
  std::unique_ptr<tensorflow::DeviceMgr> device_mgr(
      new tensorflow::DynamicDeviceMgr(std::move(devices)));

- **Apps**
  - [stable] When replica sets cannot create pods, they will now report detail via the API about the underlying reason ([kubernetes/features#120](https://github.com/kubernetes/enhancements/issues/120))

- Output for the `ScalingReplicaSet` event has changed from:
      Scaled <up|down> replica set <replica-set-name> to <new-value> from <old-value>
  to:
      Scaled <up|down> replica set <replica-set-name> from <old-value> to <new-value>. ([#125118](https://github.com/kubernetes/kubernetes/pull/125118), [@jsoref](https://github.com/jsoref)) [SIG Apps and CLI]

		if !ventry.Valid() {
			return "", errors.New("internal error: invalid version entry generated")
		}
	}
	updateVersion := false
	if fi.VersionPurgeStatus().Empty() && (fi.DeleteMarkerReplicationStatus() == "REPLICA" || fi.DeleteMarkerReplicationStatus().Empty()) {
		updateVersion = fi.MarkDeleted
	} else {
		// for replication scenario
		if fi.Deleted && fi.VersionPurgeStatus() != replication.VersionPurgeComplete {

	// Save the final object size and modtime.
	fi.Size = objectSize
	fi.ModTime = opts.MTime
	if opts.MTime.IsZero() {
		fi.ModTime = UTCNow()
	}

	// Save successfully calculated md5sum.
	// for replica, newMultipartUpload would have already sent the replication ETag
	if fi.Metadata["etag"] == "" {
		if opts.UserDefined["etag"] != "" {
			fi.Metadata["etag"] = opts.UserDefined["etag"]