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
   * certain HTTP/2 errors. https://github.com/square/okhttp/issues/5547

## Changes since v1.3.0-beta.1

### Experimental Features

### Introducing Windows CSI support alpha for Kubernetes

With the release of Kubernetes 1.18, an alpha version of CSI Proxy for Windows is getting released. CSI proxy enables non-privileged (pre-approved) containers to perform privileged storage operations on Windows. CSI drivers can now be supported in Windows by leveraging CSI proxy.
SIG Storage made a lot of progress in the 1.18 release.

- Kubeadm: during execution of the "check expiration" command, treat the etcd CA as external if there is a missing etcd CA key file (etcd/ca.key) and perform the proper validation on certificates signed by the etcd CA. Additionally, make sure that the CA for all entries in the output table is included - for both certificates on disk and in kubeconfig files. ([#106924](https://github.com/kubernetes/kub...

func (er erasureObjects) updateRestoreMetadata(ctx context.Context, bucket, object string, objInfo ObjectInfo, opts ObjectOptions) error {
	oi := objInfo.Clone()
	oi.metadataOnly = true // Perform only metadata updates.

	// allow retry in the case of failure to restore
	delete(oi.UserDefined, xhttp.AmzRestore)

	if _, err := er.CopyObject(ctx, bucket, object, bucket, object, oi, ObjectOptions{

* Fix GCE etcd scripts to pass in all required parameters for the etcd migration utility to correctly perform HA upgrades and downgrades ([#61957](https://github.com/kubernetes/kubernetes/pull/61957), [@jpbetz](https://github.com/jpbetz))

			if !rinfo.VersionPurgeStatus.Empty() {
				rinfo.VersionPurgeStatus = replication.VersionPurgeComplete
				return
			}
		case isErrReadQuorum(serr), isErrWriteQuorum(serr):
			// destination has some quorum issues, perform removeObject() anyways
			// to complete the operation.
		default:
			if err != nil && minio.IsNetworkOrHostDown(err, true) && !globalBucketTargetSys.isOffline(tgt.EndpointURL()) {

    * Add [`tf.data.experimental.get_model_proto`](https://www.tensorflow.org/api_docs/python/tf/data/experimental/get_model_proto) to allow users to peek into the analytical model inside of a dataset iterator....

computing the new offsets and applying\n * the new style.<br />\n * Prefer `scheduleUpdate` over `update` because of performance reasons.\n * @method\n * @memberof Popper\n */\nexport default function update() {\n  // if popper is destroyed, don't perform any further update\n  if (this.state.isDestroyed) {\n    return;\n  }\n\n  let data = {\n    instance: this,\n    styles: {},\n    arrowStyles: {},\n    attributes: {},\n    flipped: false,\n    offsets: {},\n  };\n\n  // compute reference element...

* Fix GCE etcd scripts to pass in all required parameters for the etcd migration utility to correctly perform HA upgrades and downgrades ([#61960](https://github.com/kubernetes/kubernetes/pull/61960), [@jpbetz](https://github.com/jpbetz))