* Pod IP allocation-related issues. Deleting the docker checkpoint prior to
restarting the daemon alleviates this issue, but hasn’t been verified to
completely eliminate the IP allocation issue. More information [here](https://github.com/kubernetes/kubernetes/issues/21523#issuecomment-191498969)

 - The intree volume plugin flocker support was completely removed from Kubernetes. ([#111618](https://github.com/kubernetes/kubernetes/pull/111618), [@Jiawei0227](https://github.com/Jiawei0227))
 - The intree volume plugin quobyte support has been completely removed from Kubernetes. ([#111619](https://github.com/kubernetes/kubernetes/pull/111619), [@Jiawei0227](https://github.com/Jiawei0227))

  Also `metricsBindAddress` and `healthzBindAddress` fields from `kubescheduler.config.k8s.io/v1beta1` are no-op and expected to be empty. Removed in `kubescheduler.config.k8s.io/v1beta2` completely. ([#96345](https://github.com/kubernetes/kubernetes/pull/96345), [@ingvagabund](https://github.com/ingvagabund))
  In addition, please be careful that:

        throw ProtocolException("unexpected end of stream")
      }
      try {
        super.close()
        complete(null)
      } catch (e: IOException) {
        throw complete(e)
      }
    }

    private fun <E : IOException?> complete(e: E): E {
      if (completed) return e
      completed = true
      return bodyComplete(bytesReceived, responseDone = false, requestDone = true, e = e)
    }
  }

 *       be "done" as soon as <i>that</i> {@code Future} completes in any way. Notably, a {@code
 *       Future} is "completed" even if it is cancelled while its underlying work continues on a
 *       thread, an RPC, etc. The {@code Future} is also "completed" if it fails "early" -- for
 *       example, if the deadline expires on a {@code Future} returned from {@link

- Improved air-gapped and offline support
  - kubeadm now handles air-gapped environments by using the local client version as a fallback.
  - Some kubeadm commands are now allowed to work in a completely offline mode.
- Certificate handling improvements:
  - Renew certs as part of upgrade.
  - New `kubeadm alpha phase certs renew` command for renewing certificates.

   * Clears fields that are no longer needed after this future has completed -- or at least all its
   * inputs have completed (more precisely, after {@link #handleAllCompleted()} has been called).
   * Often called multiple times (that is, both when the inputs complete and when the output
   * completes).
   *
   * <p>This is similar to our proposed {@code afterCommit} method but not quite the same. See the
   * description of CL 265462958.
   */

 *       be "done" as soon as <i>that</i> {@code Future} completes in any way. Notably, a {@code
 *       Future} is "completed" even if it is cancelled while its underlying work continues on a
 *       thread, an RPC, etc. The {@code Future} is also "completed" if it fails "early" -- for
 *       example, if the deadline expires on a {@code Future} returned from {@link

  TF_Operation* feed = Placeholder(graph, s);
  ASSERT_EQ(TF_OK, TF_GetCode(s)) << TF_Message(s);
  TF_Output feed_out_0 = TF_Output{feed, 0};
  int num_dims;

  // Fetch the shape, it should be completely unknown.
  num_dims = TF_GraphGetTensorNumDims(graph, feed_out_0, s);
  ASSERT_EQ(TF_OK, TF_GetCode(s)) << TF_Message(s);
  EXPECT_EQ(-1, num_dims);

  // Set the shape to be unknown, expect no change.

* Access to externally managed IP addresses via the kube-apiserver service proxy subresource is no longer allowed by default. This can be re-enabled via the `ServiceProxyAllowExternalIPs` feature gate, but will be disallowed completely in 1.11 ([#57265](https://github.com/kubernetes/kubernetes/pull/57265), [@brendandburns](https://github.com/brendandburns))