}
        message += " ";
      }
      if (shouldShowExtraNanos) {
        message += overWaitLeftoverNanos + " nanoseconds ";
      }

      message += "delay)";
    }
    // It's confusing to see a completed future in a timeout message; if isDone() returns false,
    // then we know it must have given a pending toString value earlier. If not, then the future

    print('WARNING: The NDK version in %s is %s, which is not '
          'supported by Bazel (officially supported versions: %s). Please use '
          'another version. Compiling Android targets may result in confusing '
          'errors.\n' %
          (android_ndk_home_path, ndk_version, _SUPPORTED_ANDROID_NDK_VERSIONS))
  write_action_env_to_bazelrc('ANDROID_NDK_VERSION', ndk_version)

 * socket that could receive both IPv4 and IPv6 connections (by disabling the {@code IPV6_V6ONLY}
 * socket option on an IPv6 socket). Yes, it's confusing. Nevertheless, these "mapped" addresses
 * were never supposed to be seen on the wire. That assumption was dropped, some say mistakenly, in
 * later RFCs with the apparent aim of making IPv4-to-IPv6 transition simpler.
 *

 *
 * Hostnames have different requirements and use a different encoding scheme. It consists of IDNA
 * mapping and Punycode encoding.
 *
 * In order to avoid confusion and discourage phishing attacks, [IDNA Mapping][idna] transforms
 * names to avoid confusing characters. This includes basic case folding: transforming shouting
 * `SQUARE.COM` into cool and casual `square.com`. It also handles more exotic characters. For

 * socket that could receive both IPv4 and IPv6 connections (by disabling the {@code IPV6_V6ONLY}
 * socket option on an IPv6 socket). Yes, it's confusing. Nevertheless, these "mapped" addresses
 * were never supposed to be seen on the wire. That assumption was dropped, some say mistakenly, in
 * later RFCs with the apparent aim of making IPv4-to-IPv6 transition simpler.
 *

  `I don't see,' said the Caterpillar.

  `I'm afraid I can't put it more clearly,' Alice replied very
politely, `for I can't understand it myself to begin with; and
being so many different sizes in a day is very confusing.'

  `It isn't,' said the Caterpillar.

  `Well, perhaps you haven't found it so yet,' said Alice; `but
when you have to turn into a chrysalis--you will some day, you

 *  Fix: Don't log the protocol (HTTP/1.1, h2) in HttpLoggingInterceptor if the
    protocol isn't negotiated yet! Previously we'd log HTTP/1.1 by default, and
    this was confusing.
 *  Fix: Omit the message from MockWebServer's HTTP/2 `:status` header.
 *  Fix: Handle 'Expect: 100 Continue' properly in MockWebServer.


## Version 3.8.1

_2017-06-18_

   * Support kube-master-url flag without kubeconfig

   * Fix concurrent R/Ws in dns.go

   * Fix confusing logging when initialize server

   * Fix printf in cmd/kube-dns/app/server.go

   * Fix version on startup and `--version` flag

   * Support specifying port number for nameserver in stubDomains

- Fixed binding of block PersistentVolumes / PersistentVolumeClaims when BlockVolume feature is off. ([#84049](https://github.com/kubernetes/kubernetes/pull/84049), [@jsafrane](https://github.com/jsafrane))
- Report non-confusing error for negative storage size in PVC spec. ([#82759](https://github.com/kubernetes/kubernetes/pull/82759), [@sttts](https://github.com/sttts))

          // Produced ClientHello handshake message
          // Consuming ServerHello handshake message
          // Consuming server Certificate handshake message
          // Consuming server CertificateStatus handshake message
          // Found trusted certificate
          // Consuming ECDH ServerKeyExchange handshake message
          // Consuming ServerHelloDone handshake message