// timed out. e.g. if we weren't event able to grab the lock within the timeout we would never
      // even check the guard. I don't think we care too much about this use case but it could lead
      // to a confusing error message.
      throw new TimeoutException("Timed out waiting for " + this + " to reach the RUNNING state.");
    }
  }

  @Override
  public final void awaitTerminated() {

 * <p><b>Warning:</b> The methods of {@code ForwardingList} forward <b>indiscriminately</b> to the
 * methods of the delegate. For example, overriding {@link #add} alone <b>will not</b> change the
 * behavior of {@link #addAll}, which can lead to unexpected behavior. In this case, you should
 * override {@code addAll} as well, either providing your own implementation, or delegating to the
 * provided {@code standardAddAll} method.
 *

 * <b>indiscriminately</b> to the methods of the delegate. For example, overriding {@link
 * #add(Object, int)} alone <b>will not</b> change the behavior of {@link #add(Object)}, which can
 * lead to unexpected behavior. In this case, you should override {@code add(Object)} as well,
 * either providing your own implementation, or delegating to the provided {@code standardAdd}
 * method.
 *

# This should only be necessary when running on A GCP VM, on a default
# network, which has the MTU of 1460,
# due to 40 bytes being reserved for GCP's internal usage.
# Note, an invalid sub-interface name will lead to an obscure error, e.g.:
# "The filename, directory name, or volume label syntax is incorrect."
# In such cases, check that the name of the sub-interface is valid:
# `netsh interface show interface`

      listener.cacheMiss(call)
    }

    var networkResponse: Response? = null
    try {
      networkResponse = chain.proceed(networkRequest)
    } finally {
      // If we're crashing on I/O or otherwise, don't leak the cache body.
      if (networkResponse == null && cacheCandidate != null) {
        cacheCandidate.body.closeQuietly()
      }
    }

    // If we have a cache response too, then we're doing a conditional get.

 * the methods of the delegate. For example, overriding {@link #add} alone <i>will not</i> change
 * the behavior of {@link #addAll}, which can lead to unexpected behavior. In this case, you should
 * override {@code addAll} as well, either providing your own implementation, or delegating to the
 * provided {@code standardAddAll} method.
 *

          if (taskFuture.isDone()) {
            // Since the value of oldFuture can only ever be immediateFuture(null) or setFuture of
            // a future that eventually came from immediateFuture(null), this doesn't leak
            // throwables or completion values.
            newFuture.setFuture(oldFuture);
          } else if (outputFuture.isCancelled() && taskExecutor.trySetCancelled()) {

          if (taskFuture.isDone()) {
            // Since the value of oldFuture can only ever be immediateFuture(null) or setFuture of
            // a future that eventually came from immediateFuture(null), this doesn't leak
            // throwables or completion values.
            newFuture.setFuture(oldFuture);
          } else if (outputFuture.isCancelled() && taskExecutor.trySetCancelled()) {

   * message must mention two values that were associated with the duplicate key in two different
   * calls to Builder.put (though we don't really care *which* two values if there were more than
   * two). These considerations lead us to have a field of type DuplicateKey in the Builder, which
   * will remember the first duplicate key we encountered. All later calls to buildOrThrow() can

 *
 * ### The response body must be closed.
 *
 * Each response body is backed by a limited resource like a socket (live network responses) or
 * an open file (for cached responses). Failing to close the response body will leak resources and
 * may ultimately cause the application to slow down or crash.
 *
 * Both this class and [Response] implement [Closeable]. Closing a response simply