uses: actions/checkout@v2
      with:
        # We must fetch at least the immediate parents so that if this is
        # a pull request then we can checkout the head.
        fetch-depth: 2

    # If this run was triggered by a pull request event, then checkout
    # the head of the pull request instead of the merge commit.
    - run: git checkout HEAD^2
      if: ${{ github.event_name == 'pull_request' }}

/// tip

Note that we're using async/await with the new `AsyncClient` - the request is asynchronous.

///

/// warning

If your application relies on lifespan events, the `AsyncClient` won't trigger these events. To ensure they are triggered, use `LifespanManager` from <a href="https://github.com/florimondmanca/asgi-lifespan#usage" class="external-link" target="_blank">florimondmanca/asgi-lifespan</a>.

///

## Other Asynchronous Function Calls

Additionally, we MUST accept incoming traffic during this time, both new and existing.
Typically, it is rare to get new connections after shutdown has started - once shutdown is triggered the endpoint is marked `NotReady` and removed from load balancing consideration.
However, some applications do not utilize Service readiness (connecting directly to the pod, etc).

A push occurs when Istiod detects an update of some set of configuration is needed. This results in roughly the same result as a Request (new configuration is pushed to the client), and is just triggered by a different source.

Various components described in Config Ingestion can trigger a Config Update. These are batched up ("debounced"), to avoid excessive activity when many changes happen in succession, and eventually enqueued in the Push Queue.

	// wait for an update event
	mt.Assert(EventTotals.Name(), map[string]string{"type": "update"}, monitortest.AtLeast(2))

	// wait for the pod to be annotated
	// after Pod annotated, another update event will be triggered.
	assertPodAnnotated(t, client, pod)

	// Assert expected calls actually made
	fs.AssertExpectations(t)

	// unlabelling the namespace should cause only one RemovePodFromMesh to happen

 * header, or they may decline the challenge by returning null. In this case the unauthenticated
 * response will be returned to the caller that triggered it.
 *
 * Implementations should check if the initial request already included an attempt to
 * authenticate. If so it is likely that further attempts will not be useful and the authenticator
 * should give up.
 *

# Generate an overview page describing the build
if [[ "$TFCI_INDEX_HTML_ENABLE" == 1 ]]; then
  ./ci/official/utilities/generate_index_html.sh "$TFCI_OUTPUT_DIR/index.html"
fi

# Single handler for all cleanup actions, triggered on an EXIT trap.
# TODO(angerson) Making this use different scripts may be overkill.
cleanup() {
  if [[ "$TFCI_DOCKER_ENABLE" == 1 ]]; then
    ./ci/official/utilities/cleanup_docker.sh
  fi

   */
  @get:JvmName("cacheResponse") val cacheResponse: Response?,
  /**
   * Returns the response for the HTTP redirect or authorization challenge that triggered this
   * response, or null if this response wasn't triggered by an automatic retry. The body of the
   * returned response should not be read because it has already been consumed by the redirecting
   * client.
   */

        boolean found = checkInResult(action, name, infos);
        if ( !found ) {
            // retry, the first watch may have already come back before the triggered change was active
            try {
                setupWatch(w);
                found = checkInResult(action, name, infos);
            }
            catch ( TimeoutException e ) {

   * of {@link CombinedFuture}, the user-supplied callback usually has its own references to inputs.
   */
  /*
   * In certain circumstances, this field might theoretically not be visible to an afterDone() call
   * triggered by cancel(). For details, see the comments on the fields of TimeoutFuture.
   */
  @CheckForNull @LazyInit
  private ImmutableCollection<? extends ListenableFuture<? extends InputT>> futures;