private boolean isProjectPresent = true;

    // ----------------------------------------------------------------------------
    // We need to allow per execution user and global settings as the embedder
    // might be running in a mode where it's executing many threads with totally
    // different settings.
    // ----------------------------------------------------------------------------

    private File userSettingsFile;

  TFE_TensorHandleCopyToDevice(h0_task0_new, ctx, remote_device_name, status);
  EXPECT_NE(TF_OK, TF_GetCode(status)) << TF_Message(status);

  // Copying and executing on the new remote device works.
  const char new_remote_device_name[] =
      "/job:worker/replica:0/task:1/device:CPU:0";
  const char new_local_device_name[] =
      "/job:worker/replica:0/task:0/device:CPU:0";

                .build();

        /*
        // This is the desired behaviour, however there are too many crappy poms in the repo and an issue with the
        // timing of executing the interpolation

        try
        {
        new RegexBasedModelInterpolator().interpolate( model, context );
        fail( "Should have failed to interpolate with invalid reference" );
        }

## Execution of dependencies with `yield`

The sequence of execution is more or less like this diagram. Time flows from top to bottom. And each column is one of the parts interacting or executing code.

```mermaid
sequenceDiagram

participant client as Client
participant handler as Exception handler

          call.cancel()
          System.out.printf("%.2f Canceled call.%n", (System.nanoTime() - startNanos) / 1e9f)
        }, 1, TimeUnit.SECONDS)

        System.out.printf("%.2f Executing call.%n", (System.nanoTime() - startNanos) / 1e9f)
        try {
          call.execute().use { response ->
            System.out.printf("%.2f Call was expected to fail, but completed: %s%n",

 * #startUp} and {@link #shutDown}) will never run concurrently. Notably, if any execution of {@link
 * #runOneIteration} takes longer than its schedule defines, then subsequent executions may start
 * late. Also, all life cycle methods are executed with a lock held, so subclasses can safely modify
 * shared state without additional synchronization necessary for visibility to later executions of
 * the life cycle methods.
 *
 * <h3>Usage Example</h3>
 *

     * This means that trees/sessions/connections won't be idle-disconnected even if there are no other active
     * references (currently executing code, file descriptors).
     * 
     * Depending on the usage scenario, this may have some benefit as there won't be any delays for restablishing these

   * Like all methods in this interface, this is invoked on the thread that triggered the event. But
   * while other events occur sequentially; cancels may occur concurrently with other events. For
   * example, thread A may be executing [responseBodyStart] while thread B executes [canceled].
   * Implementations must support such concurrent calls.
   *
   * Note that cancellation is best-effort and that a call may proceed normally after it has been

//   - What should be done with requests that exceed the limit?
message LimitedPriorityLevelConfiguration {
  // `assuredConcurrencyShares` (ACS) configures the execution
  // limit, which is a limit on the number of requests of this
  // priority level that may be exeucting at a given time.  ACS must
  // be a positive number. The server's concurrency limit (SCL) is
  // divided among the concurrency-controlled priority levels in

    /*
     * Wait not just until the Future's value is set (as in future.get()) but
     * also until ListeningScheduledExecutorService's wrapper task is done
     * executing listeners, as detected by yielding control to afterExecute.
     */
    completed.await();
    assertTrue(future.isDone());
    assertThat(future.get()).isEqualTo(42);
    assertListenerRunImmediately(future);