Examples with SettableLeaderRetrievalService org.apache.flink.runtime.leaderretrieval.SettableLeaderRetrievalService used on opensource projects

Example 11 with SettableLeaderRetrievalService

use of org.apache.flink.runtime.leaderretrieval.SettableLeaderRetrievalService in project flink by apache.

the class RpcGatewayRetrieverTest method testRpcGatewayRetrieval.

/**
 * Tests that the RpcGatewayRetriever can retrieve the specified gateway type from a leader
 * retrieval service.
 */
@Test
public void testRpcGatewayRetrieval() throws Exception {
    final String expectedValue = "foobar";
    final String expectedValue2 = "barfoo";
    final UUID leaderSessionId = UUID.randomUUID();
    RpcGatewayRetriever<UUID, DummyGateway> gatewayRetriever = new RpcGatewayRetriever<>(rpcService, DummyGateway.class, Function.identity(), TestingRetryStrategies.NO_RETRY_STRATEGY);
    SettableLeaderRetrievalService settableLeaderRetrievalService = new SettableLeaderRetrievalService();
    DummyRpcEndpoint dummyRpcEndpoint = new DummyRpcEndpoint(rpcService, "dummyRpcEndpoint1", expectedValue);
    DummyRpcEndpoint dummyRpcEndpoint2 = new DummyRpcEndpoint(rpcService, "dummyRpcEndpoint2", expectedValue2);
    rpcService.registerGateway(dummyRpcEndpoint.getAddress(), dummyRpcEndpoint.getSelfGateway(DummyGateway.class));
    rpcService.registerGateway(dummyRpcEndpoint2.getAddress(), dummyRpcEndpoint2.getSelfGateway(DummyGateway.class));
    try {
        dummyRpcEndpoint.start();
        dummyRpcEndpoint2.start();
        settableLeaderRetrievalService.start(gatewayRetriever);
        final CompletableFuture<DummyGateway> gatewayFuture = gatewayRetriever.getFuture();
        assertFalse(gatewayFuture.isDone());
        settableLeaderRetrievalService.notifyListener(dummyRpcEndpoint.getAddress(), leaderSessionId);
        final DummyGateway dummyGateway = gatewayFuture.get(TIMEOUT.toMilliseconds(), TimeUnit.MILLISECONDS);
        assertEquals(dummyRpcEndpoint.getAddress(), dummyGateway.getAddress());
        assertEquals(expectedValue, dummyGateway.foobar(TIMEOUT).get(TIMEOUT.toMilliseconds(), TimeUnit.MILLISECONDS));
        // elect a new leader
        settableLeaderRetrievalService.notifyListener(dummyRpcEndpoint2.getAddress(), leaderSessionId);
        final CompletableFuture<DummyGateway> gatewayFuture2 = gatewayRetriever.getFuture();
        final DummyGateway dummyGateway2 = gatewayFuture2.get(TIMEOUT.toMilliseconds(), TimeUnit.MILLISECONDS);
        assertEquals(dummyRpcEndpoint2.getAddress(), dummyGateway2.getAddress());
        assertEquals(expectedValue2, dummyGateway2.foobar(TIMEOUT).get(TIMEOUT.toMilliseconds(), TimeUnit.MILLISECONDS));
    } finally {
        RpcUtils.terminateRpcEndpoints(TIMEOUT, dummyRpcEndpoint, dummyRpcEndpoint2);
    }
}

Example 12 with SettableLeaderRetrievalService

use of org.apache.flink.runtime.leaderretrieval.SettableLeaderRetrievalService in project flink by apache.

the class TaskExecutorRecoveryTest method testRecoveredTaskExecutorWillRestoreAllocationState.

@Test
public void testRecoveredTaskExecutorWillRestoreAllocationState(@TempDir File tempDir) throws Exception {
    final ResourceID resourceId = ResourceID.generate();
    final Configuration configuration = new Configuration();
    configuration.set(TaskManagerOptions.NUM_TASK_SLOTS, 2);
    configuration.set(CheckpointingOptions.LOCAL_RECOVERY, true);
    final TestingResourceManagerGateway testingResourceManagerGateway = new TestingResourceManagerGateway();
    final ArrayBlockingQueue<TaskExecutorSlotReport> queue = new ArrayBlockingQueue<>(2);
    testingResourceManagerGateway.setSendSlotReportFunction(slotReportInformation -> {
        queue.offer(TaskExecutorSlotReport.create(slotReportInformation.f0, slotReportInformation.f2));
        return CompletableFuture.completedFuture(Acknowledge.get());
    });
    final TestingRpcService rpcService = rpcServiceExtension.getTestingRpcService();
    rpcService.registerGateway(testingResourceManagerGateway.getAddress(), testingResourceManagerGateway);
    final JobID jobId = new JobID();
    final TestingHighAvailabilityServices highAvailabilityServices = new TestingHighAvailabilityServices();
    highAvailabilityServices.setResourceManagerLeaderRetriever(new SettableLeaderRetrievalService(testingResourceManagerGateway.getAddress(), testingResourceManagerGateway.getFencingToken().toUUID()));
    final SettableLeaderRetrievalService jobMasterLeaderRetriever = new SettableLeaderRetrievalService();
    highAvailabilityServices.setJobMasterLeaderRetriever(jobId, jobMasterLeaderRetriever);
    final WorkingDirectory workingDirectory = WorkingDirectory.create(tempDir);
    final TaskExecutor taskExecutor = TaskExecutorBuilder.newBuilder(rpcService, highAvailabilityServices, workingDirectory).setConfiguration(configuration).setResourceId(resourceId).build();
    taskExecutor.start();
    final TaskExecutorGateway taskExecutorGateway = taskExecutor.getSelfGateway(TaskExecutorGateway.class);
    final TaskExecutorSlotReport taskExecutorSlotReport = queue.take();
    final SlotReport slotReport = taskExecutorSlotReport.getSlotReport();
    assertThat(slotReport.getNumSlotStatus(), is(2));
    final SlotStatus slotStatus = slotReport.iterator().next();
    final SlotID allocatedSlotID = slotStatus.getSlotID();
    final AllocationID allocationId = new AllocationID();
    taskExecutorGateway.requestSlot(allocatedSlotID, jobId, allocationId, slotStatus.getResourceProfile(), "localhost", testingResourceManagerGateway.getFencingToken(), Time.seconds(10L)).join();
    taskExecutor.close();
    final BlockingQueue<Collection<SlotOffer>> offeredSlots = new ArrayBlockingQueue<>(1);
    final TestingJobMasterGateway jobMasterGateway = new TestingJobMasterGatewayBuilder().setOfferSlotsFunction((resourceID, slotOffers) -> {
        offeredSlots.offer(new HashSet<>(slotOffers));
        return CompletableFuture.completedFuture(slotOffers);
    }).build();
    rpcService.registerGateway(jobMasterGateway.getAddress(), jobMasterGateway);
    jobMasterLeaderRetriever.notifyListener(jobMasterGateway.getAddress(), jobMasterGateway.getFencingToken().toUUID());
    // recover the TaskExecutor
    final TaskExecutor recoveredTaskExecutor = TaskExecutorBuilder.newBuilder(rpcService, highAvailabilityServices, workingDirectory).setConfiguration(configuration).setResourceId(resourceId).build();
    recoveredTaskExecutor.start();
    final TaskExecutorSlotReport recoveredSlotReport = queue.take();
    for (SlotStatus status : recoveredSlotReport.getSlotReport()) {
        if (status.getSlotID().equals(allocatedSlotID)) {
            assertThat(status.getJobID(), is(jobId));
            assertThat(status.getAllocationID(), is(allocationId));
        } else {
            assertThat(status.getJobID(), is(nullValue()));
        }
    }
    final Collection<SlotOffer> take = offeredSlots.take();
    assertThat(take, hasSize(1));
    final SlotOffer offeredSlot = take.iterator().next();
    assertThat(offeredSlot.getAllocationId(), is(allocationId));
}

Example 13 with SettableLeaderRetrievalService

use of org.apache.flink.runtime.leaderretrieval.SettableLeaderRetrievalService in project flink by apache.

the class TaskExecutorSlotLifetimeTest method testUserCodeClassLoaderIsBoundToSlot.

/**
 * Tests that the user code class loader is bound to the lifetime of the slot. This means that
 * it is being reused across a failover, for example. See FLINK-16408.
 */
@Test
public void testUserCodeClassLoaderIsBoundToSlot() throws Exception {
    final Configuration configuration = new Configuration();
    final TestingRpcService rpcService = TESTING_RPC_SERVICE_RESOURCE.getTestingRpcService();
    final TestingResourceManagerGateway resourceManagerGateway = new TestingResourceManagerGateway();
    final CompletableFuture<SlotReport> firstSlotReportFuture = new CompletableFuture<>();
    resourceManagerGateway.setSendSlotReportFunction(resourceIDInstanceIDSlotReportTuple3 -> {
        firstSlotReportFuture.complete(resourceIDInstanceIDSlotReportTuple3.f2);
        return CompletableFuture.completedFuture(Acknowledge.get());
    });
    final BlockingQueue<TaskExecutionState> taskExecutionStates = new ArrayBlockingQueue<>(3);
    final OneShotLatch slotsOfferedLatch = new OneShotLatch();
    final TestingJobMasterGateway jobMasterGateway = new TestingJobMasterGatewayBuilder().setOfferSlotsFunction((resourceID, slotOffers) -> {
        slotsOfferedLatch.trigger();
        return CompletableFuture.completedFuture(slotOffers);
    }).setUpdateTaskExecutionStateFunction(FunctionUtils.uncheckedFunction(taskExecutionState -> {
        taskExecutionStates.put(taskExecutionState);
        return CompletableFuture.completedFuture(Acknowledge.get());
    })).build();
    final LeaderRetrievalService resourceManagerLeaderRetriever = new SettableLeaderRetrievalService(resourceManagerGateway.getAddress(), resourceManagerGateway.getFencingToken().toUUID());
    final LeaderRetrievalService jobMasterLeaderRetriever = new SettableLeaderRetrievalService(jobMasterGateway.getAddress(), jobMasterGateway.getFencingToken().toUUID());
    final TestingHighAvailabilityServices haServices = new TestingHighAvailabilityServicesBuilder().setResourceManagerLeaderRetriever(resourceManagerLeaderRetriever).setJobMasterLeaderRetrieverFunction(ignored -> jobMasterLeaderRetriever).build();
    rpcService.registerGateway(resourceManagerGateway.getAddress(), resourceManagerGateway);
    rpcService.registerGateway(jobMasterGateway.getAddress(), jobMasterGateway);
    final LocalUnresolvedTaskManagerLocation unresolvedTaskManagerLocation = new LocalUnresolvedTaskManagerLocation();
    try (final TaskExecutor taskExecutor = createTaskExecutor(configuration, rpcService, haServices, unresolvedTaskManagerLocation)) {
        taskExecutor.start();
        final SlotReport slotReport = firstSlotReportFuture.join();
        final SlotID firstSlotId = slotReport.iterator().next().getSlotID();
        final TaskExecutorGateway taskExecutorGateway = taskExecutor.getSelfGateway(TaskExecutorGateway.class);
        final JobID jobId = new JobID();
        final AllocationID allocationId = new AllocationID();
        taskExecutorGateway.requestSlot(firstSlotId, jobId, allocationId, ResourceProfile.ZERO, jobMasterGateway.getAddress(), resourceManagerGateway.getFencingToken(), RpcUtils.INF_TIMEOUT).join();
        final TaskDeploymentDescriptor tdd = TaskDeploymentDescriptorBuilder.newBuilder(jobId, UserClassLoaderExtractingInvokable.class).setAllocationId(allocationId).build();
        slotsOfferedLatch.await();
        taskExecutorGateway.submitTask(tdd, jobMasterGateway.getFencingToken(), RpcUtils.INF_TIMEOUT).join();
        final ClassLoader firstClassLoader = UserClassLoaderExtractingInvokable.take();
        // wait for the first task to finish
        TaskExecutionState taskExecutionState;
        do {
            taskExecutionState = taskExecutionStates.take();
        } while (!taskExecutionState.getExecutionState().isTerminal());
        // check that a second task will re-use the same class loader
        taskExecutorGateway.submitTask(tdd, jobMasterGateway.getFencingToken(), RpcUtils.INF_TIMEOUT).join();
        final ClassLoader secondClassLoader = UserClassLoaderExtractingInvokable.take();
        assertThat(firstClassLoader, sameInstance(secondClassLoader));
    }
}

Example 14 with SettableLeaderRetrievalService

use of org.apache.flink.runtime.leaderretrieval.SettableLeaderRetrievalService in project flink by apache.

the class TaskExecutorTest method setup.

@Before
public void setup() throws IOException {
    rpc = new TestingRpcService();
    configuration = new Configuration();
    TaskExecutorResourceUtils.adjustForLocalExecution(configuration);
    unresolvedTaskManagerLocation = new LocalUnresolvedTaskManagerLocation();
    jobId = new JobID();
    jobId2 = new JobID();
    testingFatalErrorHandler = new TestingFatalErrorHandler();
    haServices = new TestingHighAvailabilityServices();
    resourceManagerLeaderRetriever = new SettableLeaderRetrievalService();
    jobManagerLeaderRetriever = new SettableLeaderRetrievalService();
    jobManagerLeaderRetriever2 = new SettableLeaderRetrievalService();
    haServices.setResourceManagerLeaderRetriever(resourceManagerLeaderRetriever);
    haServices.setJobMasterLeaderRetriever(jobId, jobManagerLeaderRetriever);
    haServices.setJobMasterLeaderRetriever(jobId2, jobManagerLeaderRetriever2);
    nettyShuffleEnvironment = new NettyShuffleEnvironmentBuilder().build();
}

Example 15 with SettableLeaderRetrievalService

use of org.apache.flink.runtime.leaderretrieval.SettableLeaderRetrievalService in project flink by apache.

the class DefaultJobLeaderIdServiceTest method testInitialJobTimeout.

/**
 * Tests that the initial job registration registers a timeout which will call {@link
 * JobLeaderIdActions#notifyJobTimeout(JobID, UUID)} when executed.
 */
@Test
public void testInitialJobTimeout() throws Exception {
    final JobID jobId = new JobID();
    TestingHighAvailabilityServices highAvailabilityServices = new TestingHighAvailabilityServices();
    SettableLeaderRetrievalService leaderRetrievalService = new SettableLeaderRetrievalService(null, null);
    highAvailabilityServices.setJobMasterLeaderRetriever(jobId, leaderRetrievalService);
    ScheduledExecutor scheduledExecutor = mock(ScheduledExecutor.class);
    Time timeout = Time.milliseconds(5000L);
    JobLeaderIdActions jobLeaderIdActions = mock(JobLeaderIdActions.class);
    JobLeaderIdService jobLeaderIdService = new DefaultJobLeaderIdService(highAvailabilityServices, scheduledExecutor, timeout);
    jobLeaderIdService.start(jobLeaderIdActions);
    jobLeaderIdService.addJob(jobId);
    assertTrue(jobLeaderIdService.containsJob(jobId));
    ArgumentCaptor<Runnable> runnableArgumentCaptor = ArgumentCaptor.forClass(Runnable.class);
    verify(scheduledExecutor).schedule(runnableArgumentCaptor.capture(), anyLong(), any(TimeUnit.class));
    Runnable timeoutRunnable = runnableArgumentCaptor.getValue();
    timeoutRunnable.run();
    ArgumentCaptor<UUID> timeoutIdArgumentCaptor = ArgumentCaptor.forClass(UUID.class);
    verify(jobLeaderIdActions, times(1)).notifyJobTimeout(eq(jobId), timeoutIdArgumentCaptor.capture());
    assertTrue(jobLeaderIdService.isValidTimeout(jobId, timeoutIdArgumentCaptor.getValue()));
}