Examples with SegmentDescriptor org.apache.druid.query.SegmentDescriptor used on opensource projects

Example 26 with SegmentDescriptor

use of org.apache.druid.query.SegmentDescriptor in project druid by druid-io.

the class RealtimeIndexTaskTest method testTransformSpec.

@Test(timeout = 60_000L)
public void testTransformSpec() throws Exception {
    final TestIndexerMetadataStorageCoordinator mdc = new TestIndexerMetadataStorageCoordinator();
    final TransformSpec transformSpec = new TransformSpec(new SelectorDimFilter("dim1", "foo", null), ImmutableList.of(new ExpressionTransform("dim1t", "concat(dim1,dim1)", ExprMacroTable.nil())));
    final RealtimeIndexTask task = makeRealtimeTask(null, transformSpec, true, 0);
    final TaskToolbox taskToolbox = makeToolbox(task, mdc, tempFolder.newFolder());
    final ListenableFuture<TaskStatus> statusFuture = runTask(task, taskToolbox);
    final DataSegment publishedSegment;
    // Wait for firehose to show up, it starts off null.
    while (task.getFirehose() == null) {
        Thread.sleep(50);
    }
    final TestFirehose firehose = (TestFirehose) task.getFirehose();
    firehose.addRows(ImmutableList.of(ImmutableMap.of("t", now.getMillis(), "dim1", "foo", "met1", "1"), ImmutableMap.of("t", now.minus(new Period("P1D")).getMillis(), "dim1", "foo", "met1", 2.0), ImmutableMap.of("t", now.getMillis(), "dim2", "bar", "met1", 2.0)));
    // Stop the firehose, this will drain out existing events.
    firehose.close();
    // Wait for publish.
    while (mdc.getPublished().isEmpty()) {
        Thread.sleep(50);
    }
    publishedSegment = Iterables.getOnlyElement(mdc.getPublished());
    // Check metrics.
    Assert.assertEquals(1, task.getMetrics().processed());
    Assert.assertEquals(2, task.getMetrics().thrownAway());
    Assert.assertEquals(0, task.getMetrics().unparseable());
    // Do some queries.
    Assert.assertEquals(1, sumMetric(task, null, "rows").longValue());
    Assert.assertEquals(1, sumMetric(task, new SelectorDimFilter("dim1t", "foofoo", null), "rows").longValue());
    if (NullHandling.replaceWithDefault()) {
        Assert.assertEquals(0, sumMetric(task, new SelectorDimFilter("dim1t", "barbar", null), "rows").longValue());
    } else {
        Assert.assertNull(sumMetric(task, new SelectorDimFilter("dim1t", "barbar", null), "rows"));
    }
    Assert.assertEquals(1, sumMetric(task, null, "met1").longValue());
    // Simulate handoff.
    for (Map.Entry<SegmentDescriptor, Pair<Executor, Runnable>> entry : handOffCallbacks.entrySet()) {
        final Pair<Executor, Runnable> executorRunnablePair = entry.getValue();
        Assert.assertEquals(new SegmentDescriptor(publishedSegment.getInterval(), publishedSegment.getVersion(), publishedSegment.getShardSpec().getPartitionNum()), entry.getKey());
        executorRunnablePair.lhs.execute(executorRunnablePair.rhs);
    }
    handOffCallbacks.clear();
    // Wait for the task to finish.
    final TaskStatus taskStatus = statusFuture.get();
    Assert.assertEquals(TaskState.SUCCESS, taskStatus.getStatusCode());
}

Example 27 with SegmentDescriptor

use of org.apache.druid.query.SegmentDescriptor in project druid by druid-io.

the class RealtimeIndexTaskTest method testRestoreAfterHandoffAttemptDuringShutdown.

@Test(timeout = 60_000L)
public void testRestoreAfterHandoffAttemptDuringShutdown() throws Exception {
    final TaskStorage taskStorage = new HeapMemoryTaskStorage(new TaskStorageConfig(null));
    final TestIndexerMetadataStorageCoordinator mdc = new TestIndexerMetadataStorageCoordinator();
    final File directory = tempFolder.newFolder();
    final RealtimeIndexTask task1 = makeRealtimeTask(null);
    final DataSegment publishedSegment;
    // First run:
    {
        final TaskToolbox taskToolbox = makeToolbox(task1, taskStorage, mdc, directory);
        final ListenableFuture<TaskStatus> statusFuture = runTask(task1, taskToolbox);
        // Wait for firehose to show up, it starts off null.
        while (task1.getFirehose() == null) {
            Thread.sleep(50);
        }
        final TestFirehose firehose = (TestFirehose) task1.getFirehose();
        firehose.addRows(ImmutableList.of(ImmutableMap.of("t", now.getMillis(), "dim1", "foo")));
        // Stop the firehose, this will trigger a finishJob.
        firehose.close();
        // Wait for publish.
        while (mdc.getPublished().isEmpty()) {
            Thread.sleep(50);
        }
        publishedSegment = Iterables.getOnlyElement(mdc.getPublished());
        // Do a query.
        Assert.assertEquals(1, sumMetric(task1, null, "rows").longValue());
        // Trigger graceful shutdown.
        task1.stopGracefully(taskToolbox.getConfig());
        // Wait for the task to finish. The status doesn't really matter.
        while (!statusFuture.isDone()) {
            Thread.sleep(50);
        }
    }
    // Second run:
    {
        final RealtimeIndexTask task2 = makeRealtimeTask(task1.getId());
        final TaskToolbox taskToolbox = makeToolbox(task2, taskStorage, mdc, directory);
        final ListenableFuture<TaskStatus> statusFuture = runTask(task2, taskToolbox);
        // Wait for firehose to show up, it starts off null.
        while (task2.getFirehose() == null) {
            Thread.sleep(50);
        }
        // Stop the firehose again, this will start another handoff.
        final TestFirehose firehose = (TestFirehose) task2.getFirehose();
        // Stop the firehose, this will trigger a finishJob.
        firehose.close();
        // publishedSegment is still published. No reason it shouldn't be.
        Assert.assertEquals(ImmutableSet.of(publishedSegment), mdc.getPublished());
        // Wait for a handoffCallback to show up.
        while (handOffCallbacks.isEmpty()) {
            Thread.sleep(50);
        }
        // Simulate handoff.
        for (Map.Entry<SegmentDescriptor, Pair<Executor, Runnable>> entry : handOffCallbacks.entrySet()) {
            final Pair<Executor, Runnable> executorRunnablePair = entry.getValue();
            Assert.assertEquals(new SegmentDescriptor(publishedSegment.getInterval(), publishedSegment.getVersion(), publishedSegment.getShardSpec().getPartitionNum()), entry.getKey());
            executorRunnablePair.lhs.execute(executorRunnablePair.rhs);
        }
        handOffCallbacks.clear();
        // Wait for the task to finish.
        final TaskStatus taskStatus = statusFuture.get();
        Assert.assertEquals(TaskState.SUCCESS, taskStatus.getStatusCode());
    }
}

Example 28 with SegmentDescriptor

use of org.apache.druid.query.SegmentDescriptor in project druid by druid-io.

the class KafkaIndexTaskTest method testRunOneTaskTwoPartitions.

@Test(timeout = 60_000L)
public void testRunOneTaskTwoPartitions() throws Exception {
    final KafkaIndexTask task = createTask(null, new KafkaIndexTaskIOConfig(0, "sequence0", new SeekableStreamStartSequenceNumbers<>(topic, ImmutableMap.of(0, 2L, 1, 0L), ImmutableSet.of()), new SeekableStreamEndSequenceNumbers<>(topic, ImmutableMap.of(0, 5L, 1, 2L)), kafkaServer.consumerProperties(), KafkaSupervisorIOConfig.DEFAULT_POLL_TIMEOUT_MILLIS, true, null, null, INPUT_FORMAT));
    final ListenableFuture<TaskStatus> future = runTask(task);
    // Insert data
    insertData();
    // Wait for tasks to exit
    Assert.assertEquals(TaskState.SUCCESS, future.get().getStatusCode());
    // Check metrics
    Assert.assertEquals(5, task.getRunner().getRowIngestionMeters().getProcessed());
    Assert.assertEquals(0, task.getRunner().getRowIngestionMeters().getUnparseable());
    Assert.assertEquals(0, task.getRunner().getRowIngestionMeters().getThrownAway());
    // Check published segments & metadata
    SegmentDescriptorAndExpectedDim1Values desc1 = sdd("2010/P1D", 0, ImmutableList.of("c"));
    SegmentDescriptorAndExpectedDim1Values desc2 = sdd("2011/P1D", 0, ImmutableList.of("d", "e", "h"));
    // desc3 will not be created in KafkaIndexTask (0.12.x) as it does not create per Kafka partition Druid segments
    @SuppressWarnings("unused") SegmentDescriptor desc3 = sd("2011/P1D", 1);
    SegmentDescriptorAndExpectedDim1Values desc4 = sdd("2012/P1D", 0, ImmutableList.of("g"));
    assertEqualsExceptVersion(ImmutableList.of(desc1, desc2, desc4), publishedDescriptors());
    Assert.assertEquals(new KafkaDataSourceMetadata(new SeekableStreamEndSequenceNumbers<>(topic, ImmutableMap.of(0, 5L, 1, 2L))), newDataSchemaMetadata());
}

Example 29 with SegmentDescriptor

use of org.apache.druid.query.SegmentDescriptor in project druid by druid-io.

the class KafkaIndexTaskTest method testRunWithTransformSpec.

@Test(timeout = 60_000L)
public void testRunWithTransformSpec() throws Exception {
    final KafkaIndexTask task = createTask(null, NEW_DATA_SCHEMA.withTransformSpec(new TransformSpec(new SelectorDimFilter("dim1", "b", null), ImmutableList.of(new ExpressionTransform("dim1t", "concat(dim1,dim1)", ExprMacroTable.nil())))), new KafkaIndexTaskIOConfig(0, "sequence0", new SeekableStreamStartSequenceNumbers<>(topic, ImmutableMap.of(0, 0L), ImmutableSet.of()), new SeekableStreamEndSequenceNumbers<>(topic, ImmutableMap.of(0, 5L)), kafkaServer.consumerProperties(), KafkaSupervisorIOConfig.DEFAULT_POLL_TIMEOUT_MILLIS, true, null, null, INPUT_FORMAT));
    final ListenableFuture<TaskStatus> future = runTask(task);
    // Wait for the task to start reading
    while (task.getRunner().getStatus() != Status.READING) {
        Thread.sleep(10);
    }
    // Insert data
    insertData();
    // Wait for task to exit
    Assert.assertEquals(TaskState.SUCCESS, future.get().getStatusCode());
    // Check metrics
    Assert.assertEquals(1, task.getRunner().getRowIngestionMeters().getProcessed());
    Assert.assertEquals(0, task.getRunner().getRowIngestionMeters().getUnparseable());
    Assert.assertEquals(4, task.getRunner().getRowIngestionMeters().getThrownAway());
    // Check published metadata
    final List<SegmentDescriptor> publishedDescriptors = publishedDescriptors();
    assertEqualsExceptVersion(ImmutableList.of(sdd("2009/P1D", 0)), publishedDescriptors);
    Assert.assertEquals(new KafkaDataSourceMetadata(new SeekableStreamEndSequenceNumbers<>(topic, ImmutableMap.of(0, 5L))), newDataSchemaMetadata());
    // Check segments in deep storage
    Assert.assertEquals(ImmutableList.of("b"), readSegmentColumn("dim1", publishedDescriptors.get(0)));
    Assert.assertEquals(ImmutableList.of("bb"), readSegmentColumn("dim1t", publishedDescriptors.get(0)));
}

Example 30 with SegmentDescriptor

use of org.apache.druid.query.SegmentDescriptor in project druid by druid-io.

the class KinesisIndexTaskTest method makeToolboxFactory.

private void makeToolboxFactory() throws IOException {
    directory = tempFolder.newFolder();
    final TestUtils testUtils = new TestUtils();
    final ObjectMapper objectMapper = testUtils.getTestObjectMapper();
    objectMapper.setInjectableValues(((InjectableValues.Std) objectMapper.getInjectableValues()).addValue(AWSCredentialsConfig.class, new AWSCredentialsConfig()));
    for (Module module : new KinesisIndexingServiceModule().getJacksonModules()) {
        objectMapper.registerModule(module);
    }
    final TaskConfig taskConfig = new TaskConfig(new File(directory, "baseDir").getPath(), new File(directory, "baseTaskDir").getPath(), null, 50000, null, true, null, null, null, false, false, TaskConfig.BATCH_PROCESSING_MODE_DEFAULT.name());
    final TestDerbyConnector derbyConnector = derby.getConnector();
    derbyConnector.createDataSourceTable();
    derbyConnector.createPendingSegmentsTable();
    derbyConnector.createSegmentTable();
    derbyConnector.createRulesTable();
    derbyConnector.createConfigTable();
    derbyConnector.createTaskTables();
    derbyConnector.createAuditTable();
    taskStorage = new MetadataTaskStorage(derbyConnector, new TaskStorageConfig(null), new DerbyMetadataStorageActionHandlerFactory(derbyConnector, derby.metadataTablesConfigSupplier().get(), objectMapper));
    metadataStorageCoordinator = new IndexerSQLMetadataStorageCoordinator(testUtils.getTestObjectMapper(), derby.metadataTablesConfigSupplier().get(), derbyConnector);
    taskLockbox = new TaskLockbox(taskStorage, metadataStorageCoordinator);
    final TaskActionToolbox taskActionToolbox = new TaskActionToolbox(taskLockbox, taskStorage, metadataStorageCoordinator, emitter, new SupervisorManager(null) {

        @Override
        public boolean checkPointDataSourceMetadata(String supervisorId, int taskGroupId, @Nullable DataSourceMetadata checkpointMetadata) {
            LOG.info("Adding checkpoint hash to the set");
            checkpointRequestsHash.add(Objects.hash(supervisorId, taskGroupId, checkpointMetadata));
            return true;
        }
    });
    final TaskActionClientFactory taskActionClientFactory = new LocalTaskActionClientFactory(taskStorage, taskActionToolbox, new TaskAuditLogConfig(false));
    final SegmentHandoffNotifierFactory handoffNotifierFactory = dataSource -> new SegmentHandoffNotifier() {

        @Override
        public boolean registerSegmentHandoffCallback(SegmentDescriptor descriptor, Executor exec, Runnable handOffRunnable) {
            if (doHandoff) {
                // Simulate immediate handoff
                exec.execute(handOffRunnable);
            }
            return true;
        }

        @Override
        public void start() {
        // Noop
        }

        @Override
        public void close() {
        // Noop
        }
    };
    final LocalDataSegmentPusherConfig dataSegmentPusherConfig = new LocalDataSegmentPusherConfig();
    dataSegmentPusherConfig.storageDirectory = getSegmentDirectory();
    final DataSegmentPusher dataSegmentPusher = new LocalDataSegmentPusher(dataSegmentPusherConfig);
    toolboxFactory = new TaskToolboxFactory(taskConfig, // taskExecutorNode
    null, taskActionClientFactory, emitter, dataSegmentPusher, new TestDataSegmentKiller(), // DataSegmentMover
    null, // DataSegmentArchiver
    null, new TestDataSegmentAnnouncer(), EasyMock.createNiceMock(DataSegmentServerAnnouncer.class), handoffNotifierFactory, this::makeTimeseriesOnlyConglomerate, DirectQueryProcessingPool.INSTANCE, NoopJoinableFactory.INSTANCE, () -> EasyMock.createMock(MonitorScheduler.class), new SegmentCacheManagerFactory(testUtils.getTestObjectMapper()), testUtils.getTestObjectMapper(), testUtils.getTestIndexIO(), MapCache.create(1024), new CacheConfig(), new CachePopulatorStats(), testUtils.getTestIndexMergerV9(), EasyMock.createNiceMock(DruidNodeAnnouncer.class), EasyMock.createNiceMock(DruidNode.class), new LookupNodeService("tier"), new DataNodeService("tier", 1, ServerType.INDEXER_EXECUTOR, 0), new SingleFileTaskReportFileWriter(reportsFile), null, AuthTestUtils.TEST_AUTHORIZER_MAPPER, new NoopChatHandlerProvider(), testUtils.getRowIngestionMetersFactory(), new TestAppenderatorsManager(), new NoopIndexingServiceClient(), null, null, null);
}