flink-connectors/flink-connector-kafka/src/test/java/org/apache/flink/streaming/connectors/kafka/FlinkKafkaConsumerBaseMigrationTest.java

/*
 * Licensed to the Apache Software Foundation (ASF) under one
 * or more contributor license agreements.  See the NOTICE file
 * distributed with this work for additional information
 * regarding copyright ownership.  The ASF licenses this file
 * to you under the Apache License, Version 2.0 (the
 * "License"); you may not use this file except in compliance
 * with the License.  You may obtain a copy of the License at
 *
 *     http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

package org.apache.flink.streaming.connectors.kafka;

import org.apache.flink.api.common.eventtime.WatermarkStrategy;
import org.apache.flink.core.testutils.OneShotLatch;
import org.apache.flink.metrics.MetricGroup;
import org.apache.flink.runtime.checkpoint.OperatorSubtaskState;
import org.apache.flink.streaming.api.TimeCharacteristic;
import org.apache.flink.streaming.api.functions.source.SourceFunction;
import org.apache.flink.streaming.api.operators.StreamSource;
import org.apache.flink.streaming.api.operators.StreamingRuntimeContext;
import org.apache.flink.streaming.api.watermark.Watermark;
import org.apache.flink.streaming.connectors.kafka.config.OffsetCommitMode;
import org.apache.flink.streaming.connectors.kafka.internals.AbstractFetcher;
import org.apache.flink.streaming.connectors.kafka.internals.AbstractPartitionDiscoverer;
import org.apache.flink.streaming.connectors.kafka.internals.KafkaTopicPartition;
import org.apache.flink.streaming.connectors.kafka.internals.KafkaTopicPartitionStateSentinel;
import org.apache.flink.streaming.connectors.kafka.internals.KafkaTopicsDescriptor;
import org.apache.flink.streaming.util.AbstractStreamOperatorTestHarness;
import org.apache.flink.streaming.util.OperatorSnapshotUtil;
import org.apache.flink.testutils.migration.MigrationVersion;
import org.apache.flink.util.SerializedValue;

import org.junit.Ignore;
import org.junit.Test;
import org.junit.runner.RunWith;
import org.junit.runners.Parameterized;
import org.mockito.invocation.InvocationOnMock;
import org.mockito.stubbing.Answer;

import java.util.ArrayList;
import java.util.Arrays;
import java.util.Collection;
import java.util.Collections;
import java.util.HashMap;
import java.util.List;
import java.util.Map;
import java.util.stream.Collectors;

import static org.junit.Assert.assertEquals;
import static org.junit.Assert.assertTrue;
import static org.mockito.Matchers.any;
import static org.mockito.Mockito.mock;
import static org.powermock.api.mockito.PowerMockito.doAnswer;
import static org.powermock.api.mockito.PowerMockito.when;

/**
 * Tests for checking whether {@link FlinkKafkaConsumerBase} can restore from snapshots that were
 * done using previous Flink versions' {@link FlinkKafkaConsumerBase}.
 *
 * <p>For regenerating the binary snapshot files run {@link #writeSnapshot()} on the corresponding
 * Flink release-* branch.
 */
@RunWith(Parameterized.class)
public class FlinkKafkaConsumerBaseMigrationTest {

    /**
     * TODO change this to the corresponding savepoint version to be written (e.g. {@link
     * MigrationVersion#v1_3} for 1.3) TODO and remove all @Ignore annotations on write*Snapshot()
     * methods to generate savepoints TODO Note: You should generate the savepoint based on the
     * release branch instead of the master.
     */
    private final MigrationVersion flinkGenerateSavepointVersion = null;

    private static final HashMap<KafkaTopicPartition, Long> PARTITION_STATE = new HashMap<>();

    static {
        PARTITION_STATE.put(new KafkaTopicPartition("abc", 13), 16768L);
        PARTITION_STATE.put(new KafkaTopicPartition("def", 7), 987654321L);
    }

    private static final List<String> TOPICS =
            new ArrayList<>(PARTITION_STATE.keySet())
                    .stream().map(p -> p.getTopic()).distinct().collect(Collectors.toList());

    private final MigrationVersion testMigrateVersion;

    @Parameterized.Parameters(name = "Migration Savepoint: {0}")
    public static Collection<MigrationVersion> parameters() {
        return Arrays.asList(
                MigrationVersion.v1_4,
                MigrationVersion.v1_5,
                MigrationVersion.v1_6,
                MigrationVersion.v1_7,
                MigrationVersion.v1_8,
                MigrationVersion.v1_9,
                MigrationVersion.v1_10,
                MigrationVersion.v1_11,
                MigrationVersion.v1_12,
                MigrationVersion.v1_13);
    }

    public FlinkKafkaConsumerBaseMigrationTest(MigrationVersion testMigrateVersion) {
        this.testMigrateVersion = testMigrateVersion;
    }

    /** Manually run this to write binary snapshot data. */
    @Ignore
    @Test
    public void writeSnapshot() throws Exception {
        writeSnapshot(
                "src/test/resources/kafka-consumer-migration-test-flink"
                        + flinkGenerateSavepointVersion
                        + "-snapshot",
                PARTITION_STATE);

        final HashMap<KafkaTopicPartition, Long> emptyState = new HashMap<>();
        writeSnapshot(
                "src/test/resources/kafka-consumer-migration-test-flink"
                        + flinkGenerateSavepointVersion
                        + "-empty-state-snapshot",
                emptyState);
    }

    private void writeSnapshot(String path, HashMap<KafkaTopicPartition, Long> state)
            throws Exception {

        final OneShotLatch latch = new OneShotLatch();
        final AbstractFetcher<String, ?> fetcher = mock(AbstractFetcher.class);

        doAnswer(
                        new Answer<Void>() {
                            @Override
                            public Void answer(InvocationOnMock invocation) throws Throwable {
                                latch.trigger();
                                return null;
                            }
                        })
                .when(fetcher)
                .runFetchLoop();

        when(fetcher.snapshotCurrentState()).thenReturn(state);

        final List<KafkaTopicPartition> partitions = new ArrayList<>(PARTITION_STATE.keySet());

        final DummyFlinkKafkaConsumer<String> consumerFunction =
                new DummyFlinkKafkaConsumer<>(
                        fetcher,
                        TOPICS,
                        partitions,
                        FlinkKafkaConsumerBase.PARTITION_DISCOVERY_DISABLED);

        StreamSource<String, DummyFlinkKafkaConsumer<String>> consumerOperator =
                new StreamSource<>(consumerFunction);

        final AbstractStreamOperatorTestHarness<String> testHarness =
                new AbstractStreamOperatorTestHarness<>(consumerOperator, 1, 1, 0);

        testHarness.setTimeCharacteristic(TimeCharacteristic.ProcessingTime);

        testHarness.setup();
        testHarness.open();

        final Throwable[] error = new Throwable[1];

        // run the source asynchronously
        Thread runner =
                new Thread() {
                    @Override
                    public void run() {
                        try {
                            consumerFunction.run(
                                    new DummySourceContext() {
                                        @Override
                                        public void collect(String element) {}
                                    });
                        } catch (Throwable t) {
                            t.printStackTrace();
                            error[0] = t;
                        }
                    }
                };
        runner.start();

        if (!latch.isTriggered()) {
            latch.await();
        }

        final OperatorSubtaskState snapshot;
        synchronized (testHarness.getCheckpointLock()) {
            snapshot = testHarness.snapshot(0L, 0L);
        }

        OperatorSnapshotUtil.writeStateHandle(snapshot, path);

        consumerOperator.close();
        runner.join();
    }

    /** Test restoring from an legacy empty state, when no partitions could be found for topics. */
    @Test
    public void testRestoreFromEmptyStateNoPartitions() throws Exception {
        final DummyFlinkKafkaConsumer<String> consumerFunction =
                new DummyFlinkKafkaConsumer<>(
                        Collections.singletonList("dummy-topic"),
                        Collections.<KafkaTopicPartition>emptyList(),
                        FlinkKafkaConsumerBase.PARTITION_DISCOVERY_DISABLED);

        StreamSource<String, DummyFlinkKafkaConsumer<String>> consumerOperator =
                new StreamSource<>(consumerFunction);

        final AbstractStreamOperatorTestHarness<String> testHarness =
                new AbstractStreamOperatorTestHarness<>(consumerOperator, 1, 1, 0);

        testHarness.setTimeCharacteristic(TimeCharacteristic.ProcessingTime);

        testHarness.setup();

        // restore state from binary snapshot file
        testHarness.initializeState(
                OperatorSnapshotUtil.getResourceFilename(
                        "kafka-consumer-migration-test-flink"
                                + testMigrateVersion
                                + "-empty-state-snapshot"));

        testHarness.open();

        // assert that no partitions were found and is empty
        assertTrue(consumerFunction.getSubscribedPartitionsToStartOffsets() != null);
        assertTrue(consumerFunction.getSubscribedPartitionsToStartOffsets().isEmpty());

        // assert that no state was restored
        assertTrue(consumerFunction.getRestoredState().isEmpty());

        consumerOperator.close();
        consumerOperator.cancel();
    }

    /**
     * Test restoring from an empty state taken using a previous Flink version, when some partitions
     * could be found for topics.
     */
    @Test
    public void testRestoreFromEmptyStateWithPartitions() throws Exception {
        final List<KafkaTopicPartition> partitions = new ArrayList<>(PARTITION_STATE.keySet());

        final DummyFlinkKafkaConsumer<String> consumerFunction =
                new DummyFlinkKafkaConsumer<>(
                        TOPICS, partitions, FlinkKafkaConsumerBase.PARTITION_DISCOVERY_DISABLED);

        StreamSource<String, DummyFlinkKafkaConsumer<String>> consumerOperator =
                new StreamSource<>(consumerFunction);

        final AbstractStreamOperatorTestHarness<String> testHarness =
                new AbstractStreamOperatorTestHarness<>(consumerOperator, 1, 1, 0);

        testHarness.setTimeCharacteristic(TimeCharacteristic.ProcessingTime);

        testHarness.setup();

        // restore state from binary snapshot file
        testHarness.initializeState(
                OperatorSnapshotUtil.getResourceFilename(
                        "kafka-consumer-migration-test-flink"
                                + testMigrateVersion
                                + "-empty-state-snapshot"));

        testHarness.open();

        // the expected state in "kafka-consumer-migration-test-flink1.x-snapshot-empty-state";
        // all new partitions after the snapshot are considered as partitions that were created
        // while the
        // consumer wasn't running, and should start from the earliest offset.
        final HashMap<KafkaTopicPartition, Long> expectedSubscribedPartitionsWithStartOffsets =
                new HashMap<>();
        for (KafkaTopicPartition partition : PARTITION_STATE.keySet()) {
            expectedSubscribedPartitionsWithStartOffsets.put(
                    partition, KafkaTopicPartitionStateSentinel.EARLIEST_OFFSET);
        }

        // assert that there are partitions and is identical to expected list
        assertTrue(consumerFunction.getSubscribedPartitionsToStartOffsets() != null);
        assertTrue(!consumerFunction.getSubscribedPartitionsToStartOffsets().isEmpty());
        assertEquals(
                expectedSubscribedPartitionsWithStartOffsets,
                consumerFunction.getSubscribedPartitionsToStartOffsets());

        // the new partitions should have been considered as restored state
        assertTrue(consumerFunction.getRestoredState() != null);
        assertTrue(!consumerFunction.getSubscribedPartitionsToStartOffsets().isEmpty());
        for (Map.Entry<KafkaTopicPartition, Long> expectedEntry :
                expectedSubscribedPartitionsWithStartOffsets.entrySet()) {
            assertEquals(
                    expectedEntry.getValue(),
                    consumerFunction.getRestoredState().get(expectedEntry.getKey()));
        }

        consumerOperator.close();
        consumerOperator.cancel();
    }

    /**
     * Test restoring from a non-empty state taken using a previous Flink version, when some
     * partitions could be found for topics.
     */
    @Test
    public void testRestore() throws Exception {
        final List<KafkaTopicPartition> partitions = new ArrayList<>(PARTITION_STATE.keySet());

        final DummyFlinkKafkaConsumer<String> consumerFunction =
                new DummyFlinkKafkaConsumer<>(
                        TOPICS, partitions, FlinkKafkaConsumerBase.PARTITION_DISCOVERY_DISABLED);

        StreamSource<String, DummyFlinkKafkaConsumer<String>> consumerOperator =
                new StreamSource<>(consumerFunction);

        final AbstractStreamOperatorTestHarness<String> testHarness =
                new AbstractStreamOperatorTestHarness<>(consumerOperator, 1, 1, 0);

        testHarness.setTimeCharacteristic(TimeCharacteristic.ProcessingTime);

        testHarness.setup();

        // restore state from binary snapshot file
        testHarness.initializeState(
                OperatorSnapshotUtil.getResourceFilename(
                        "kafka-consumer-migration-test-flink" + testMigrateVersion + "-snapshot"));

        testHarness.open();

        // assert that there are partitions and is identical to expected list
        assertTrue(consumerFunction.getSubscribedPartitionsToStartOffsets() != null);
        assertTrue(!consumerFunction.getSubscribedPartitionsToStartOffsets().isEmpty());

        // on restore, subscribedPartitionsToStartOffsets should be identical to the restored state
        assertEquals(PARTITION_STATE, consumerFunction.getSubscribedPartitionsToStartOffsets());

        // assert that state is correctly restored from legacy checkpoint
        assertTrue(consumerFunction.getRestoredState() != null);
        assertEquals(PARTITION_STATE, consumerFunction.getRestoredState());

        consumerOperator.close();
        consumerOperator.cancel();
    }

    // ------------------------------------------------------------------------

    private static class DummyFlinkKafkaConsumer<T> extends FlinkKafkaConsumerBase<T> {
        private static final long serialVersionUID = 1L;

        private final List<KafkaTopicPartition> partitions;

        private final AbstractFetcher<T, ?> fetcher;

        @SuppressWarnings("unchecked")
        DummyFlinkKafkaConsumer(
                AbstractFetcher<T, ?> fetcher,
                List<String> topics,
                List<KafkaTopicPartition> partitions,
                long discoveryInterval) {

            super(
                    topics,
                    null,
                    (KafkaDeserializationSchema<T>) mock(KafkaDeserializationSchema.class),
                    discoveryInterval,
                    false);

            this.fetcher = fetcher;
            this.partitions = partitions;
        }

        DummyFlinkKafkaConsumer(
                List<String> topics, List<KafkaTopicPartition> partitions, long discoveryInterval) {
            this(mock(AbstractFetcher.class), topics, partitions, discoveryInterval);
        }

        @Override
        protected AbstractFetcher<T, ?> createFetcher(
                SourceContext<T> sourceContext,
                Map<KafkaTopicPartition, Long> thisSubtaskPartitionsWithStartOffsets,
                SerializedValue<WatermarkStrategy<T>> watermarkStrategy,
                StreamingRuntimeContext runtimeContext,
                OffsetCommitMode offsetCommitMode,
                MetricGroup consumerMetricGroup,
                boolean useMetrics)
                throws Exception {
            return fetcher;
        }

        @Override
        protected AbstractPartitionDiscoverer createPartitionDiscoverer(
                KafkaTopicsDescriptor topicsDescriptor,
                int indexOfThisSubtask,
                int numParallelSubtasks) {

            AbstractPartitionDiscoverer mockPartitionDiscoverer =
                    mock(AbstractPartitionDiscoverer.class);

            try {
                when(mockPartitionDiscoverer.discoverPartitions()).thenReturn(partitions);
            } catch (Exception e) {
                // ignore
            }
            when(mockPartitionDiscoverer.setAndCheckDiscoveredPartition(
                            any(KafkaTopicPartition.class)))
                    .thenReturn(true);

            return mockPartitionDiscoverer;
        }

        @Override
        protected boolean getIsAutoCommitEnabled() {
            return false;
        }

        @Override
        protected Map<KafkaTopicPartition, Long> fetchOffsetsWithTimestamp(
                Collection<KafkaTopicPartition> partitions, long timestamp) {
            throw new UnsupportedOperationException();
        }
    }

    private abstract static class DummySourceContext
            implements SourceFunction.SourceContext<String> {

        private final Object lock = new Object();

        @Override
        public void collectWithTimestamp(String element, long timestamp) {}

        @Override
        public void emitWatermark(Watermark mark) {}

        @Override
        public Object getCheckpointLock() {
            return lock;
        }

        @Override
        public void close() {}

        @Override
        public void markAsTemporarilyIdle() {}
    }
}