pvary commented on code in PR #8553: URL: https://github.com/apache/iceberg/pull/8553#discussion_r1400220361
########## flink/v1.17/flink/src/test/java/org/apache/iceberg/flink/source/TestIcebergSourceWithWatermarkExtractor.java: ########## @@ -0,0 +1,359 @@ +/* + * 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.iceberg.flink.source; + +import static org.apache.flink.connector.testframe.utils.ConnectorTestConstants.DEFAULT_COLLECT_DATA_TIMEOUT; +import static org.assertj.core.api.AssertionsForClassTypes.assertThat; + +import java.io.Serializable; +import java.time.Duration; +import java.time.Instant; +import java.time.LocalDateTime; +import java.time.ZoneId; +import java.util.List; +import java.util.Optional; +import java.util.Set; +import java.util.concurrent.CompletableFuture; +import java.util.concurrent.TimeUnit; +import java.util.stream.Collectors; +import org.apache.flink.api.common.JobID; +import org.apache.flink.api.common.eventtime.SerializableTimestampAssigner; +import org.apache.flink.api.common.eventtime.WatermarkStrategy; +import org.apache.flink.api.common.typeinfo.TypeInformation; +import org.apache.flink.configuration.Configuration; +import org.apache.flink.core.execution.JobClient; +import org.apache.flink.metrics.Gauge; +import org.apache.flink.runtime.metrics.MetricNames; +import org.apache.flink.runtime.minicluster.RpcServiceSharing; +import org.apache.flink.runtime.testutils.InMemoryReporter; +import org.apache.flink.runtime.testutils.MiniClusterResourceConfiguration; +import org.apache.flink.streaming.api.datastream.DataStream; +import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment; +import org.apache.flink.streaming.api.functions.windowing.AllWindowFunction; +import org.apache.flink.streaming.api.windowing.assigners.TumblingEventTimeWindows; +import org.apache.flink.streaming.api.windowing.time.Time; +import org.apache.flink.streaming.api.windowing.windows.TimeWindow; +import org.apache.flink.table.data.RowData; +import org.apache.flink.test.util.MiniClusterWithClientResource; +import org.apache.flink.util.CloseableIterator; +import org.apache.flink.util.Collector; +import org.apache.iceberg.FileFormat; +import org.apache.iceberg.data.GenericAppenderHelper; +import org.apache.iceberg.data.GenericRecord; +import org.apache.iceberg.data.Record; +import org.apache.iceberg.flink.HadoopTableResource; +import org.apache.iceberg.flink.RowDataConverter; +import org.apache.iceberg.flink.TestFixtures; +import org.apache.iceberg.relocated.com.google.common.collect.ImmutableList; +import org.apache.iceberg.relocated.com.google.common.collect.Lists; +import org.apache.iceberg.relocated.com.google.common.collect.Sets; +import org.awaitility.Awaitility; +import org.junit.Assert; +import org.junit.ClassRule; +import org.junit.Rule; +import org.junit.Test; +import org.junit.rules.TemporaryFolder; + +public class TestIcebergSourceWithWatermarkExtractor implements Serializable { + private static final InMemoryReporter reporter = InMemoryReporter.createWithRetainedMetrics(); + private static final int PARALLELISM = 4; + private static final String SOURCE_NAME = "IcebergSource"; + private static final int RECORD_NUM_FOR_2_SPLITS = 200; + + @ClassRule public static final TemporaryFolder TEMPORARY_FOLDER = new TemporaryFolder(); + + @Rule + public final MiniClusterWithClientResource miniClusterResource = + new MiniClusterWithClientResource( + new MiniClusterResourceConfiguration.Builder() + .setNumberTaskManagers(1) + .setNumberSlotsPerTaskManager(PARALLELISM) + .setRpcServiceSharing(RpcServiceSharing.DEDICATED) + .setConfiguration(reporter.addToConfiguration(new Configuration())) + .withHaLeadershipControl() + .build()); + + @Rule + public final HadoopTableResource sourceTableResource = + new HadoopTableResource( + TEMPORARY_FOLDER, TestFixtures.DATABASE, TestFixtures.TABLE, TestFixtures.TS_SCHEMA); + + @Test + public void testWindowing() throws Exception { + GenericAppenderHelper dataAppender = appender(); + List<Record> expectedRecords = Lists.newArrayList(); + + // Generate records with the following pattern: + // - File 1 - Later records (Watermark 6000000) + // - Split 1 - 2 records (100, "file_1-recordTs_100"), (103, "file_1-recordTs_103") + // - File 2 - First records (Watermark 0) + // - Split 1 - 100 records (0, "file_2-recordTs_0"), (1, "file_2-recordTs_1"),... + // - Split 2 - 100 records (0, "file_2-recordTs_0"), (1, "file_2-recordTs_1"),... + // - File 3 - Parallel write for the first records (Watermark 60000) + // - Split 1 - 2 records (1, "file_3-recordTs_1"), (3, "file_3-recordTs_3") + List<Record> batch = ImmutableList.of(generateRecord(100, "100"), generateRecord(103, "103")); + expectedRecords.addAll(batch); + dataAppender.appendToTable(batch); + + batch = Lists.newArrayListWithCapacity(100); + for (int i = 0; i < RECORD_NUM_FOR_2_SPLITS; ++i) { + batch.add(generateRecord(i % 5, "file_2-recordTs_" + i)); + } + expectedRecords.addAll(batch); + dataAppender.appendToTable(batch); + + batch = + ImmutableList.of( + generateRecord(1, "file_3-recordTs_1"), generateRecord(3, "file_3-recordTs_3")); + expectedRecords.addAll(batch); + dataAppender.appendToTable(batch); + + StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment(); + env.setParallelism(2); + + DataStream<RowData> stream = + env.fromSource( + sourceBuilder() + .streaming(true) + .monitorInterval(Duration.ofMillis(10)) + .streamingStartingStrategy(StreamingStartingStrategy.TABLE_SCAN_THEN_INCREMENTAL) + .build(), + WatermarkStrategy.<RowData>noWatermarks() + .withTimestampAssigner(new RowDataTimestampAssigner()), + SOURCE_NAME, + TypeInformation.of(RowData.class)); + DataStream<RowData> windowed = + stream + .windowAll(TumblingEventTimeWindows.of(Time.minutes(5))) + .apply( + new AllWindowFunction<RowData, RowData, TimeWindow>() { + @Override + public void apply( + TimeWindow window, Iterable<RowData> values, Collector<RowData> out) { + // Just print all the data to confirm everything has arrived + values.forEach(out::collect); + } + }); + + try (CloseableIterator<RowData> resultIterator = windowed.collectAsync()) { + env.executeAsync("Iceberg Source Windowing Test"); + + // Write data so the windows containing test data are closed + dataAppender.appendToTable(ImmutableList.of(generateRecord(1500, "last-record"))); + dataAppender.appendToTable(ImmutableList.of(generateRecord(1500, "last-record"))); + dataAppender.appendToTable(ImmutableList.of(generateRecord(1500, "last-record"))); + + assertRecords(resultIterator, expectedRecords); + } + } + + @Test + public void testThrottling() throws Exception { + GenericAppenderHelper dataAppender = appender(); + + // Generate records with the following pattern: + // - File 1 - Later records (Watermark 6000000) + // - Split 1 - 2 records (100, "file_1-recordTs_100"), (103, "file_1-recordTs_103") + // - File 2 - First records (Watermark 0) + // - Split 1 - 100 records (0, "file_2-recordTs_0"), (1, "file_2-recordTs_1"),... + // - Split 2 - 100 records (0, "file_2-recordTs_0"), (1, "file_2-recordTs_1"),... + List<Record> batch; + batch = + ImmutableList.of( + generateRecord(100, "file_1-recordTs_100"), generateRecord(103, "file_1-recordTs_103")); + dataAppender.appendToTable(batch); + + batch = Lists.newArrayListWithCapacity(100); + for (int i = 0; i < RECORD_NUM_FOR_2_SPLITS; ++i) { + batch.add(generateRecord(i % 5, "file_2-recordTs_" + i)); + } + + dataAppender.appendToTable(batch); + + StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment(); + env.setParallelism(2); + + DataStream<RowData> stream = + env.fromSource( + sourceBuilder() + .streaming(true) + .monitorInterval(Duration.ofMillis(10)) + .streamingStartingStrategy(StreamingStartingStrategy.TABLE_SCAN_THEN_INCREMENTAL) + .build(), + WatermarkStrategy.<RowData>noWatermarks() + .withWatermarkAlignment("iceberg", Duration.ofMinutes(20), Duration.ofMillis(10)), + SOURCE_NAME, + TypeInformation.of(RowData.class)); + + try (CloseableIterator<RowData> resultIterator = stream.collectAsync()) { + JobClient jobClient = env.executeAsync("Continuous Iceberg Source Failover Test"); + + // Check that the read the non-blocked data + // The first RECORD_NUM_FOR_2_SPLITS should be read + // 1 or more from the runaway reader should be arrived depending on thread scheduling + waitForRecords(resultIterator, RECORD_NUM_FOR_2_SPLITS + 1); + + // Get the drift metric, wait for it to be created and reach the expected state + // (100 min - 20 min - 0 min) Review Comment: Exact testing is not working for throttling, as there is an inherently time sensitive element in global watermark propagation which happens asynchronously. So we can not be sure how many of the too early records will be propagated before the throttling kicks in, and the reader is stopped. That is the reason why even Flink internal tests are depend on the the `WATERMARK_ALIGNMENT_DRIFT` metrics to check if the throttling is happened. We follow the same pattern here: - Add some new records, add some old records - Verify that the old records are received fully - Check the alignment metrics - Add some more new records - Verify that all the records are received - Verify the alignment metrics -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: issues-unsubscr...@iceberg.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@iceberg.apache.org For additional commands, e-mail: issues-h...@iceberg.apache.org
