1
0

[HUDI-1325] [RFC-15] Merge updates of unsynced instants to metadata table (apache#2342)

[RFC-15] Fix partition key in metadata table when bootstrapping from file system (apache#2387)

Co-authored-by: Ryan Pifer <ryanpife@amazon.com>
This commit is contained in:
Ryan Pifer
2020-12-31 08:57:13 -08:00
committed by vinoth chandar
parent 2bd4a68731
commit 4b94529aaf
11 changed files with 909 additions and 468 deletions

View File

@@ -37,6 +37,7 @@ import org.apache.hudi.testutils.HoodieClientTestBase;
import org.apache.log4j.LogManager;
import org.apache.log4j.Logger;
import org.junit.jupiter.api.AfterEach;
import org.junit.jupiter.api.BeforeEach;
import org.junit.jupiter.api.Test;
@@ -70,6 +71,11 @@ public class TestCompactionAdminClient extends HoodieClientTestBase {
client = new CompactionAdminClient(context, basePath);
}
@AfterEach
public void cleanUp() throws Exception {
cleanupResources();
}
@Test
public void testUnscheduleCompactionPlan() throws Exception {
int numEntriesPerInstant = 10;

View File

@@ -79,8 +79,8 @@ import org.junit.jupiter.api.io.TempDir;
import org.junit.jupiter.params.ParameterizedTest;
import org.junit.jupiter.params.provider.ValueSource;
public class TestHoodieFsMetadata extends HoodieClientTestHarness {
private static final Logger LOG = LogManager.getLogger(TestHoodieFsMetadata.class);
public class TestHoodieBackedMetadata extends HoodieClientTestHarness {
private static final Logger LOG = LogManager.getLogger(TestHoodieBackedMetadata.class);
@TempDir
public java.nio.file.Path tempFolder;
@@ -95,7 +95,7 @@ public class TestHoodieFsMetadata extends HoodieClientTestHarness {
initSparkContexts("TestHoodieMetadata");
initFileSystem();
fs.mkdirs(new Path(basePath));
initMetaClient();
initMetaClient(tableType);
initTestDataGenerator();
metadataTableBasePath = HoodieTableMetadata.getMetadataTableBasePath(basePath);
@@ -371,7 +371,41 @@ public class TestHoodieFsMetadata extends HoodieClientTestHarness {
client.syncTableMetadata();
validateMetadata(client);
}
}
/**
* Test when syncing rollback to metadata if the commit being rolled back has not been synced that essentially a no-op
* occurs to metadata.
* @throws Exception
*/
@Test
public void testRollbackUnsyncedCommit() throws Exception {
init(HoodieTableType.COPY_ON_WRITE);
HoodieSparkEngineContext engineContext = new HoodieSparkEngineContext(jsc);
try (SparkRDDWriteClient client = new SparkRDDWriteClient(engineContext, getWriteConfig(true, true))) {
// Initialize table with metadata
String newCommitTime = HoodieActiveTimeline.createNewInstantTime();
List<HoodieRecord> records = dataGen.generateInserts(newCommitTime, 20);
client.startCommitWithTime(newCommitTime);
List<WriteStatus> writeStatuses = client.bulkInsert(jsc.parallelize(records, 1), newCommitTime).collect();
assertNoWriteErrors(writeStatuses);
validateMetadata(client);
}
String newCommitTime = HoodieActiveTimeline.createNewInstantTime();
try (SparkRDDWriteClient client = new SparkRDDWriteClient(engineContext, getWriteConfig(true, false))) {
// Commit with metadata disabled
client.startCommitWithTime(newCommitTime);
List<HoodieRecord> records = dataGen.generateUpdates(newCommitTime, 10);
List<WriteStatus> writeStatuses = client.upsert(jsc.parallelize(records, 1), newCommitTime).collect();
assertNoWriteErrors(writeStatuses);
client.rollback(newCommitTime);
}
try (SparkRDDWriteClient client = new SparkRDDWriteClient<>(engineContext, getWriteConfig(true, true))) {
validateMetadata(client);
}
}
/**
@@ -636,6 +670,84 @@ public class TestHoodieFsMetadata extends HoodieClientTestHarness {
}
}
/**
* Test when reading from metadata table which is out of sync with dataset that results are still consistent.
*/
// @ParameterizedTest
// @EnumSource(HoodieTableType.class)
@Test
public void testMetadataOutOfSync() throws Exception {
init(HoodieTableType.COPY_ON_WRITE);
HoodieSparkEngineContext engineContext = new HoodieSparkEngineContext(jsc);
SparkRDDWriteClient unsyncedClient = new SparkRDDWriteClient(engineContext, getWriteConfig(true, true));
// Enable metadata so table is initialized
try (SparkRDDWriteClient client = new SparkRDDWriteClient(engineContext, getWriteConfig(true, true))) {
// Perform Bulk Insert
String newCommitTime = "001";
client.startCommitWithTime(newCommitTime);
List<HoodieRecord> records = dataGen.generateInserts(newCommitTime, 20);
client.bulkInsert(jsc.parallelize(records, 1), newCommitTime).collect();
}
// Perform commit operations with metadata disabled
try (SparkRDDWriteClient client = new SparkRDDWriteClient(engineContext, getWriteConfig(true, false))) {
// Perform Insert
String newCommitTime = "002";
client.startCommitWithTime(newCommitTime);
List<HoodieRecord> records = dataGen.generateInserts(newCommitTime, 20);
client.insert(jsc.parallelize(records, 1), newCommitTime).collect();
// Perform Upsert
newCommitTime = "003";
client.startCommitWithTime(newCommitTime);
records = dataGen.generateUniqueUpdates(newCommitTime, 20);
client.upsert(jsc.parallelize(records, 1), newCommitTime).collect();
// Compaction
if (metaClient.getTableType() == HoodieTableType.MERGE_ON_READ) {
newCommitTime = "004";
client.scheduleCompactionAtInstant(newCommitTime, Option.empty());
client.compact(newCommitTime);
}
}
assertFalse(metadata(unsyncedClient).isInSync());
validateMetadata(unsyncedClient);
// Perform clean operation with metadata disabled
try (SparkRDDWriteClient client = new SparkRDDWriteClient(engineContext, getWriteConfig(true, false))) {
// One more commit needed to trigger clean so upsert and compact
String newCommitTime = "005";
client.startCommitWithTime(newCommitTime);
List<HoodieRecord> records = dataGen.generateUpdates(newCommitTime, 20);
client.upsert(jsc.parallelize(records, 1), newCommitTime).collect();
if (metaClient.getTableType() == HoodieTableType.MERGE_ON_READ) {
newCommitTime = "006";
client.scheduleCompactionAtInstant(newCommitTime, Option.empty());
client.compact(newCommitTime);
}
// Clean
newCommitTime = "007";
client.clean(newCommitTime);
}
assertFalse(metadata(unsyncedClient).isInSync());
validateMetadata(unsyncedClient);
// Perform restore with metadata disabled
try (SparkRDDWriteClient client = new SparkRDDWriteClient(engineContext, getWriteConfig(true, false))) {
client.restoreToInstant("004");
}
assertFalse(metadata(unsyncedClient).isInSync());
validateMetadata(unsyncedClient);
}
/**
* Validate the metadata tables contents to ensure it matches what is on the file system.
*
@@ -643,8 +755,9 @@ public class TestHoodieFsMetadata extends HoodieClientTestHarness {
*/
private void validateMetadata(SparkRDDWriteClient client) throws IOException {
HoodieWriteConfig config = client.getConfig();
HoodieBackedTableMetadataWriter metadataWriter = metadataWriter(client);
assertNotNull(metadataWriter, "MetadataWriter should have been initialized");
HoodieBackedTableMetadata tableMetadata = metadata(client);
assertNotNull(tableMetadata, "MetadataReader should have been initialized");
if (!config.useFileListingMetadata()) {
return;
}
@@ -652,17 +765,9 @@ public class TestHoodieFsMetadata extends HoodieClientTestHarness {
HoodieTimer timer = new HoodieTimer().startTimer();
HoodieSparkEngineContext engineContext = new HoodieSparkEngineContext(jsc);
// Validate write config for metadata table
HoodieWriteConfig metadataWriteConfig = metadataWriter.getWriteConfig();
assertFalse(metadataWriteConfig.useFileListingMetadata(), "No metadata table for metadata table");
assertFalse(metadataWriteConfig.getFileListingMetadataVerify(), "No verify for metadata table");
// Metadata table should be in sync with the dataset
assertTrue(metadata(client).isInSync());
// Partitions should match
List<String> fsPartitions = FSUtils.getAllFoldersWithPartitionMetaFile(fs, basePath);
List<String> metadataPartitions = metadataWriter.metadata().getAllPartitionPaths();
List<String> metadataPartitions = tableMetadata.getAllPartitionPaths();
Collections.sort(fsPartitions);
Collections.sort(metadataPartitions);
@@ -684,7 +789,7 @@ public class TestHoodieFsMetadata extends HoodieClientTestHarness {
partitionPath = new Path(basePath, partition);
}
FileStatus[] fsStatuses = FSUtils.getAllDataFilesInPartition(fs, partitionPath);
FileStatus[] metaStatuses = metadataWriter.metadata().getAllFilesInPartition(partitionPath);
FileStatus[] metaStatuses = tableMetadata.getAllFilesInPartition(partitionPath);
List<String> fsFileNames = Arrays.stream(fsStatuses)
.map(s -> s.getPath().getName()).collect(Collectors.toList());
List<String> metadataFilenames = Arrays.stream(metaStatuses)
@@ -705,9 +810,9 @@ public class TestHoodieFsMetadata extends HoodieClientTestHarness {
// FileSystemView should expose the same data
List<HoodieFileGroup> fileGroups = tableView.getAllFileGroups(partition).collect(Collectors.toList());
fileGroups.forEach(g -> LogManager.getLogger(TestHoodieFsMetadata.class).info(g));
fileGroups.forEach(g -> g.getAllBaseFiles().forEach(b -> LogManager.getLogger(TestHoodieFsMetadata.class).info(b)));
fileGroups.forEach(g -> g.getAllFileSlices().forEach(s -> LogManager.getLogger(TestHoodieFsMetadata.class).info(s)));
fileGroups.forEach(g -> LogManager.getLogger(TestHoodieBackedMetadata.class).info(g));
fileGroups.forEach(g -> g.getAllBaseFiles().forEach(b -> LogManager.getLogger(TestHoodieBackedMetadata.class).info(b)));
fileGroups.forEach(g -> g.getAllFileSlices().forEach(s -> LogManager.getLogger(TestHoodieBackedMetadata.class).info(s)));
long numFiles = fileGroups.stream()
.mapToLong(g -> g.getAllBaseFiles().count() + g.getAllFileSlices().mapToLong(s -> s.getLogFiles().count()).sum())
@@ -720,10 +825,17 @@ public class TestHoodieFsMetadata extends HoodieClientTestHarness {
}
});
HoodieTableMetaClient metadataMetaClient = new HoodieTableMetaClient(hadoopConf, metadataTableBasePath);
HoodieBackedTableMetadataWriter metadataWriter = metadataWriter(client);
assertNotNull(metadataWriter, "MetadataWriter should have been initialized");
// Validate write config for metadata table
HoodieWriteConfig metadataWriteConfig = metadataWriter.getWriteConfig();
assertFalse(metadataWriteConfig.useFileListingMetadata(), "No metadata table for metadata table");
assertFalse(metadataWriteConfig.getFileListingMetadataVerify(), "No verify for metadata table");
// Metadata table should be in sync with the dataset
assertTrue(metadataWriter.metadata().isInSync());
assertTrue(metadata(client).isInSync());
HoodieTableMetaClient metadataMetaClient = new HoodieTableMetaClient(hadoopConf, metadataTableBasePath);
// Metadata table is MOR
assertEquals(metadataMetaClient.getTableType(), HoodieTableType.MERGE_ON_READ, "Metadata Table should be MOR");

View File

@@ -46,6 +46,7 @@ import org.apache.hadoop.fs.Path;
import org.apache.spark.api.java.JavaRDD;
import org.apache.spark.sql.Dataset;
import org.apache.spark.sql.Row;
import org.junit.jupiter.api.AfterEach;
import org.junit.jupiter.api.BeforeEach;
import org.junit.jupiter.api.Test;
import org.junit.jupiter.params.ParameterizedTest;
@@ -91,6 +92,11 @@ public class TestUpgradeDowngrade extends HoodieClientTestBase {
initDFSMetaClient();
}
@AfterEach
public void cleanUp() throws Exception {
cleanupResources();
}
@Test
public void testLeftOverUpdatedPropFileCleanup() throws IOException {
testUpgradeInternal(true, true, HoodieTableType.MERGE_ON_READ);

View File

@@ -24,6 +24,7 @@ import org.apache.hudi.client.common.HoodieSparkEngineContext;
import org.apache.hudi.common.fs.FSUtils;
import org.apache.hudi.common.model.HoodieRecord;
import org.apache.hudi.common.model.HoodieRecordLocation;
import org.apache.hudi.common.model.HoodieTableType;
import org.apache.hudi.common.table.HoodieTableMetaClient;
import org.apache.hudi.common.table.timeline.HoodieTimeline;
import org.apache.hudi.common.table.view.HoodieTableFileSystemView;
@@ -204,6 +205,10 @@ public abstract class HoodieClientTestHarness extends HoodieCommonTestHarness im
* @throws IOException
*/
protected void initMetaClient() throws IOException {
initMetaClient(getTableType());
}
protected void initMetaClient(HoodieTableType tableType) throws IOException {
if (basePath == null) {
throw new IllegalStateException("The base path has not been initialized.");
}
@@ -212,7 +217,7 @@ public abstract class HoodieClientTestHarness extends HoodieCommonTestHarness im
throw new IllegalStateException("The Spark context has not been initialized.");
}
metaClient = HoodieTestUtils.init(context.getHadoopConf().get(), basePath, getTableType());
metaClient = HoodieTestUtils.init(hadoopConf, basePath, tableType);
}
/**