You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@phoenix.apache.org by ka...@apache.org on 2020/03/01 03:07:20 UTC
[phoenix] 01/02: PHOENIX-5709 Base changes for simplifying index
update generation code for consistent global indexes
This is an automated email from the ASF dual-hosted git repository.
kadir pushed a commit to branch PHOENIX-5748-4.x-HBase-1.5
in repository https://gitbox.apache.org/repos/asf/phoenix.git
commit 690944530d2fe27d62720bc9871c86b839cdca9e
Author: Kadir <ko...@salesforce.com>
AuthorDate: Wed Feb 26 10:19:48 2020 -0800
PHOENIX-5709 Base changes for simplifying index update generation code for consistent global indexes
---
.../end2end/ConcurrentMutationsExtendedIT.java | 35 +-
.../phoenix/end2end/ConcurrentMutationsIT.java | 3 +-
.../org/apache/phoenix/end2end/IndexToolIT.java | 15 +-
.../end2end/index/GlobalIndexCheckerIT.java | 70 ++
.../phoenix/compile/ServerBuildIndexCompiler.java | 15 +-
.../coprocessor/IndexRebuildRegionScanner.java | 921 +++++++++++++++------
.../UngroupedAggregateRegionObserver.java | 17 +-
.../phoenix/hbase/index/IndexRegionObserver.java | 855 ++++++++++++-------
.../org/apache/phoenix/index/IndexMaintainer.java | 2 +-
9 files changed, 1350 insertions(+), 583 deletions(-)
diff --git a/phoenix-core/src/it/java/org/apache/phoenix/end2end/ConcurrentMutationsExtendedIT.java b/phoenix-core/src/it/java/org/apache/phoenix/end2end/ConcurrentMutationsExtendedIT.java
index 571961d..546c1fe 100644
--- a/phoenix-core/src/it/java/org/apache/phoenix/end2end/ConcurrentMutationsExtendedIT.java
+++ b/phoenix-core/src/it/java/org/apache/phoenix/end2end/ConcurrentMutationsExtendedIT.java
@@ -25,7 +25,9 @@ import org.apache.hadoop.hbase.coprocessor.RegionCoprocessorEnvironment;
import org.apache.hadoop.hbase.coprocessor.SimpleRegionObserver;
import org.apache.hadoop.hbase.regionserver.MiniBatchOperationInProgress;
import org.apache.hadoop.hbase.util.Bytes;
+
import org.apache.phoenix.jdbc.PhoenixConnection;
+import org.apache.phoenix.mapreduce.index.IndexTool;
import org.apache.phoenix.util.*;
import org.junit.Ignore;
import org.junit.Test;
@@ -51,6 +53,21 @@ public class ConcurrentMutationsExtendedIT extends ParallelStatsDisabledIT {
private final Object lock = new Object();
+ private long verifyIndexTable(String tableName, String indexName, Connection conn) throws Exception {
+ // This checks the state of every raw index row without rebuilding any row
+ IndexToolIT.runIndexTool(true, false, "", tableName, indexName, null,
+ 0, IndexTool.IndexVerifyType.ONLY);
+ // This checks the state of an index row after it is repaired
+ long actualRowCount = IndexScrutiny.scrutinizeIndex(conn, tableName, indexName);
+ // We want to check the index rows again as they may be modified by the read repair
+ IndexToolIT.runIndexTool(true, false, "", tableName, indexName, null,
+ 0, IndexTool.IndexVerifyType.ONLY);
+ // Now we rebuild the entire index table and expect that it is still good after the full rebuild
+ IndexToolIT.runIndexTool(true, false, "", tableName, indexName, null,
+ 0, IndexTool.IndexVerifyType.AFTER);
+ return actualRowCount;
+ }
+
@Test
public void testSynchronousDeletesAndUpsertValues() throws Exception {
final String tableName = generateUniqueName();
@@ -130,7 +147,7 @@ public class ConcurrentMutationsExtendedIT extends ParallelStatsDisabledIT {
t2.start();
doneSignal.await(60, TimeUnit.SECONDS);
- IndexScrutiny.scrutinizeIndex(conn, tableName, indexName);
+ verifyIndexTable(tableName, indexName, conn);
}
@Test
@@ -191,7 +208,7 @@ public class ConcurrentMutationsExtendedIT extends ParallelStatsDisabledIT {
t2.start();
doneSignal.await(60, TimeUnit.SECONDS);
- IndexScrutiny.scrutinizeIndex(conn, tableName, indexName);
+ verifyIndexTable(tableName, indexName, conn);
}
@Test @Repeat(5)
@@ -204,9 +221,10 @@ public class ConcurrentMutationsExtendedIT extends ParallelStatsDisabledIT {
final String indexName = generateUniqueName();
Connection conn = DriverManager.getConnection(getUrl());
conn.createStatement().execute("CREATE TABLE " + tableName
- + "(k1 INTEGER NOT NULL, k2 INTEGER NOT NULL, v1 INTEGER, CONSTRAINT pk PRIMARY KEY (k1,k2)) COLUMN_ENCODED_BYTES = 0, VERSIONS=1");
+ + "(k1 INTEGER NOT NULL, k2 INTEGER NOT NULL, a.v1 INTEGER, b.v2 INTEGER, c.v3 INTEGER, d.v4 INTEGER," +
+ "CONSTRAINT pk PRIMARY KEY (k1,k2)) COLUMN_ENCODED_BYTES = 0, VERSIONS=1");
TestUtil.addCoprocessor(conn, tableName, DelayingRegionObserver.class);
- conn.createStatement().execute("CREATE INDEX " + indexName + " ON " + tableName + "(v1)");
+ conn.createStatement().execute("CREATE INDEX " + indexName + " ON " + tableName + "(v1) INCLUDE(v2, v3)");
final CountDownLatch doneSignal = new CountDownLatch(nThreads);
Runnable[] runnables = new Runnable[nThreads];
for (int i = 0; i < nThreads; i++) {
@@ -216,11 +234,12 @@ public class ConcurrentMutationsExtendedIT extends ParallelStatsDisabledIT {
try {
Connection conn = DriverManager.getConnection(getUrl());
for (int i = 0; i < 10000; i++) {
- boolean isNull = RAND.nextBoolean();
- int randInt = RAND.nextInt() % nIndexValues;
conn.createStatement().execute(
"UPSERT INTO " + tableName + " VALUES (" + (i % nRows) + ", 0, "
- + (isNull ? null : randInt) + ")");
+ + (RAND.nextBoolean() ? null : (RAND.nextInt() % nIndexValues)) + ", "
+ + (RAND.nextBoolean() ? null : RAND.nextInt()) + ", "
+ + (RAND.nextBoolean() ? null : RAND.nextInt()) + ", "
+ + (RAND.nextBoolean() ? null : RAND.nextInt()) + ")");
if ((i % batchSize) == 0) {
conn.commit();
}
@@ -241,7 +260,7 @@ public class ConcurrentMutationsExtendedIT extends ParallelStatsDisabledIT {
}
assertTrue("Ran out of time", doneSignal.await(120, TimeUnit.SECONDS));
- long actualRowCount = IndexScrutiny.scrutinizeIndex(conn, tableName, indexName);
+ long actualRowCount = verifyIndexTable(tableName, indexName, conn);
assertEquals(nRows, actualRowCount);
}
diff --git a/phoenix-core/src/it/java/org/apache/phoenix/end2end/ConcurrentMutationsIT.java b/phoenix-core/src/it/java/org/apache/phoenix/end2end/ConcurrentMutationsIT.java
index f312df0..6562af0 100644
--- a/phoenix-core/src/it/java/org/apache/phoenix/end2end/ConcurrentMutationsIT.java
+++ b/phoenix-core/src/it/java/org/apache/phoenix/end2end/ConcurrentMutationsIT.java
@@ -274,7 +274,8 @@ public class ConcurrentMutationsIT extends ParallelStatsDisabledIT {
EnvironmentEdgeManager.injectEdge(null);
}
}
-
+ @Ignore ("It is not possible to assign the same timestamp two separately committed mutations in the current model\n" +
+ " except when the server time goes backward. In that case, the behavior is not deterministic")
@Test
public void testDeleteRowAndUpsertValueAtSameTS1() throws Exception {
try {
diff --git a/phoenix-core/src/it/java/org/apache/phoenix/end2end/IndexToolIT.java b/phoenix-core/src/it/java/org/apache/phoenix/end2end/IndexToolIT.java
index d3ca67d..30fd018 100644
--- a/phoenix-core/src/it/java/org/apache/phoenix/end2end/IndexToolIT.java
+++ b/phoenix-core/src/it/java/org/apache/phoenix/end2end/IndexToolIT.java
@@ -229,8 +229,6 @@ public class IndexToolIT extends BaseUniqueNamesOwnClusterIT {
assertEquals(NROWS, indexTool.getJob().getCounters().findCounter(INPUT_RECORDS).getValue());
long actualRowCount = IndexScrutiny.scrutinizeIndex(conn, dataTableFullName, indexTableFullName);
assertEquals(NROWS, actualRowCount);
- // Check after compaction
- TestUtil.doMajorCompaction(conn, dataTableFullName);
actualRowCount = IndexScrutiny.scrutinizeIndex(conn, dataTableFullName, indexTableFullName);
assertEquals(NROWS, actualRowCount);
setEveryNthRowWithNull(NROWS, 5, stmt);
@@ -241,7 +239,6 @@ public class IndexToolIT extends BaseUniqueNamesOwnClusterIT {
conn.commit();
actualRowCount = IndexScrutiny.scrutinizeIndex(conn, dataTableFullName, indexTableFullName);
assertEquals(NROWS, actualRowCount);
- TestUtil.doMajorCompaction(conn, dataTableFullName);
actualRowCount = IndexScrutiny.scrutinizeIndex(conn, dataTableFullName, indexTableFullName);
assertEquals(NROWS, actualRowCount);
indexTool = runIndexTool(directApi, useSnapshot, schemaName, dataTableName, indexTableName, null,
@@ -526,7 +523,7 @@ public class IndexToolIT extends BaseUniqueNamesOwnClusterIT {
null, 0, IndexTool.IndexVerifyType.AFTER);
assertEquals(1, MutationCountingRegionObserver.getMutationCount());
MutationCountingRegionObserver.setMutationCount(0);
- // Since all the rows are in the index table, running the index tool with the "-v BEFORE" option should
+ // Since all the rows are in the index table, running the index tool with the "-v BEFORE" option should not
// write any index rows
runIndexTool(directApi, useSnapshot, schemaName, viewName, indexTableName,
null, 0, IndexTool.IndexVerifyType.BEFORE);
@@ -614,16 +611,8 @@ public class IndexToolIT extends BaseUniqueNamesOwnClusterIT {
// Run the index tool to populate the index while verifying rows
runIndexTool(directApi, useSnapshot, schemaName, dataTableName, indexTableName,
null, 0, IndexTool.IndexVerifyType.AFTER);
- // Corrupt one cell by writing directly into the index table
- conn.createStatement().execute("upsert into " + indexTableFullName + " values ('Phoenix', 1, 'B')");
- conn.commit();
- // Run the index tool using the only-verify option to detect this mismatch between the data and index table
runIndexTool(directApi, useSnapshot, schemaName, dataTableName, indexTableName,
- null, -1, IndexTool.IndexVerifyType.ONLY);
- cell = getErrorMessageFromIndexToolOutputTable(conn, dataTableFullName, indexTableFullName);
- expectedValueBytes = Bytes.toBytes("Not matching value for 0:0:CODE E:A A:B");
- assertTrue(Bytes.compareTo(cell.getValueArray(), cell.getValueOffset(), cell.getValueLength(),
- expectedValueBytes, 0, expectedValueBytes.length) == 0);
+ null, 0, IndexTool.IndexVerifyType.ONLY);
dropIndexToolTables(conn);
}
}
diff --git a/phoenix-core/src/it/java/org/apache/phoenix/end2end/index/GlobalIndexCheckerIT.java b/phoenix-core/src/it/java/org/apache/phoenix/end2end/index/GlobalIndexCheckerIT.java
index f9c50fd..35489d0 100644
--- a/phoenix-core/src/it/java/org/apache/phoenix/end2end/index/GlobalIndexCheckerIT.java
+++ b/phoenix-core/src/it/java/org/apache/phoenix/end2end/index/GlobalIndexCheckerIT.java
@@ -117,6 +117,76 @@ public class GlobalIndexCheckerIT extends BaseUniqueNamesOwnClusterIT {
}
@Test
+ public void testDelete() throws Exception {
+ try (Connection conn = DriverManager.getConnection(getUrl())) {
+ String dataTableName = generateUniqueName();
+ populateTable(dataTableName); // with two rows ('a', 'ab', 'abc', 'abcd') and ('b', 'bc', 'bcd', 'bcde')
+ String dml = "DELETE from " + dataTableName + " WHERE id = 'a'";
+ assertEquals(1, conn.createStatement().executeUpdate(dml));
+ conn.commit();
+ String indexTableName = generateUniqueName();
+ conn.createStatement().execute("CREATE INDEX " + indexTableName + " on " +
+ dataTableName + " (val1) include (val2, val3)" + (async ? "ASYNC" : ""));
+ if (async) {
+ // run the index MR job.
+ IndexToolIT.runIndexTool(true, false, null, dataTableName, indexTableName);
+ }
+ // Count the number of index rows
+ String query = "SELECT COUNT(*) from " + indexTableName;
+ // There should be one row in the index table
+ ResultSet rs = conn.createStatement().executeQuery(query);
+ assertTrue(rs.next());
+ assertEquals(1, rs.getInt(1));
+ // Add rows and check everything is still okay
+ verifyTableHealth(conn, dataTableName, indexTableName);
+ }
+ }
+
+ @Test
+ public void testSimulateConcurrentUpdates() throws Exception {
+ try (Connection conn = DriverManager.getConnection(getUrl())) {
+ String dataTableName = generateUniqueName();
+ populateTable(dataTableName); // with two rows ('a', 'ab', 'abc', 'abcd') and ('b', 'bc', 'bcd', 'bcde')
+ String indexTableName = generateUniqueName();
+ conn.createStatement().execute("CREATE INDEX " + indexTableName + " on " +
+ dataTableName + " (val1) include (val2, val3)" + (async ? "ASYNC" : ""));
+ if (async) {
+ // run the index MR job.
+ IndexToolIT.runIndexTool(true, false, null, dataTableName, indexTableName);
+ }
+ // For the concurrent updates on the same row, the last write phase is ignored.
+ // Configure IndexRegionObserver to fail the last write phase (i.e., the post index update phase) where the
+ // verify flag is set to true and/or index rows are deleted and check that this does not impact the
+ // correctness.
+ IndexRegionObserver.setFailPostIndexUpdatesForTesting(true);
+ // Do multiple updates on the same data row
+ conn.createStatement().execute("upsert into " + dataTableName + " (id, val2) values ('a', 'abcc')");
+ conn.createStatement().execute("upsert into " + dataTableName + " (id, val1) values ('a', 'aa')");
+ conn.commit();
+ // The expected state of the index table is {('aa', 'a', 'abcc', 'abcd'), ('bc', 'b', 'bcd', 'bcde')}
+ // Do more multiple updates on the same data row
+ conn.createStatement().execute("upsert into " + dataTableName + " (id, val1, val3) values ('a', null, null)");
+ conn.createStatement().execute("upsert into " + dataTableName + " (id, val1) values ('a', 'ab')");
+ conn.createStatement().execute("upsert into " + dataTableName + " (id, val1) values ('b', 'ab')");
+ conn.createStatement().execute("upsert into " + dataTableName + " (id, val1, val2) values ('b', 'ab', null)");
+ conn.commit();
+ // Now the expected state of the index table is {('ab', 'a', 'abcc' , null), ('ab', 'b', null, 'bcde')}
+ ResultSet rs = conn.createStatement().executeQuery("SELECT * from " + indexTableName);
+ assertTrue(rs.next());
+ assertEquals("ab", rs.getString(1));
+ assertEquals("a", rs.getString(2));
+ assertEquals("abcc", rs.getString(3));
+ assertEquals(null, rs.getString(4));
+ assertTrue(rs.next());
+ assertEquals("ab", rs.getString(1));
+ assertEquals("b", rs.getString(2));
+ assertEquals(null, rs.getString(3));
+ assertEquals("bcde", rs.getString(4));
+ assertFalse(rs.next());
+ }
+ }
+
+ @Test
public void testFailPostIndexDeleteUpdate() throws Exception {
String dataTableName = generateUniqueName();
populateTable(dataTableName); // with two rows ('a', 'ab', 'abc', 'abcd') and ('b', 'bc', 'bcd', 'bcde')
diff --git a/phoenix-core/src/main/java/org/apache/phoenix/compile/ServerBuildIndexCompiler.java b/phoenix-core/src/main/java/org/apache/phoenix/compile/ServerBuildIndexCompiler.java
index 4392e23..99caa6e 100644
--- a/phoenix-core/src/main/java/org/apache/phoenix/compile/ServerBuildIndexCompiler.java
+++ b/phoenix-core/src/main/java/org/apache/phoenix/compile/ServerBuildIndexCompiler.java
@@ -17,27 +17,30 @@
*/
package org.apache.phoenix.compile;
-import java.sql.SQLException;
-import java.util.Collections;
-
import org.apache.hadoop.hbase.Cell;
import org.apache.hadoop.hbase.client.Scan;
import org.apache.hadoop.hbase.io.ImmutableBytesWritable;
import org.apache.phoenix.coprocessor.BaseScannerRegionObserver;
import org.apache.phoenix.coprocessor.MetaDataProtocol;
+import org.apache.phoenix.execute.BaseQueryPlan;
import org.apache.phoenix.execute.MutationState;
import org.apache.phoenix.hbase.index.covered.update.ColumnReference;
import org.apache.phoenix.index.IndexMaintainer;
import org.apache.phoenix.index.PhoenixIndexCodec;
import org.apache.phoenix.jdbc.PhoenixConnection;
import org.apache.phoenix.jdbc.PhoenixStatement;
-import org.apache.phoenix.schema.*;
+import org.apache.phoenix.schema.PTable;
+import org.apache.phoenix.schema.SortOrder;
+import org.apache.phoenix.schema.TableRef;
import org.apache.phoenix.schema.tuple.Tuple;
import org.apache.phoenix.schema.types.PLong;
import org.apache.phoenix.util.ByteUtil;
import org.apache.phoenix.util.IndexUtil;
import org.apache.phoenix.util.ScanUtil;
+import java.sql.SQLException;
+import java.util.Collections;
+
import static org.apache.phoenix.schema.types.PDataType.TRUE_BYTES;
import static org.apache.phoenix.util.IndexUtil.addEmptyColumnToScan;
@@ -96,9 +99,9 @@ public class ServerBuildIndexCompiler {
throw new IllegalArgumentException(
"ServerBuildIndexCompiler does not support global indexes on transactional tables");
}
+ IndexMaintainer indexMaintainer = index.getIndexMaintainer(dataTable, connection);
// By default, we'd use a FirstKeyOnly filter as nothing else needs to be projected for count(*).
// However, in this case, we need to project all of the data columns that contribute to the index.
- IndexMaintainer indexMaintainer = index.getIndexMaintainer(dataTable, connection);
for (ColumnReference columnRef : indexMaintainer.getAllColumns()) {
if (index.getImmutableStorageScheme() == PTable.ImmutableStorageScheme.SINGLE_CELL_ARRAY_WITH_OFFSETS) {
scan.addFamily(columnRef.getFamily());
@@ -121,6 +124,8 @@ public class ServerBuildIndexCompiler {
scan.setAttribute(PhoenixIndexCodec.INDEX_PROTO_MD, ByteUtil.copyKeyBytesIfNecessary(ptr));
scan.setAttribute(BaseScannerRegionObserver.REBUILD_INDEXES, TRUE_BYTES);
ScanUtil.setClientVersion(scan, MetaDataProtocol.PHOENIX_VERSION);
+ scan.setAttribute(BaseScannerRegionObserver.INDEX_REBUILD_PAGING, TRUE_BYTES);
+ BaseQueryPlan.serializeViewConstantsIntoScan(scan, dataTable);
addEmptyColumnToScan(scan, indexMaintainer.getDataEmptyKeyValueCF(), indexMaintainer.getEmptyKeyValueQualifier());
}
if (dataTable.isTransactional()) {
diff --git a/phoenix-core/src/main/java/org/apache/phoenix/coprocessor/IndexRebuildRegionScanner.java b/phoenix-core/src/main/java/org/apache/phoenix/coprocessor/IndexRebuildRegionScanner.java
index 598eb5f..a52bddf 100644
--- a/phoenix-core/src/main/java/org/apache/phoenix/coprocessor/IndexRebuildRegionScanner.java
+++ b/phoenix-core/src/main/java/org/apache/phoenix/coprocessor/IndexRebuildRegionScanner.java
@@ -17,8 +17,8 @@
*/
package org.apache.phoenix.coprocessor;
+import static org.apache.phoenix.hbase.index.IndexRegionObserver.UNVERIFIED_BYTES;
import static org.apache.phoenix.hbase.index.IndexRegionObserver.VERIFIED_BYTES;
-import static org.apache.phoenix.hbase.index.IndexRegionObserver.removeEmptyColumn;
import static org.apache.phoenix.hbase.index.write.AbstractParallelWriterIndexCommitter.INDEX_WRITER_KEEP_ALIVE_TIME_CONF_KEY;
import static org.apache.phoenix.mapreduce.index.IndexTool.AFTER_REBUILD_EXPIRED_INDEX_ROW_COUNT_BYTES;
import static org.apache.phoenix.mapreduce.index.IndexTool.AFTER_REBUILD_INVALID_INDEX_ROW_COUNT_BYTES;
@@ -42,13 +42,15 @@ import static org.apache.phoenix.query.QueryServices.MUTATE_BATCH_SIZE_BYTES_ATT
import java.io.IOException;
import java.util.ArrayList;
import java.util.Arrays;
+import java.util.Collections;
+import java.util.Comparator;
import java.util.Iterator;
import java.util.List;
import java.util.Map;
-import java.util.Map.Entry;
-import java.util.Set;
+import java.util.NavigableSet;
import java.util.concurrent.ExecutionException;
+import com.google.common.collect.Lists;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.hbase.Cell;
import org.apache.hadoop.hbase.CellUtil;
@@ -70,6 +72,7 @@ import org.apache.hadoop.hbase.regionserver.Region;
import org.apache.hadoop.hbase.regionserver.RegionScanner;
import org.apache.hadoop.hbase.util.Bytes;
import org.apache.hadoop.hbase.util.EnvironmentEdgeManager;
+import org.apache.hadoop.hbase.util.Pair;
import org.apache.phoenix.cache.ServerCacheClient;
import org.apache.phoenix.compile.ScanRanges;
import org.apache.phoenix.filter.SkipScanFilter;
@@ -85,6 +88,7 @@ import org.apache.phoenix.hbase.index.parallel.WaitForCompletionTaskRunner;
import org.apache.phoenix.hbase.index.table.HTableFactory;
import org.apache.phoenix.hbase.index.util.GenericKeyValueBuilder;
import org.apache.phoenix.hbase.index.util.ImmutableBytesPtr;
+import org.apache.phoenix.hbase.index.util.IndexManagementUtil;
import org.apache.phoenix.index.GlobalIndexChecker;
import org.apache.phoenix.index.IndexMaintainer;
import org.apache.phoenix.index.PhoenixIndexCodec;
@@ -93,6 +97,7 @@ import org.apache.phoenix.query.KeyRange;
import org.apache.phoenix.query.QueryServicesOptions;
import org.apache.phoenix.schema.types.PLong;
import org.apache.phoenix.schema.types.PVarbinary;
+import org.apache.phoenix.util.IndexUtil;
import org.apache.phoenix.util.KeyValueUtil;
import org.apache.phoenix.util.ServerUtil;
import org.slf4j.Logger;
@@ -100,7 +105,6 @@ import org.slf4j.LoggerFactory;
import com.google.common.base.Throwables;
import com.google.common.collect.Maps;
-import com.google.common.collect.Sets;
public class IndexRebuildRegionScanner extends BaseRegionScanner {
@@ -309,20 +313,14 @@ public class IndexRebuildRegionScanner extends BaseRegionScanner {
public boolean isVerificationFailed(IndexTool.IndexVerifyType verifyType) {
if (verifyType == IndexTool.IndexVerifyType.BEFORE || verifyType == IndexTool.IndexVerifyType.NONE) {
return false;
- }
- if (verifyType == IndexTool.IndexVerifyType.ONLY) {
- if (before.validIndexRowCount + before.expiredIndexRowCount != scannedDataRowCount) {
+ } else if (verifyType == IndexTool.IndexVerifyType.ONLY) {
+ if (before.invalidIndexRowCount + before.missingIndexRowCount > 0) {
return true;
}
- }
- if (verifyType == IndexTool.IndexVerifyType.BOTH || verifyType == IndexTool.IndexVerifyType.AFTER) {
+ } else if (verifyType == IndexTool.IndexVerifyType.BOTH || verifyType == IndexTool.IndexVerifyType.AFTER) {
if (after.invalidIndexRowCount + after.missingIndexRowCount > 0) {
return true;
}
- if (before.validIndexRowCount + before.expiredIndexRowCount +
- after.expiredIndexRowCount + after.validIndexRowCount != scannedDataRowCount) {
- return true;
- }
}
return false;
}
@@ -360,8 +358,8 @@ public class IndexRebuildRegionScanner extends BaseRegionScanner {
private Table resultHTable = null;
private IndexTool.IndexVerifyType verifyType = IndexTool.IndexVerifyType.NONE;
private boolean verify = false;
- private Map<byte[], Put> indexKeyToDataPutMap;
- private Map<byte[], Put> dataKeyToDataPutMap;
+ private Map<byte[], List<Mutation>> indexKeyToMutationMap;
+ private Map<byte[], Pair<Put, Delete>> dataKeyToMutationMap;
private TaskRunner pool;
private TaskBatch<Boolean> tasks;
private String exceptionMessage;
@@ -371,15 +369,21 @@ public class IndexRebuildRegionScanner extends BaseRegionScanner {
private int indexTableTTL;
private VerificationResult verificationResult;
private boolean isBeforeRebuilt = true;
-
- IndexRebuildRegionScanner (final RegionScanner innerScanner, final Region region, final Scan scan,
- final RegionCoprocessorEnvironment env,
- UngroupedAggregateRegionObserver ungroupedAggregateRegionObserver) throws IOException {
+ private boolean partialRebuild = false;
+ private int singleRowRebuildReturnCode;
+ private Map<byte[], NavigableSet<byte[]>> familyMap;
+ private byte[][] viewConstants;
+
+ IndexRebuildRegionScanner(final RegionScanner innerScanner, final Region region, final Scan scan,
+ final RegionCoprocessorEnvironment env,
+ UngroupedAggregateRegionObserver ungroupedAggregateRegionObserver) throws IOException {
super(innerScanner);
final Configuration config = env.getConfiguration();
if (scan.getAttribute(BaseScannerRegionObserver.INDEX_REBUILD_PAGING) != null) {
pageSizeInRows = config.getLong(INDEX_REBUILD_PAGE_SIZE_IN_ROWS,
QueryServicesOptions.DEFAULT_INDEX_REBUILD_PAGE_SIZE_IN_ROWS);
+ } else {
+ partialRebuild = true;
}
maxBatchSize = config.getInt(MUTATE_BATCH_SIZE_ATTRIB, QueryServicesOptions.DEFAULT_MUTATE_BATCH_SIZE);
mutations = new UngroupedAggregateRegionObserver.MutationList(maxBatchSize);
@@ -392,31 +396,38 @@ public class IndexRebuildRegionScanner extends BaseRegionScanner {
useProto = false;
indexMetaData = scan.getAttribute(PhoenixIndexCodec.INDEX_MD);
}
- if (!scan.isRaw()) {
- // No need to deserialize index maintainers when the scan is raw. Raw scan is used by partial rebuilds
- List<IndexMaintainer> maintainers = IndexMaintainer.deserialize(indexMetaData, true);
- indexMaintainer = maintainers.get(0);
- }
+ List<IndexMaintainer> maintainers = IndexMaintainer.deserialize(indexMetaData, true);
+ indexMaintainer = maintainers.get(0);
this.scan = scan;
+ familyMap = scan.getFamilyMap();
+ if (familyMap.isEmpty()) {
+ familyMap = null;
+ }
+
this.innerScanner = innerScanner;
this.region = region;
this.env = env;
this.ungroupedAggregateRegionObserver = ungroupedAggregateRegionObserver;
indexRowKey = scan.getAttribute(BaseScannerRegionObserver.INDEX_ROW_KEY);
+ if (indexRowKey != null) {
+ setReturnCodeForSingleRowRebuild();
+ pageSizeInRows = 1;
+ }
byte[] valueBytes = scan.getAttribute(BaseScannerRegionObserver.INDEX_REBUILD_VERIFY_TYPE);
if (valueBytes != null) {
verificationResult = new VerificationResult();
verifyType = IndexTool.IndexVerifyType.fromValue(valueBytes);
if (verifyType != IndexTool.IndexVerifyType.NONE) {
verify = true;
+ viewConstants = IndexUtil.deserializeViewConstantsFromScan(scan);
// Create the following objects only for rebuilds by IndexTool
hTableFactory = ServerUtil.getDelegateHTableFactory(env, ServerUtil.ConnectionType.INDEX_WRITER_CONNECTION);
indexHTable = hTableFactory.getTable(new ImmutableBytesPtr(indexMaintainer.getIndexTableName()));
indexTableTTL = indexHTable.getTableDescriptor().getColumnFamilies()[0].getTimeToLive();
outputHTable = hTableFactory.getTable(new ImmutableBytesPtr(IndexTool.OUTPUT_TABLE_NAME_BYTES));
resultHTable = hTableFactory.getTable(new ImmutableBytesPtr(IndexTool.RESULT_TABLE_NAME_BYTES));
- indexKeyToDataPutMap = Maps.newTreeMap(Bytes.BYTES_COMPARATOR);
- dataKeyToDataPutMap = Maps.newTreeMap(Bytes.BYTES_COMPARATOR);
+ indexKeyToMutationMap = Maps.newTreeMap(Bytes.BYTES_COMPARATOR);
+ dataKeyToMutationMap = Maps.newTreeMap(Bytes.BYTES_COMPARATOR);
pool = new WaitForCompletionTaskRunner(ThreadPoolManager.getExecutor(
new ThreadPoolBuilder("IndexVerify",
env.getConfiguration()).setMaxThread(NUM_CONCURRENT_INDEX_VERIFY_THREADS_CONF_KEY,
@@ -428,13 +439,38 @@ public class IndexRebuildRegionScanner extends BaseRegionScanner {
}
}
+ private void setReturnCodeForSingleRowRebuild() throws IOException {
+ try (RegionScanner scanner = region.getScanner(scan)) {
+ List<Cell> row = new ArrayList<>();
+ scanner.next(row);
+ // Check if the data table row we have just scanned matches with the index row key.
+ // If not, there is no need to build the index row from this data table row,
+ // and just return zero row count.
+ if (row.isEmpty()) {
+ singleRowRebuildReturnCode = GlobalIndexChecker.RebuildReturnCode.NO_DATA_ROW.getValue();
+ } else {
+ Put put = new Put(CellUtil.cloneRow(row.get(0)));
+ for (Cell cell : row) {
+ put.add(cell);
+ }
+ if (checkIndexRow(indexRowKey, put)) {
+ singleRowRebuildReturnCode = GlobalIndexChecker.RebuildReturnCode.INDEX_ROW_EXISTS.getValue();
+ } else {
+ singleRowRebuildReturnCode = GlobalIndexChecker.RebuildReturnCode.NO_INDEX_ROW.getValue();
+ }
+ }
+ }
+ }
+
@Override
public HRegionInfo getRegionInfo() {
return region.getRegionInfo();
}
@Override
- public boolean isFilterDone() { return false; }
+ public boolean isFilterDone() {
+ return false;
+ }
private void logToIndexToolResultTable() throws IOException {
long scanMaxTs = scan.getTimeRange().getMax();
@@ -505,38 +541,6 @@ public class IndexRebuildRegionScanner extends BaseRegionScanner {
m.setDurability(Durability.SKIP_WAL);
}
- private Delete generateDeleteMarkers(Put put) {
- Set<ColumnReference> allColumns = indexMaintainer.getAllColumns();
- int cellCount = put.size();
- if (cellCount == allColumns.size() + 1) {
- // We have all the columns for the index table. So, no delete marker is needed
- return null;
- }
- Set<ColumnReference> includedColumns = Sets.newLinkedHashSetWithExpectedSize(cellCount);
- long ts = 0;
- for (List<Cell> cells : put.getFamilyCellMap().values()) {
- if (cells == null) {
- break;
- }
- for (Cell cell : cells) {
- includedColumns.add(new ColumnReference(CellUtil.cloneFamily(cell), CellUtil.cloneQualifier(cell)));
- if (ts < cell.getTimestamp()) {
- ts = cell.getTimestamp();
- }
- }
- }
- Delete del = null;
- for (ColumnReference column : allColumns) {
- if (!includedColumns.contains(column)) {
- if (del == null) {
- del = new Delete(put.getRow());
- }
- del.addColumns(column.getFamily(), column.getQualifier(), ts);
- }
- }
- return del;
- }
-
private byte[] commitIfReady(byte[] uuidValue, UngroupedAggregateRegionObserver.MutationList mutationList) throws IOException {
if (ServerUtil.readyToCommit(mutationList.size(), mutationList.byteSize(), maxBatchSize, maxBatchSizeBytes)) {
ungroupedAggregateRegionObserver.checkForRegionClosing();
@@ -547,12 +551,14 @@ public class IndexRebuildRegionScanner extends BaseRegionScanner {
return uuidValue;
}
- private class SimpleValueGetter implements ValueGetter {
+ public static class SimpleValueGetter implements ValueGetter {
final ImmutableBytesWritable valuePtr = new ImmutableBytesWritable();
final Put put;
- SimpleValueGetter (final Put put) {
+
+ public SimpleValueGetter(final Put put) {
this.put = put;
}
+
@Override
public ImmutableBytesWritable getLatestValue(ColumnReference ref, long ts) throws IOException {
List<Cell> cellList = put.get(ref.getFamily(), ref.getQualifier());
@@ -595,7 +601,7 @@ public class IndexRebuildRegionScanner extends BaseRegionScanner {
}
private void logToIndexToolOutputTable(byte[] dataRowKey, byte[] indexRowKey, long dataRowTs, long indexRowTs,
- String errorMsg, byte[] expectedValue, byte[] actualValue) throws IOException {
+ String errorMsg, byte[] expectedValue, byte[] actualValue) throws IOException {
final byte[] E_VALUE_PREFIX_BYTES = Bytes.toBytes(" E:");
final byte[] A_VALUE_PREFIX_BYTES = Bytes.toBytes(" A:");
final int PREFIX_LENGTH = 3;
@@ -643,8 +649,7 @@ public class IndexRebuildRegionScanner extends BaseRegionScanner {
length += PREFIX_LENGTH;
Bytes.putBytes(errorMessageBytes, length, actualValue, 0, actualValue.length);
- }
- else {
+ } else {
errorMessageBytes = Bytes.toBytes(errorMsg);
}
put.addColumn(IndexTool.OUTPUT_TABLE_COLUMN_FAMILY, IndexTool.ERROR_MESSAGE_BYTES, scanMaxTs, errorMessageBytes);
@@ -656,21 +661,11 @@ public class IndexRebuildRegionScanner extends BaseRegionScanner {
outputHTable.put(put);
}
- private long getMaxTimestamp(Result result) {
+ private static long getMaxTimestamp(Mutation m) {
long ts = 0;
- for (Cell cell : result.rawCells()) {
- if (ts < cell.getTimestamp()) {
- ts = cell.getTimestamp();
- }
- }
- return ts;
- }
-
- private long getMaxTimestamp(Put put) {
- long ts = 0;
- for (List<Cell> cells : put.getFamilyCellMap().values()) {
+ for (List<Cell> cells : m.getFamilyCellMap().values()) {
if (cells == null) {
- break;
+ continue;
}
for (Cell cell : cells) {
if (ts < cell.getTimestamp()) {
@@ -681,132 +676,383 @@ public class IndexRebuildRegionScanner extends BaseRegionScanner {
return ts;
}
- private boolean verifySingleIndexRow(Result indexRow, final Put dataRow) throws IOException {
- ValueGetter valueGetter = new SimpleValueGetter(dataRow);
- long ts = getMaxTimestamp(dataRow);
- Put indexPut = indexMaintainer.buildUpdateMutation(GenericKeyValueBuilder.INSTANCE,
- valueGetter, new ImmutableBytesWritable(dataRow.getRow()), ts, null, null);
- if (indexPut == null) {
- // This means the data row does not have any covered column values
- indexPut = new Put(indexRow.getRow());
+ private static Cell getCell(Mutation m, byte[] family, byte[] qualifier) {
+ List<Cell> cellList = m.getFamilyCellMap().get(family);
+ if (cellList == null) {
+ return null;
}
- else {
- // Remove the empty column prepared by Index codec as we need to change its value
- removeEmptyColumn(indexPut, indexMaintainer.getEmptyKeyValueFamily().copyBytesIfNecessary(),
- indexMaintainer.getEmptyKeyValueQualifier());
+ for (Cell cell : cellList) {
+ if (CellUtil.matchingQualifier(cell, qualifier)) {
+ return cell;
+ }
}
- // Add the empty column
- indexPut.addColumn(indexMaintainer.getEmptyKeyValueFamily().copyBytesIfNecessary(),
- indexMaintainer.getEmptyKeyValueQualifier(), ts, VERIFIED_BYTES);
- int cellCount = 0;
- long currentTime = EnvironmentEdgeManager.currentTime();
- for (List<Cell> cells : indexPut.getFamilyCellMap().values()) {
+ return null;
+ }
+
+ private boolean isMatchingMutation(Mutation expected, Mutation actual, int version) throws IOException {
+ if (getTimestamp(expected) != getTimestamp(actual)) {
+ String errorMsg = "Not matching timestamp";
+ byte[] dataKey = indexMaintainer.buildDataRowKey(new ImmutableBytesWritable(expected.getRow()), viewConstants);
+ logToIndexToolOutputTable(dataKey, expected.getRow(), getTimestamp(expected), getTimestamp(actual),
+ errorMsg, null, null);
+ return false;
+ }
+ for (List<Cell> cells : expected.getFamilyCellMap().values()) {
if (cells == null) {
- break;
+ continue;
}
for (Cell expectedCell : cells) {
byte[] family = CellUtil.cloneFamily(expectedCell);
byte[] qualifier = CellUtil.cloneQualifier(expectedCell);
- Cell actualCell = indexRow.getColumnLatestCell(family, qualifier);
- if (actualCell == null) {
- // Check if cell expired as per the current server's time and data table ttl
- // Index table should have the same ttl as the data table, hence we might not
- // get a value back from index if it has already expired between our rebuild and
- // verify
- // TODO: have a metric to update for these cases
- if (isTimestampBeforeTTL(currentTime, expectedCell.getTimestamp())) {
- continue;
- }
- String errorMsg = " Missing cell " + Bytes.toString(family) + ":" +
- Bytes.toString(qualifier);
- logToIndexToolOutputTable(dataRow.getRow(), indexRow.getRow(), ts, getMaxTimestamp(indexRow), errorMsg);
+ Cell actualCell = getCell(actual, family, qualifier);
+ if (actualCell == null ||
+ !CellUtil.matchingType(expectedCell, actualCell)) {
+ byte[] dataKey = indexMaintainer.buildDataRowKey(new ImmutableBytesWritable(expected.getRow()), viewConstants);
+ String errorMsg = "Missing cell (in version " + version + ") " + Bytes.toString(family) + ":" + Bytes.toString(qualifier);
+ logToIndexToolOutputTable(dataKey, expected.getRow(), getTimestamp(expected), getTimestamp(actual), errorMsg);
return false;
}
- if (actualCell.getTimestamp() < ts) {
- // Skip older cells since a Phoenix index row is composed of cells with the same timestamp
- continue;
- }
- // Check all columns
if (!CellUtil.matchingValue(actualCell, expectedCell)) {
- String errorMsg = "Not matching value for " + Bytes.toString(family) + ":" +
- Bytes.toString(qualifier);
- logToIndexToolOutputTable(dataRow.getRow(), indexRow.getRow(), ts, getMaxTimestamp(indexRow),
+ if (Bytes.compareTo(family, indexMaintainer.getEmptyKeyValueFamily().copyBytesIfNecessary()) == 0 &&
+ Bytes.compareTo(qualifier, indexMaintainer.getEmptyKeyValueQualifier()) == 0) {
+ if (Bytes.compareTo(actualCell.getValueArray(), actualCell.getValueOffset(), actualCell.getValueLength(),
+ UNVERIFIED_BYTES, 0, UNVERIFIED_BYTES.length) == 0) {
+ // We will not flag this as mismatch but will log it
+ byte[] dataKey = indexMaintainer.buildDataRowKey(new ImmutableBytesWritable(expected.getRow()), viewConstants);
+ String errorMsg = "Unverified index row (in version " + version + ")";
+ logToIndexToolOutputTable(dataKey, expected.getRow(), getTimestamp(expected), getTimestamp(actual), errorMsg);
+ continue;
+ }
+ }
+ String errorMsg = "Not matching value (in version " + version + ") for " + Bytes.toString(family) + ":" + Bytes.toString(qualifier);
+ byte[] dataKey = indexMaintainer.buildDataRowKey(new ImmutableBytesWritable(expected.getRow()), viewConstants);
+ logToIndexToolOutputTable(dataKey, expected.getRow(), getTimestamp(expected), getTimestamp(actual),
errorMsg, CellUtil.cloneValue(expectedCell), CellUtil.cloneValue(actualCell));
return false;
- } else if (actualCell.getTimestamp() != ts) {
- String errorMsg = "Not matching timestamp for " + Bytes.toString(family) + ":" +
- Bytes.toString(qualifier) + " E: " + ts + " A: " +
- actualCell.getTimestamp();
- logToIndexToolOutputTable(dataRow.getRow(), indexRow.getRow(), ts, getMaxTimestamp(indexRow),
- errorMsg, null, null);
+ }
+ }
+ }
+ return true;
+ }
+
+ private boolean isVerified(Put mutation) throws IOException {
+ List<Cell> cellList = mutation.get(indexMaintainer.getEmptyKeyValueFamily().copyBytesIfNecessary(),
+ indexMaintainer.getEmptyKeyValueQualifier());
+ Cell cell = (cellList != null && !cellList.isEmpty()) ? cellList.get(0) : null;
+ if (cell == null) {
+ throw new DoNotRetryIOException("No empty column cell");
+ }
+ if (Bytes.compareTo(cell.getValueArray(), cell.getValueOffset(), cell.getValueLength(),
+ VERIFIED_BYTES, 0, VERIFIED_BYTES.length) == 0) {
+ return true;
+ }
+ return false;
+ }
+
+ /**
+ * This is to reorder the mutations in descending order by the tuple of timestamp and mutation type where
+ * delete comes before put
+ */
+ public static final Comparator<Mutation> MUTATION_TS_DESC_COMPARATOR = new Comparator<Mutation>() {
+ @Override
+ public int compare(Mutation o1, Mutation o2) {
+ long ts1 = getTimestamp(o1);
+ long ts2 = getTimestamp(o2);
+ if (ts1 > ts2) {
+ return -1;
+ }
+ if (ts1 < ts2) {
+ return 1;
+ }
+ if (o1 instanceof Delete && o2 instanceof Put) {
+ return -1;
+ }
+ if (o1 instanceof Put && o2 instanceof Delete) {
+ return 1;
+ }
+ return 0;
+ }
+ };
+
+ private boolean isDeleteFamily(Mutation mutation) {
+ for (List<Cell> cells : mutation.getFamilyCellMap().values()) {
+ for (Cell cell : cells) {
+ if (KeyValue.Type.codeToType(cell.getTypeByte()) == KeyValue.Type.DeleteFamily) {
+ return true;
+ }
+ }
+ }
+ return false;
+ }
+
+ private boolean isDeleteFamilyVersion(Mutation mutation) {
+ for (List<Cell> cells : mutation.getFamilyCellMap().values()) {
+ for (Cell cell : cells) {
+ if (KeyValue.Type.codeToType(cell.getTypeByte()) == KeyValue.Type.DeleteFamilyVersion) {
+ return true;
+ }
+ }
+ }
+ return false;
+ }
+
+ private static List<Mutation> prepareActualIndexMutations(Result indexRow) throws IOException {
+ Put put = null;
+ Delete del = null;
+ for (Cell cell : indexRow.rawCells()) {
+ if (KeyValue.Type.codeToType(cell.getTypeByte()) == KeyValue.Type.Put) {
+ if (put == null) {
+ put = new Put(CellUtil.cloneRow(cell));
+ }
+ put.add(cell);
+ } else {
+ if (del == null) {
+ del = new Delete(CellUtil.cloneRow(cell));
+ }
+ del.addDeleteMarker(cell);
+ }
+ }
+ return getMutationsWithSameTS(put, del);
+ }
+
+ /**
+ * indexRow is the set of all cells of all the row version of an index row from the index table. These are actual
+ * cells. We group these cells based on timestamp and type (put vs delete), and form the actual set of
+ * index mutations. indexKeyToMutationMap is a map from an index row key to a set of mutations that are generated
+ * using the rebuild process (i.e., by replaying raw data table mutations). These sets are sets of expected
+ * index mutations, one set for each index row key. Since not all mutations in the index table have both phase
+ * (i.e., pre and post data phase) mutations, we cannot compare actual index mutations with expected one by one
+ * and expect to find them identical. We need to consider concurrent data mutation effects, data table row write
+ * failures, post index write failures. Thus, we need to allow some expected and actual mutations to be skipped
+ * during comparing actual mutations to index mutations.
+ *
+ * The main idea for the verification algorithm used here is to match every expected verified put with an actual
+ * put such that these two mutations are the same except that actual mutation can be unverified.
+ *
+ * Some background on why we can skip some of the actual unverified puts and delete markers due to concurrent data
+ * table updates is as follows:
+ *
+ * For each data table mutation, two operations are done on the data table. One is to read the existing row state,
+ * and the second is to write to the data table. The processing of concurrent data mutations are serialized once
+ * for reading the existing row states, and then serialized again for updating data table. In other words,
+ * they go through locking twice, i.e., [lock, read, unlock] and [lock, write, unlock]. Because of this two phase
+ * locking, for a pair of concurrent mutations (for the same row), the same row state can be read from the data
+ * table. This means the same existing index row can be made unverified twice with different timestamps, one for
+ * each concurrent mutation. These unverified mutation are then repaired from the data table. Since expected
+ * mutations are used for rebuild (which is also used by the read repair), skipping these unverified put mutations
+ * that are not matching with expected mutation are safe as they will go through the same process during
+ * read repair and will be skipped and eventually cleaned up by the read repair. We can skip the delete markers
+ * safely too as they are placed to clean up these unverified mutations. When the data table rows are rebuilt,
+ * the rebuild process generates the delete family markers. The timestamp of delete markers are the timestamp of
+ * the data table mutation for which the delete marker is added. Thus, the timestamp of these delete markers will be
+ * higher than the timestamp of index row to be deleted.
+ */
+ private boolean verifySingleIndexRow(Result indexRow, VerificationResult.PhaseResult verificationPhaseResult)
+ throws IOException {
+ List<Mutation> expectedMutationList = indexKeyToMutationMap.get(indexRow.getRow());
+ if (expectedMutationList == null) {
+ throw new DoNotRetryIOException("No expected mutation");
+ }
+ List<Mutation> actualMutationList = prepareActualIndexMutations(indexRow);
+ if (actualMutationList == null || actualMutationList.isEmpty()) {
+ throw new DoNotRetryIOException("actualMutationList is null or empty");
+ }
+ Collections.sort(expectedMutationList, MUTATION_TS_DESC_COMPARATOR);
+ Collections.sort(actualMutationList, MUTATION_TS_DESC_COMPARATOR);
+ long currentTime = EnvironmentEdgeManager.currentTime();
+ int actualIndex = 0;
+ int expectedIndex = 0;
+ int matchingCount = 0;
+ int expectedSize = expectedMutationList.size();
+ int actualSize = actualMutationList.size();
+ Mutation expected = null;
+ Mutation previousExpected;
+ Mutation actual;
+ while (expectedIndex < expectedSize && actualIndex <actualSize) {
+ previousExpected = expected;
+ expected = expectedMutationList.get(expectedIndex);
+ // Check if cell expired as per the current server's time and data table ttl
+ // Index table should have the same ttl as the data table, hence we might not
+ // get a value back from index if it has already expired between our rebuild and
+ // verify
+ // TODO: have a metric to update for these cases
+ if (isTimestampBeforeTTL(currentTime, getTimestamp(expected))) {
+ verificationPhaseResult.expiredIndexRowCount++;
+ return true;
+ }
+ actual = actualMutationList.get(actualIndex);
+ if (expected instanceof Put) {
+ if (previousExpected == null || previousExpected instanceof Put) {
+ // This expected put is either the first mutation or a put just comes after another expected mutation
+ // on the expected mutation list which is sorted by the mutation timestamps. The cell timestamps
+ // within each mutation here are the same.
+ // Go down the list of actual mutations and find the corresponding actual put mutation with the same
+ // timestamp. Stop if a verified put or delete family mutation is encountered on the way. Skip
+ // unverified puts or delete family version delete markers.
+ while (getTimestamp(actual) > getTimestamp(expected) &&
+ ((actual instanceof Put && !isVerified((Put) actual)) ||
+ (actual instanceof Delete && isDeleteFamilyVersion(actual)))) {
+ actualIndex++;
+ if (actualIndex == actualSize) {
+ break;
+ }
+ actual = actualMutationList.get(actualIndex);
+ }
+ } else { // previousExpected instanceof Delete
+ // Between an expected delete and put, there cannot be any types of mutation even verified put
+ while (getTimestamp(actual) > getTimestamp(expected)) {
+ actualIndex++;
+ if (actualIndex == actualSize) {
+ break;
+ }
+ actual = actualMutationList.get(actualIndex);
+ }
+ }
+ if (actualIndex == actualSize) {
+ break;
+ }
+ // Now the expected and actual mutations should match
+ if (isMatchingMutation(expected, actual, expectedIndex)) {
+ expectedIndex++;
+ actualIndex++;
+ matchingCount++;
+ continue;
+ }
+ verificationPhaseResult.invalidIndexRowCount++;
+ return false;
+ } else { // expected instanceof Delete
+ // Between put and delete, delete and delete, or before first delete, there can be other deletes and
+ // unverified puts. Skip all of them if any
+ while (getTimestamp(actual) > getTimestamp(expected) &&
+ ((actual instanceof Put && !isVerified((Put) actual)) || actual instanceof Delete)) {
+ actualIndex++;
+ if (actualIndex == actualSize) {
+ break;
+ }
+ actual = actualMutationList.get(actualIndex);
+ }
+ if (actualIndex == actualSize) {
+ break;
+ }
+ // If this is first expected mutation is delete, there should be an actual delete mutation with the
+ // same timestamp or an unverified put with the same or older timestamp
+ if (getTimestamp(actual) == getTimestamp(expected) &&
+ (actual instanceof Delete && isDeleteFamily(actual))) {
+ expectedIndex++;
+ actualIndex++;
+ matchingCount++;
+ continue;
+ } else if (getTimestamp(actual) <= getTimestamp(expected) &&
+ (actual instanceof Put && !isVerified((Put) actual))) {
+ expectedIndex++;
+ if (previousExpected == null) {
+ matchingCount++;
+ }
+ continue;
+ }
+ if (previousExpected == null) {
+ String errorMsg = "First delete check failure";
+ byte[] dataKey = indexMaintainer.buildDataRowKey(new ImmutableBytesWritable(indexRow.getRow()), viewConstants);
+ logToIndexToolOutputTable(dataKey, indexRow.getRow(),
+ getTimestamp(expected),
+ getTimestamp(actual), errorMsg);
+ verificationPhaseResult.invalidIndexRowCount++;
return false;
}
- cellCount++;
}
}
- if (cellCount != indexRow.rawCells().length) {
- String errorMsg = "Expected to find " + cellCount + " cells but got "
- + indexRow.rawCells().length + " cells";
- logToIndexToolOutputTable(dataRow.getRow(), indexRow.getRow(), ts, getMaxTimestamp(indexRow), errorMsg);
- return false;
+ if ((expectedIndex != expectedSize) || actualIndex != actualSize) {
+ if (matchingCount > 0) {
+ // We do not consider this as a verification issue but log it for further information.
+ // This may happen due to compaction
+ byte[] dataKey = indexMaintainer.buildDataRowKey(new ImmutableBytesWritable(indexRow.getRow()), viewConstants);
+ String errorMsg = "Expected to find " + expectedMutationList.size() + " mutations but got "
+ + actualMutationList.size();
+ logToIndexToolOutputTable(dataKey, indexRow.getRow(),
+ getTimestamp(expectedMutationList.get(0)),
+ getTimestamp(actualMutationList.get(0)), errorMsg);
+ } else {
+ byte[] dataKey = indexMaintainer.buildDataRowKey(new ImmutableBytesWritable(indexRow.getRow()), viewConstants);
+ String errorMsg = "Not matching index row";
+ logToIndexToolOutputTable(dataKey, indexRow.getRow(),
+ getTimestamp(expectedMutationList.get(0)),
+ getTimestamp(actualMutationList.get(0)), errorMsg);
+ verificationPhaseResult.invalidIndexRowCount++;
+ return false;
+ }
}
+ verificationPhaseResult.validIndexRowCount++;
return true;
}
- private void verifyIndexRows(List<KeyRange> keys, Map<byte[], Put> perTaskDataKeyToDataPutMap,
+ private static long getMaxTimestamp(Pair<Put, Delete> pair) {
+ Put put = pair.getFirst();
+ long ts1 = 0;
+ if (put != null) {
+ ts1 = getMaxTimestamp((Mutation)put);
+ }
+ Delete del = pair.getSecond();
+ long ts2 = 0;
+ if (del != null) {
+ ts1 = getMaxTimestamp((Mutation)del);
+ }
+ return (ts1 > ts2) ? ts1 : ts2;
+ }
+
+ private void verifyIndexRows(List<KeyRange> keys,
VerificationResult.PhaseResult verificationPhaseResult) throws IOException {
- int expectedRowCount = keys.size();
+ List<KeyRange> invalidKeys = new ArrayList<>();
ScanRanges scanRanges = ScanRanges.createPointLookup(keys);
Scan indexScan = new Scan();
indexScan.setTimeRange(scan.getTimeRange().getMin(), scan.getTimeRange().getMax());
scanRanges.initializeScan(indexScan);
SkipScanFilter skipScanFilter = scanRanges.getSkipScanFilter();
- indexScan.setFilter(skipScanFilter);
- int rowCount = 0;
+ indexScan.setFilter(new SkipScanFilter(skipScanFilter, true));
+ indexScan.setRaw(true);
+ indexScan.setMaxVersions();
try (ResultScanner resultScanner = indexHTable.getScanner(indexScan)) {
for (Result result = resultScanner.next(); (result != null); result = resultScanner.next()) {
- Put dataPut = indexKeyToDataPutMap.get(result.getRow());
- if (dataPut == null) {
- // This should never happen
- String errorMsg = "Missing data row";
- logToIndexToolOutputTable(null, result.getRow(), 0, getMaxTimestamp(result), errorMsg);
- exceptionMessage = "Index verify failed - Missing data row - " + indexHTable.getName();
- throw new IOException(exceptionMessage);
+ KeyRange keyRange = PVarbinary.INSTANCE.getKeyRange(result.getRow());
+ if (!keys.contains(keyRange)) {
+ continue;
}
- if (verifySingleIndexRow(result, dataPut)) {
- verificationPhaseResult.validIndexRowCount++;
- perTaskDataKeyToDataPutMap.remove(dataPut.getRow());
- } else {
- verificationPhaseResult.invalidIndexRowCount++;
+ if (!verifySingleIndexRow(result, verificationPhaseResult)) {
+ invalidKeys.add(keyRange);
}
- rowCount++;
+ keys.remove(keyRange);
}
} catch (Throwable t) {
ServerUtil.throwIOException(indexHTable.getName().toString(), t);
}
// Check if any expected rows from index(which we didn't get) are already expired due to TTL
// TODO: metrics for expired rows
- if (!perTaskDataKeyToDataPutMap.isEmpty()) {
- Iterator<Entry<byte[], Put>> itr = perTaskDataKeyToDataPutMap.entrySet().iterator();
+ if (!keys.isEmpty()) {
+ Iterator<KeyRange> itr = keys.iterator();
long currentTime = EnvironmentEdgeManager.currentTime();
while(itr.hasNext()) {
- Entry<byte[], Put> entry = itr.next();
- long ts = getMaxTimestamp(entry.getValue());
- if (isTimestampBeforeTTL(currentTime, ts)) {
+ KeyRange keyRange = itr.next();
+ byte[] key = keyRange.getLowerRange();
+ List<Mutation> mutationList = indexKeyToMutationMap.get(key);
+ if (isTimestampBeforeTTL(currentTime, getTimestamp(mutationList.get(mutationList.size() - 1)))) {
itr.remove();
- rowCount++;
verificationPhaseResult.expiredIndexRowCount++;
}
}
}
- if (rowCount != expectedRowCount) {
- for (Map.Entry<byte[], Put> entry : perTaskDataKeyToDataPutMap.entrySet()) {
+ if (keys.size() > 0) {
+ for (KeyRange keyRange : keys) {
String errorMsg = "Missing index row";
- logToIndexToolOutputTable(entry.getKey(), null, getMaxTimestamp(entry.getValue()),
- 0, errorMsg);
+ byte[] key = keyRange.getLowerRange();
+ List<Mutation> mutationList = indexKeyToMutationMap.get(key);
+ byte[] dataKey = indexMaintainer.buildDataRowKey(new ImmutableBytesWritable(keyRange.getLowerRange()), viewConstants);
+ logToIndexToolOutputTable(dataKey,
+ keyRange.getLowerRange(),
+ getMaxTimestamp(dataKeyToMutationMap.get(dataKey)),
+ getTimestamp(mutationList.get(mutationList.size() - 1)), errorMsg);
}
- verificationPhaseResult.missingIndexRowCount += expectedRowCount - rowCount;
+ verificationPhaseResult.missingIndexRowCount += keys.size();
}
+ keys.addAll(invalidKeys);
}
private boolean isTimestampBeforeTTL(long currentTime, long tsToCheck) {
@@ -816,7 +1062,7 @@ public class IndexRebuildRegionScanner extends BaseRegionScanner {
return tsToCheck < (currentTime - (long) indexTableTTL * 1000);
}
- private void addVerifyTask(final List<KeyRange> keys, final Map<byte[], Put> perTaskDataKeyToDataPutMap,
+ private void addVerifyTask(final List<KeyRange> keys,
final VerificationResult.PhaseResult verificationPhaseResult) {
tasks.add(new Task<Boolean>() {
@Override
@@ -826,7 +1072,7 @@ public class IndexRebuildRegionScanner extends BaseRegionScanner {
exceptionMessage = "Pool closed, not attempting to verify index rows! " + indexHTable.getName();
throw new IOException(exceptionMessage);
}
- verifyIndexRows(keys, perTaskDataKeyToDataPutMap, verificationPhaseResult);
+ verifyIndexRows(keys, verificationPhaseResult);
} catch (Exception e) {
throw e;
}
@@ -836,32 +1082,26 @@ public class IndexRebuildRegionScanner extends BaseRegionScanner {
}
private void parallelizeIndexVerify(VerificationResult.PhaseResult verificationPhaseResult) throws IOException {
- for (Mutation mutation : mutations) {
- indexKeyToDataPutMap.put(getIndexRowKey((Put)mutation), (Put)mutation);
- }
- int taskCount = (indexKeyToDataPutMap.size() + rowCountPerTask - 1) / rowCountPerTask;
+ int taskCount = (indexKeyToMutationMap.size() + rowCountPerTask - 1) / rowCountPerTask;
tasks = new TaskBatch<>(taskCount);
- List<Map<byte[], Put>> dataPutMapList = new ArrayList<>(taskCount);
+ List<List<KeyRange>> listOfKeyRangeList = new ArrayList<>(taskCount);
List<VerificationResult.PhaseResult> verificationPhaseResultList = new ArrayList<>(taskCount);
List<KeyRange> keys = new ArrayList<>(rowCountPerTask);
- Map<byte[], Put> perTaskDataKeyToDataPutMap = Maps.newTreeMap(Bytes.BYTES_COMPARATOR);
- dataPutMapList.add(perTaskDataKeyToDataPutMap);
+ listOfKeyRangeList.add(keys);
VerificationResult.PhaseResult perTaskVerificationPhaseResult = new VerificationResult.PhaseResult();
verificationPhaseResultList.add(perTaskVerificationPhaseResult);
- for (Map.Entry<byte[], Put> entry: indexKeyToDataPutMap.entrySet()) {
- keys.add(PVarbinary.INSTANCE.getKeyRange(entry.getKey()));
- perTaskDataKeyToDataPutMap.put(entry.getValue().getRow(), entry.getValue());
+ for (byte[] indexKey: indexKeyToMutationMap.keySet()) {
+ keys.add(PVarbinary.INSTANCE.getKeyRange(indexKey));
if (keys.size() == rowCountPerTask) {
- addVerifyTask(keys, perTaskDataKeyToDataPutMap, perTaskVerificationPhaseResult);
+ addVerifyTask(keys, perTaskVerificationPhaseResult);
keys = new ArrayList<>(rowCountPerTask);
- perTaskDataKeyToDataPutMap = Maps.newTreeMap(Bytes.BYTES_COMPARATOR);
- dataPutMapList.add(perTaskDataKeyToDataPutMap);
+ listOfKeyRangeList.add(keys);
perTaskVerificationPhaseResult = new VerificationResult.PhaseResult();
verificationPhaseResultList.add(perTaskVerificationPhaseResult);
}
}
if (keys.size() > 0) {
- addVerifyTask(keys, perTaskDataKeyToDataPutMap, perTaskVerificationPhaseResult);
+ addVerifyTask(keys, perTaskVerificationPhaseResult);
}
List<Boolean> taskResultList = null;
try {
@@ -878,30 +1118,41 @@ public class IndexRebuildRegionScanner extends BaseRegionScanner {
throw new IOException(exceptionMessage);
}
}
- if (verifyType == IndexTool.IndexVerifyType.BEFORE || verifyType == IndexTool.IndexVerifyType.BOTH) {
- for (Map<byte[], Put> dataPutMap : dataPutMapList) {
- dataKeyToDataPutMap.putAll(dataPutMap);
- }
- }
for (VerificationResult.PhaseResult result : verificationPhaseResultList) {
verificationPhaseResult.add(result);
}
+ if (verifyType == IndexTool.IndexVerifyType.BEFORE || verifyType == IndexTool.IndexVerifyType.BOTH) {
+ Map<byte[], Pair<Put, Delete>> newDataKeyToMutationMap = Maps.newTreeMap(Bytes.BYTES_COMPARATOR);
+ for (List<KeyRange> keyRangeList : listOfKeyRangeList) {
+ for (KeyRange keyRange : keyRangeList) {
+ byte[] dataKey = indexMaintainer.buildDataRowKey(new ImmutableBytesWritable(keyRange.getLowerRange()), viewConstants);
+ newDataKeyToMutationMap.put(dataKey, dataKeyToMutationMap.get(dataKey));
+ }
+ }
+ dataKeyToMutationMap.clear();
+ dataKeyToMutationMap = newDataKeyToMutationMap;
+ }
}
private void rebuildIndexRows(UngroupedAggregateRegionObserver.MutationList mutationList) throws IOException {
byte[] uuidValue = ServerCacheClient.generateId();
UngroupedAggregateRegionObserver.MutationList currentMutationList =
new UngroupedAggregateRegionObserver.MutationList(maxBatchSize);
+ Put put = null;
for (Mutation mutation : mutationList) {
- Put put = (Put) mutation;
- currentMutationList.add(mutation);
- setMutationAttributes(put, uuidValue);
- uuidValue = commitIfReady(uuidValue, currentMutationList);
- Delete deleteMarkers = generateDeleteMarkers(put);
- if (deleteMarkers != null) {
- setMutationAttributes(deleteMarkers, uuidValue);
- currentMutationList.add(deleteMarkers);
+ if (mutation instanceof Put) {
+ if (put != null) {
+ // back to back put, i.e., no delete in between. we can commit the previous put
+ uuidValue = commitIfReady(uuidValue, currentMutationList);
+ }
+ currentMutationList.add(mutation);
+ setMutationAttributes(mutation, uuidValue);
+ put = (Put)mutation;
+ } else {
+ currentMutationList.add(mutation);
+ setMutationAttributes(mutation, uuidValue);
uuidValue = commitIfReady(uuidValue, currentMutationList);
+ put = null;
}
}
if (!currentMutationList.isEmpty()) {
@@ -912,11 +1163,11 @@ public class IndexRebuildRegionScanner extends BaseRegionScanner {
private void verifyAndOrRebuildIndex() throws IOException {
VerificationResult nextVerificationResult = new VerificationResult();
- nextVerificationResult.scannedDataRowCount = mutations.size();
+ nextVerificationResult.scannedDataRowCount = dataKeyToMutationMap.size();
if (verifyType == IndexTool.IndexVerifyType.AFTER || verifyType == IndexTool.IndexVerifyType.NONE) {
// For these options we start with rebuilding index rows
rebuildIndexRows(mutations);
- nextVerificationResult.rebuiltIndexRowCount = mutations.size();
+ nextVerificationResult.rebuiltIndexRowCount = dataKeyToMutationMap.size();
isBeforeRebuilt = false;
}
if (verifyType == IndexTool.IndexVerifyType.NONE) {
@@ -928,73 +1179,285 @@ public class IndexRebuildRegionScanner extends BaseRegionScanner {
// For these options we start with verifying index rows
parallelizeIndexVerify(verificationPhaseResult);
nextVerificationResult.before.add(verificationPhaseResult);
- if (mutations.size() != verificationPhaseResult.getTotalCount()) {
- throw new DoNotRetryIOException(
- "mutations.size() != verificationPhaseResult.getTotalCount() at the before phase " +
- nextVerificationResult + " dataKeyToDataPutMap.size() = " + dataKeyToDataPutMap.size());
- }
}
if (verifyType == IndexTool.IndexVerifyType.BEFORE || verifyType == IndexTool.IndexVerifyType.BOTH) {
// For these options, we have identified the rows to be rebuilt and now need to rebuild them
// At this point, dataKeyToDataPutMap includes mapping only for the rows to be rebuilt
mutations.clear();
- for (Map.Entry<byte[], Put> entry: dataKeyToDataPutMap.entrySet()) {
- mutations.add(entry.getValue());
+
+ for (Map.Entry<byte[], Pair<Put, Delete>> entry: dataKeyToMutationMap.entrySet()) {
+ if (entry.getValue().getFirst() != null) {
+ mutations.add(entry.getValue().getFirst());
+ }
+ if (entry.getValue().getSecond() != null) {
+ mutations.add(entry.getValue().getSecond());
+ }
}
rebuildIndexRows(mutations);
- nextVerificationResult.rebuiltIndexRowCount += mutations.size();
+ nextVerificationResult.rebuiltIndexRowCount += dataKeyToMutationMap.size();
isBeforeRebuilt = false;
}
if (verifyType == IndexTool.IndexVerifyType.AFTER || verifyType == IndexTool.IndexVerifyType.BOTH) {
// We have rebuilt index row and now we need to verify them
- indexKeyToDataPutMap.clear();
VerificationResult.PhaseResult verificationPhaseResult = new VerificationResult.PhaseResult();
+ indexKeyToMutationMap.clear();
+ for (Map.Entry<byte[], Pair<Put, Delete>> entry: dataKeyToMutationMap.entrySet()) {
+ prepareIndexMutations(entry.getValue().getFirst(), entry.getValue().getSecond());
+ }
parallelizeIndexVerify(verificationPhaseResult);
nextVerificationResult.after.add(verificationPhaseResult);
- if (mutations.size() != verificationPhaseResult.getTotalCount()) {
- throw new DoNotRetryIOException(
- "mutations.size() != verificationPhaseResult.getTotalCount() at the after phase " +
- nextVerificationResult + " dataKeyToDataPutMap.size() = " + dataKeyToDataPutMap.size());
- }
}
- indexKeyToDataPutMap.clear();
verificationResult.add(nextVerificationResult);
}
+ private boolean isColumnIncluded(Cell cell) {
+ byte[] family = CellUtil.cloneFamily(cell);
+ if (!familyMap.containsKey(family)) {
+ return false;
+ }
+ NavigableSet<byte[]> set = familyMap.get(family);
+ if (set == null || set.isEmpty()) {
+ return true;
+ }
+ byte[] qualifier = CellUtil.cloneQualifier(cell);
+ return set.contains(qualifier);
+ }
+
+ public static long getTimestamp(Mutation m) {
+ for (List<Cell> cells : m.getFamilyCellMap().values()) {
+ for (Cell cell : cells) {
+ return cell.getTimestamp();
+ }
+ }
+ throw new IllegalStateException("No cell found");
+ }
+
+ /**
+ * This is to reorder the mutations in ascending order by the tuple of timestamp and mutation type where
+ * delete comes before put
+ */
+ public static final Comparator<Mutation> MUTATION_TS_COMPARATOR = new Comparator<Mutation>() {
+ @Override
+ public int compare(Mutation o1, Mutation o2) {
+ long ts1 = getTimestamp(o1);
+ long ts2 = getTimestamp(o2);
+ if (ts1 < ts2) {
+ return -1;
+ }
+ if (ts1 > ts2) {
+ return 1;
+ }
+ if (o1 instanceof Delete && o2 instanceof Put) {
+ return -1;
+ }
+ if (o1 instanceof Put && o2 instanceof Delete) {
+ return 1;
+ }
+ return 0;
+ }
+ };
+
+ public static List<Mutation> getMutationsWithSameTS(Put put, Delete del) {
+ List<Mutation> mutationList = Lists.newArrayListWithExpectedSize(2);
+ if (put != null) {
+ mutationList.add(put);
+ }
+ if (del != null) {
+ mutationList.add(del);
+ }
+ // Group the cells within a mutation based on their timestamps and create a separate mutation for each group
+ mutationList = (List<Mutation>) IndexManagementUtil.flattenMutationsByTimestamp(mutationList);
+ // Reorder the mutations on the same row so that delete comes before put when they have the same timestamp
+ Collections.sort(mutationList, MUTATION_TS_COMPARATOR);
+ return mutationList;
+ }
+
+ private static Put prepareIndexPutForRebuid(IndexMaintainer indexMaintainer, ImmutableBytesPtr rowKeyPtr,
+ ValueGetter mergedRowVG, long ts)
+ throws IOException {
+ Put indexPut = indexMaintainer.buildUpdateMutation(GenericKeyValueBuilder.INSTANCE,
+ mergedRowVG, rowKeyPtr, ts, null, null);
+ if (indexPut == null) {
+ // No covered column. Just prepare an index row with the empty column
+ byte[] indexRowKey = indexMaintainer.buildRowKey(mergedRowVG, rowKeyPtr,
+ null, null, HConstants.LATEST_TIMESTAMP);
+ indexPut = new Put(indexRowKey);
+ }
+ indexPut.addColumn(indexMaintainer.getEmptyKeyValueFamily().copyBytesIfNecessary(),
+ indexMaintainer.getEmptyKeyValueQualifier(), ts, VERIFIED_BYTES);
+ return indexPut;
+ }
+
+ public static void removeColumn(Put put, Cell deleteCell) {
+ byte[] family = CellUtil.cloneFamily(deleteCell);
+ List<Cell> cellList = put.getFamilyCellMap().get(family);
+ if (cellList == null) {
+ return;
+ }
+ Iterator<Cell> cellIterator = cellList.iterator();
+ while (cellIterator.hasNext()) {
+ Cell cell = cellIterator.next();
+ if (CellUtil.matchingQualifier(cell, deleteCell)) {
+ cellIterator.remove();
+ if (cellList.isEmpty()) {
+ put.getFamilyCellMap().remove(family);
+ }
+ return;
+ }
+ }
+ }
+
+ public static void apply(Put destination, Put source) throws IOException {
+ for (List<Cell> cells : source.getFamilyCellMap().values()) {
+ for (Cell cell : cells) {
+ if (!destination.has(CellUtil.cloneFamily(cell), CellUtil.cloneQualifier(cell))) {
+ destination.add(cell);
+ }
+ }
+ }
+ }
+
+ public static Put applyNew(Put destination, Put source) throws IOException {
+ Put next = new Put(destination);
+ apply(next, source);
+ return next;
+ }
+
+ /**
+ * Generate the index update for a data row from the mutation that are obtained by merging the previous data row
+ * state with the pending row mutation for index rebuild. This method is called only for global indexes.
+ * pendingMutations is a sorted list of data table mutations that are used to replay index table mutations.
+ * This list is sorted in ascending order by the tuple of row key, timestamp and mutation type where delete comes
+ * after put.
+ */
+ public static List<Mutation> prepareIndexMutationsForRebuild(IndexMaintainer indexMaintainer,
+ Put dataPut, Delete dataDel) throws IOException {
+ List<Mutation> dataMutations = getMutationsWithSameTS(dataPut, dataDel);
+ List<Mutation> indexMutations = Lists.newArrayListWithExpectedSize(dataMutations.size());
+ // The row key ptr of the data table row for which we will build index rows here
+ ImmutableBytesPtr rowKeyPtr = (dataPut != null) ? new ImmutableBytesPtr(dataPut.getRow()) :
+ new ImmutableBytesPtr(dataDel.getRow());
+ // Start with empty data table row
+ Put currentDataRow = null;
+ // The index row key corresponding to the current data row
+ byte[] indexRowKeyForCurrentDataRow = null;
+ for (Mutation mutation : dataMutations) {
+ long ts = getTimestamp(mutation);
+ if (mutation instanceof Put) {
+ // Add this put on top of the current data row state to get the next data row state
+ Put nextDataRow = (currentDataRow == null) ? new Put((Put)mutation) : applyNew((Put)mutation, currentDataRow);
+ ValueGetter nextDataRowVG = new IndexRebuildRegionScanner.SimpleValueGetter(nextDataRow);
+ Put indexPut = prepareIndexPutForRebuid(indexMaintainer, rowKeyPtr, nextDataRowVG, ts);
+ indexMutations.add(indexPut);
+ // Delete the current index row if the new index key is different than the current one
+ if (indexRowKeyForCurrentDataRow != null) {
+ if (Bytes.compareTo(indexPut.getRow(), indexRowKeyForCurrentDataRow) != 0) {
+ Mutation del = indexMaintainer.buildRowDeleteMutation(indexRowKeyForCurrentDataRow,
+ IndexMaintainer.DeleteType.ALL_VERSIONS, ts);
+ indexMutations.add(del);
+ }
+ }
+ // For the next iteration
+ currentDataRow = nextDataRow;
+ indexRowKeyForCurrentDataRow = indexPut.getRow();
+ } else if (currentDataRow != null) {
+ // We apply delete column mutations only on the current data row state. For the index table,
+ // we are only interested in if the current data row is deleted or not. There is no need to apply
+ // column deletes to index rows since index rows are always full rows and all the cells in an index
+ // row have the same timestamp value. Because of this index rows versions do not share cells.
+ for (List<Cell> cells : mutation.getFamilyCellMap().values()) {
+ for (Cell cell : cells) {
+ switch ((KeyValue.Type.codeToType(cell.getTypeByte()))) {
+ case DeleteFamily:
+ case DeleteFamilyVersion:
+ currentDataRow.getFamilyCellMap().remove(CellUtil.cloneFamily(cell));
+ break;
+ case DeleteColumn:
+ case Delete:
+ removeColumn(currentDataRow, cell);
+ }
+ }
+ }
+ if (currentDataRow.getFamilyCellMap().size() == 0) {
+ Mutation del = indexMaintainer.buildRowDeleteMutation(indexRowKeyForCurrentDataRow,
+ IndexMaintainer.DeleteType.ALL_VERSIONS, ts);
+ indexMutations.add(del);
+ currentDataRow = null;
+ indexRowKeyForCurrentDataRow = null;
+ }
+ }
+ }
+ return indexMutations;
+ }
+
+ private void prepareIndexMutations(Put put, Delete del) throws IOException{
+ List<Mutation> indexMutations = prepareIndexMutationsForRebuild(indexMaintainer, put, del);
+ for (Mutation mutation : indexMutations) {
+ byte[] indexRowKey = mutation.getRow();
+ List<Mutation> mutationList = indexKeyToMutationMap.get(indexRowKey);
+ if (mutationList == null) {
+ mutationList = new ArrayList<>();
+ mutationList.add(mutation);
+ indexKeyToMutationMap.put(indexRowKey, mutationList);
+ } else {
+ mutationList.add(mutation);
+ }
+ }
+ }
+
@Override
public boolean next(List<Cell> results) throws IOException {
+ if (indexRowKey != null &&
+ singleRowRebuildReturnCode == GlobalIndexChecker.RebuildReturnCode.NO_DATA_ROW.getValue()) {
+ byte[] rowCountBytes =
+ PLong.INSTANCE.toBytes(Long.valueOf(singleRowRebuildReturnCode));
+ final Cell aggKeyValue = KeyValueUtil.newKeyValue(UNGROUPED_AGG_ROW_KEY, SINGLE_COLUMN_FAMILY,
+ SINGLE_COLUMN, AGG_TIMESTAMP, rowCountBytes, 0, rowCountBytes.length);
+ results.add(aggKeyValue);
+ return false;
+ }
Cell lastCell = null;
int rowCount = 0;
region.startRegionOperation();
try {
- // Partial rebuilds by MetadataRegionObserver use raw scan. Inline verification is not supported for them
- boolean partialRebuild = scan.isRaw();
byte[] uuidValue = ServerCacheClient.generateId();
synchronized (innerScanner) {
do {
List<Cell> row = new ArrayList<Cell>();
hasMore = innerScanner.nextRaw(row);
if (!row.isEmpty()) {
- lastCell = row.get(0);
+ lastCell = row.get(0); // lastCell is any cell from the last visited row
Put put = null;
Delete del = null;
for (Cell cell : row) {
if (KeyValue.Type.codeToType(cell.getTypeByte()) == KeyValue.Type.Put) {
+ if (!partialRebuild && familyMap != null && !isColumnIncluded(cell)) {
+ continue;
+ }
if (put == null) {
put = new Put(CellUtil.cloneRow(cell));
- mutations.add(put);
}
put.add(cell);
} else {
if (del == null) {
del = new Delete(CellUtil.cloneRow(cell));
- mutations.add(del);
}
del.addDeleteMarker(cell);
}
}
- if (partialRebuild) {
+ if (put == null && del == null) {
+ continue;
+ }
+ // Always add the put first and then delete for a given row. This simplifies the logic in
+ // IndexRegionObserver
+ if (put != null) {
+ mutations.add(put);
+ }
+ if (del != null) {
+ mutations.add(del);
+ }
+ if (!verify) {
if (put != null) {
setMutationAttributes(put, uuidValue);
}
@@ -1002,35 +1465,18 @@ public class IndexRebuildRegionScanner extends BaseRegionScanner {
setMutationAttributes(del, uuidValue);
}
uuidValue = commitIfReady(uuidValue, mutations);
- }
- if (indexRowKey != null) {
- if (put != null) {
- setMutationAttributes(put, uuidValue);
- }
- Delete deleteMarkers = generateDeleteMarkers(put);
- if (deleteMarkers != null) {
- setMutationAttributes(deleteMarkers, uuidValue);
- mutations.add(deleteMarkers);
- uuidValue = commitIfReady(uuidValue, mutations);
- }
- // GlobalIndexChecker passed the index row key. This is to build a single index row.
- // Check if the data table row we have just scanned matches with the index row key.
- // If not, there is no need to build the index row from this data table row,
- // and just return zero row count.
- if (checkIndexRow(indexRowKey, put)) {
- rowCount = GlobalIndexChecker.RebuildReturnCode.INDEX_ROW_EXISTS.getValue();
- } else {
- rowCount = GlobalIndexChecker.RebuildReturnCode.NO_INDEX_ROW.getValue();
- }
- break;
+ } else {
+ byte[] dataKey = (put != null) ? put.getRow() : del.getRow();
+ prepareIndexMutations(put, del);
+ dataKeyToMutationMap.put(dataKey, new Pair<Put, Delete>(put, del));
}
rowCount++;
}
} while (hasMore && rowCount < pageSizeInRows);
- if (!partialRebuild && indexRowKey == null) {
- verifyAndOrRebuildIndex();
- } else {
- if (!mutations.isEmpty()) {
+ if (!mutations.isEmpty()) {
+ if (verify) {
+ verifyAndOrRebuildIndex();
+ } else {
ungroupedAggregateRegionObserver.checkForRegionClosing();
ungroupedAggregateRegionObserver.commitBatchWithRetries(region, mutations, blockingMemstoreSize);
}
@@ -1043,10 +1489,13 @@ public class IndexRebuildRegionScanner extends BaseRegionScanner {
region.closeRegionOperation();
mutations.clear();
if (verify) {
- indexKeyToDataPutMap.clear();
- dataKeyToDataPutMap.clear();
+ dataKeyToMutationMap.clear();
+ indexKeyToMutationMap.clear();
}
}
+ if (indexRowKey != null) {
+ rowCount = singleRowRebuildReturnCode;
+ }
byte[] rowCountBytes = PLong.INSTANCE.toBytes(Long.valueOf(rowCount));
final Cell aggKeyValue;
if (lastCell == null) {
diff --git a/phoenix-core/src/main/java/org/apache/phoenix/coprocessor/UngroupedAggregateRegionObserver.java b/phoenix-core/src/main/java/org/apache/phoenix/coprocessor/UngroupedAggregateRegionObserver.java
index 4f21511..fa42ce9 100644
--- a/phoenix-core/src/main/java/org/apache/phoenix/coprocessor/UngroupedAggregateRegionObserver.java
+++ b/phoenix-core/src/main/java/org/apache/phoenix/coprocessor/UngroupedAggregateRegionObserver.java
@@ -1060,9 +1060,20 @@ public class UngroupedAggregateRegionObserver extends BaseScannerRegionObserver
private RegionScanner rebuildIndices(final RegionScanner innerScanner, final Region region, final Scan scan,
final RegionCoprocessorEnvironment env) throws IOException {
-
- RegionScanner scanner = new IndexRebuildRegionScanner(innerScanner, region, scan, env, this);
- return scanner;
+ if (!scan.isRaw() && scan.getAttribute(BaseScannerRegionObserver.INDEX_ROW_KEY) == null) {
+ Scan rawScan = new Scan(scan);
+ rawScan.setRaw(true);
+ rawScan.setMaxVersions();
+ rawScan.getFamilyMap().clear();
+ rawScan.setFilter(null);
+ for (byte[] family : scan.getFamilyMap().keySet()) {
+ rawScan.addFamily(family);
+ }
+ innerScanner.close();
+ RegionScanner scanner = region.getScanner(rawScan);
+ return new IndexRebuildRegionScanner(scanner, region, scan, env, this);
+ }
+ return new IndexRebuildRegionScanner(innerScanner, region, scan, env, this);
}
private RegionScanner collectStats(final RegionScanner innerScanner, StatisticsCollector stats,
diff --git a/phoenix-core/src/main/java/org/apache/phoenix/hbase/index/IndexRegionObserver.java b/phoenix-core/src/main/java/org/apache/phoenix/hbase/index/IndexRegionObserver.java
index 08a120e..0092c5c 100644
--- a/phoenix-core/src/main/java/org/apache/phoenix/hbase/index/IndexRegionObserver.java
+++ b/phoenix-core/src/main/java/org/apache/phoenix/hbase/index/IndexRegionObserver.java
@@ -31,6 +31,7 @@ import java.util.concurrent.ConcurrentHashMap;
import com.google.common.collect.ArrayListMultimap;
import com.google.common.collect.ListMultimap;
+import com.google.common.collect.Lists;
import org.apache.commons.logging.Log;
import org.apache.commons.logging.LogFactory;
import org.apache.hadoop.conf.Configuration;
@@ -41,25 +42,31 @@ import org.apache.hadoop.hbase.DoNotRetryIOException;
import org.apache.hadoop.hbase.HConstants;
import org.apache.hadoop.hbase.HConstants.OperationStatusCode;
import org.apache.hadoop.hbase.HTableDescriptor;
+import org.apache.hadoop.hbase.KeyValue;
import org.apache.hadoop.hbase.client.Delete;
import org.apache.hadoop.hbase.client.Durability;
import org.apache.hadoop.hbase.client.Increment;
import org.apache.hadoop.hbase.client.Mutation;
import org.apache.hadoop.hbase.client.Put;
import org.apache.hadoop.hbase.client.Result;
+import org.apache.hadoop.hbase.client.Scan;
import org.apache.hadoop.hbase.coprocessor.BaseRegionObserver;
import org.apache.hadoop.hbase.coprocessor.ObserverContext;
import org.apache.hadoop.hbase.coprocessor.RegionCoprocessorEnvironment;
import org.apache.hadoop.hbase.regionserver.MiniBatchOperationInProgress;
import org.apache.hadoop.hbase.regionserver.OperationStatus;
import org.apache.hadoop.hbase.regionserver.Region;
+import org.apache.hadoop.hbase.regionserver.RegionScanner;
import org.apache.hadoop.hbase.util.Bytes;
import org.apache.hadoop.hbase.util.Pair;
import org.apache.htrace.Span;
import org.apache.htrace.Trace;
import org.apache.htrace.TraceScope;
+import org.apache.phoenix.compile.ScanRanges;
import org.apache.phoenix.coprocessor.BaseScannerRegionObserver.ReplayWrite;
import org.apache.phoenix.coprocessor.DelegateRegionCoprocessorEnvironment;
+import org.apache.phoenix.coprocessor.IndexRebuildRegionScanner;
+import org.apache.phoenix.filter.SkipScanFilter;
import org.apache.phoenix.hbase.index.LockManager.RowLock;
import org.apache.phoenix.hbase.index.builder.IndexBuildManager;
import org.apache.phoenix.hbase.index.builder.IndexBuilder;
@@ -67,20 +74,27 @@ import org.apache.phoenix.hbase.index.covered.IndexMetaData;
import org.apache.phoenix.hbase.index.metrics.MetricsIndexerSource;
import org.apache.phoenix.hbase.index.metrics.MetricsIndexerSourceFactory;
import org.apache.phoenix.hbase.index.table.HTableInterfaceReference;
+import org.apache.phoenix.hbase.index.util.GenericKeyValueBuilder;
import org.apache.phoenix.hbase.index.util.ImmutableBytesPtr;
import org.apache.phoenix.hbase.index.util.IndexManagementUtil;
import org.apache.phoenix.hbase.index.write.IndexWriter;
import org.apache.phoenix.hbase.index.write.LazyParallelWriterIndexCommitter;
import org.apache.phoenix.index.IndexMaintainer;
import org.apache.phoenix.index.PhoenixIndexMetaData;
+import org.apache.phoenix.query.KeyRange;
import org.apache.phoenix.query.QueryServicesOptions;
+import org.apache.phoenix.schema.types.PVarbinary;
import org.apache.phoenix.trace.TracingUtils;
import org.apache.phoenix.trace.util.NullSpan;
import org.apache.phoenix.util.EnvironmentEdgeManager;
import org.apache.phoenix.util.ServerUtil;
import org.apache.phoenix.util.ServerUtil.ConnectionType;
-import com.google.common.collect.Lists;
+import java.util.Set;
+
+import static org.apache.phoenix.coprocessor.IndexRebuildRegionScanner.applyNew;
+import static org.apache.phoenix.coprocessor.IndexRebuildRegionScanner.prepareIndexMutationsForRebuild;
+import static org.apache.phoenix.coprocessor.IndexRebuildRegionScanner.removeColumn;
/**
* Do all the work of managing index updates from a single coprocessor. All Puts/Delets are passed
@@ -155,14 +169,16 @@ public class IndexRegionObserver extends BaseRegionObserver {
private List<RowLock> rowLocks = Lists.newArrayListWithExpectedSize(QueryServicesOptions.DEFAULT_MUTATE_BATCH_SIZE);
private HashSet<ImmutableBytesPtr> rowsToLock = new HashSet<>();
private boolean rebuild;
+ // The current and next states of the data rows corresponding to the pending mutations
+ private HashMap<ImmutableBytesPtr, Pair<Put, Put>> dataRowStates;
private BatchMutateContext(int clientVersion) {
this.clientVersion = clientVersion;
}
}
-
+
private ThreadLocal<BatchMutateContext> batchMutateContext =
new ThreadLocal<BatchMutateContext>();
-
+
/** Configuration key for the {@link IndexBuilder} to use */
public static final String INDEX_BUILDER_CONF_KEY = "index.builder";
@@ -172,17 +188,11 @@ public class IndexRegionObserver extends BaseRegionObserver {
*/
public static final String CHECK_VERSION_CONF_KEY = "com.saleforce.hbase.index.checkversion";
- private static final String INDEX_RECOVERY_FAILURE_POLICY_KEY = "org.apache.hadoop.hbase.index.recovery.failurepolicy";
-
public static final String INDEX_LAZY_POST_BATCH_WRITE = "org.apache.hadoop.hbase.index.lazy.post_batch.write";
private static final boolean INDEX_LAZY_POST_BATCH_WRITE_DEFAULT = false;
private static final String INDEXER_INDEX_WRITE_SLOW_THRESHOLD_KEY = "phoenix.indexer.slow.post.batch.mutate.threshold";
private static final long INDEXER_INDEX_WRITE_SLOW_THRESHOLD_DEFAULT = 3_000;
- private static final String INDEXER_INDEX_PREPARE_SLOW_THRESHOLD_KEY = "phoenix.indexer.slow.pre.batch.mutate.threshold";
- private static final long INDEXER_INDEX_PREPARE_SLOW_THREHSOLD_DEFAULT = 3_000;
- private static final String INDEXER_POST_OPEN_SLOW_THRESHOLD_KEY = "phoenix.indexer.slow.open.threshold";
- private static final long INDEXER_POST_OPEN_SLOW_THRESHOLD_DEFAULT = 3_000;
private static final String INDEXER_PRE_INCREMENT_SLOW_THRESHOLD_KEY = "phoenix.indexer.slow.pre.increment";
private static final long INDEXER_PRE_INCREMENT_SLOW_THRESHOLD_DEFAULT = 3_000;
@@ -343,13 +353,9 @@ public class IndexRegionObserver extends BaseRegionObserver {
public static long getMaxTimestamp(Mutation m) {
long maxTs = 0;
- long ts = 0;
- Iterator iterator = m.getFamilyCellMap().entrySet().iterator();
- while (iterator.hasNext()) {
- Map.Entry<byte[], List<Cell>> entry = (Map.Entry) iterator.next();
- Iterator<Cell> cellIterator = entry.getValue().iterator();
- while (cellIterator.hasNext()) {
- Cell cell = cellIterator.next();
+ long ts;
+ for (List<Cell> cells : m.getFamilyCellMap().values()) {
+ for (Cell cell : cells) {
ts = cell.getTimestamp();
if (ts > maxTs) {
maxTs = ts;
@@ -403,316 +409,541 @@ public class IndexRegionObserver extends BaseRegionObserver {
}
}
- private Collection<? extends Mutation> groupMutations(MiniBatchOperationInProgress<Mutation> miniBatchOp,
- long now, ReplayWrite replayWrite) throws IOException {
- Map<ImmutableBytesPtr, MultiMutation> mutationsMap = new HashMap<>();
- boolean copyMutations = false;
- for (int i = 0; i < miniBatchOp.size(); i++) {
- if (miniBatchOp.getOperationStatus(i) == IGNORE) {
- continue;
- }
- Mutation m = miniBatchOp.getOperation(i);
- if (this.builder.isEnabled(m)) {
- // Track whether or not we need to
- ImmutableBytesPtr row = new ImmutableBytesPtr(m.getRow());
- if (mutationsMap.containsKey(row)) {
- copyMutations = true;
- } else {
- mutationsMap.put(row, null);
- }
- }
- }
- // early exit if it turns out we don't have any edits
- if (mutationsMap.isEmpty()) {
- return null;
- }
- // If we're copying the mutations
- Collection<Mutation> originalMutations;
- Collection<? extends Mutation> mutations;
- if (copyMutations) {
- originalMutations = null;
- mutations = mutationsMap.values();
- } else {
- originalMutations = Lists.newArrayListWithExpectedSize(mutationsMap.size());
- mutations = originalMutations;
- }
-
- boolean resetTimeStamp = replayWrite == null;
+ /**
+ * This method is only used for local indexes
+ */
+ private Collection<? extends Mutation> groupMutations(MiniBatchOperationInProgress<Mutation> miniBatchOp,
+ long now) throws IOException {
+ Map<ImmutableBytesPtr, MultiMutation> mutationsMap = new HashMap<>();
+ boolean copyMutations = false;
+ for (int i = 0; i < miniBatchOp.size(); i++) {
+ if (miniBatchOp.getOperationStatus(i) == IGNORE) {
+ continue;
+ }
+ Mutation m = miniBatchOp.getOperation(i);
+ if (this.builder.isEnabled(m)) {
+ // Track whether or not we need to
+ ImmutableBytesPtr row = new ImmutableBytesPtr(m.getRow());
+ if (mutationsMap.containsKey(row)) {
+ copyMutations = true;
+ } else {
+ mutationsMap.put(row, null);
+ }
+ }
+ }
+ // early exit if it turns out we don't have any edits
+ if (mutationsMap.isEmpty()) {
+ return null;
+ }
+ // If we're copying the mutations
+ Collection<Mutation> originalMutations;
+ Collection<? extends Mutation> mutations;
+ if (copyMutations) {
+ originalMutations = null;
+ mutations = mutationsMap.values();
+ } else {
+ originalMutations = Lists.newArrayListWithExpectedSize(mutationsMap.size());
+ mutations = originalMutations;
+ }
+ for (int i = 0; i < miniBatchOp.size(); i++) {
+ Mutation m = miniBatchOp.getOperation(i);
+ // skip this mutation if we aren't enabling indexing
+ // unfortunately, we really should ask if the raw mutation (rather than the combined mutation)
+ // should be indexed, which means we need to expose another method on the builder. Such is the
+ // way optimization go though.
+ if (miniBatchOp.getOperationStatus(i) != IGNORE && this.builder.isEnabled(m)) {
+ // Unless we're replaying edits to rebuild the index, we update the time stamp
+ // of the data table to prevent overlapping time stamps (which prevents index
+ // inconsistencies as this case isn't handled correctly currently).
+ for (List<Cell> cells : m.getFamilyCellMap().values()) {
+ for (Cell cell : cells) {
+ CellUtil.setTimestamp(cell, now);
+ }
+ }
+ // Only copy mutations if we found duplicate rows
+ // which only occurs when we're partially rebuilding
+ // the index (since we'll potentially have both a
+ // Put and a Delete mutation for the same row).
+ if (copyMutations) {
+ // Add the mutation to the batch set
+ ImmutableBytesPtr row = new ImmutableBytesPtr(m.getRow());
+ MultiMutation stored = mutationsMap.get(row);
+ // we haven't seen this row before, so add it
+ if (stored == null) {
+ stored = new MultiMutation(row);
+ mutationsMap.put(row, stored);
+ }
+ stored.addAll(m);
+ } else {
+ originalMutations.add(m);
+ }
+ }
+ }
+ if (copyMutations) {
+ mutations = IndexManagementUtil.flattenMutationsByTimestamp(mutations);
+ }
+ return mutations;
+ }
- for (int i = 0; i < miniBatchOp.size(); i++) {
- Mutation m = miniBatchOp.getOperation(i);
- // skip this mutation if we aren't enabling indexing
- // unfortunately, we really should ask if the raw mutation (rather than the combined mutation)
- // should be indexed, which means we need to expose another method on the builder. Such is the
- // way optimization go though.
- if (miniBatchOp.getOperationStatus(i) != IGNORE && this.builder.isEnabled(m)) {
- if (resetTimeStamp) {
- // Unless we're replaying edits to rebuild the index, we update the time stamp
- // of the data table to prevent overlapping time stamps (which prevents index
- // inconsistencies as this case isn't handled correctly currently).
- for (List<Cell> cells : m.getFamilyCellMap().values()) {
- for (Cell cell : cells) {
- CellUtil.setTimestamp(cell, now);
- }
- }
- }
- // No need to write the table mutations when we're rebuilding
- // the index as they're already written and just being replayed.
- if (replayWrite == ReplayWrite.INDEX_ONLY
- || replayWrite == ReplayWrite.REBUILD_INDEX_ONLY) {
- miniBatchOp.setOperationStatus(i, NOWRITE);
- }
+ public static void setTimestamp(Mutation m, long ts) throws IOException {
+ for (List<Cell> cells : m.getFamilyCellMap().values()) {
+ for (Cell cell : cells) {
+ CellUtil.setTimestamp(cell, ts);
+ }
+ }
+ }
- // Only copy mutations if we found duplicate rows
- // which only occurs when we're partially rebuilding
- // the index (since we'll potentially have both a
- // Put and a Delete mutation for the same row).
- if (copyMutations) {
- // Add the mutation to the batch set
-
- ImmutableBytesPtr row = new ImmutableBytesPtr(m.getRow());
- MultiMutation stored = mutationsMap.get(row);
- // we haven't seen this row before, so add it
- if (stored == null) {
- stored = new MultiMutation(row);
- mutationsMap.put(row, stored);
- }
- stored.addAll(m);
- } else {
- originalMutations.add(m);
- }
- }
- }
+ /**
+ * This method applies pending delete mutations on the next row states
+ */
+ private void applyPendingDeleteMutations(MiniBatchOperationInProgress<Mutation> miniBatchOp,
+ BatchMutateContext context) throws IOException {
+ for (Integer i = 0; i < miniBatchOp.size(); i++) {
+ if (miniBatchOp.getOperationStatus(i) == IGNORE) {
+ continue;
+ }
+ Mutation m = miniBatchOp.getOperation(i);
+ if (!this.builder.isEnabled(m)) {
+ continue;
+ }
+ if (!(m instanceof Delete)) {
+ continue;
+ }
+ ImmutableBytesPtr rowKeyPtr = new ImmutableBytesPtr(m.getRow());
+ Pair<Put, Put> dataRowState = context.dataRowStates.get(rowKeyPtr);
+ Put nextDataRowState = dataRowState.getSecond();
+ if (nextDataRowState == null) {
+ continue;
+ }
+ for (List<Cell> cells : m.getFamilyCellMap().values()) {
+ for (Cell cell : cells) {
+ switch (KeyValue.Type.codeToType(cell.getTypeByte())) {
+ case DeleteFamily:
+ case DeleteFamilyVersion:
+ nextDataRowState.getFamilyCellMap().remove(CellUtil.cloneFamily(cell));
+ break;
+ case DeleteColumn:
+ case Delete:
+ removeColumn(nextDataRowState, cell);
+ }
+ }
+ }
+ if (nextDataRowState != null && nextDataRowState.getFamilyCellMap().size() == 0) {
+ dataRowState.setSecond(null);
+ }
+ }
+ }
- if (copyMutations || replayWrite != null) {
- mutations = IndexManagementUtil.flattenMutationsByTimestamp(mutations);
- }
- return mutations;
- }
+ /**
+ * This method applies the pending put mutations on the the next row states.
+ * Before this method is called, the next row states is set to current row states.
+ */
+ private void applyPendingPutMutations(MiniBatchOperationInProgress<Mutation> miniBatchOp,
+ BatchMutateContext context, long now) throws IOException {
+ for (Integer i = 0; i < miniBatchOp.size(); i++) {
+ if (miniBatchOp.getOperationStatus(i) == IGNORE) {
+ continue;
+ }
+ Mutation m = miniBatchOp.getOperation(i);
+ // skip this mutation if we aren't enabling indexing
+ if (!this.builder.isEnabled(m)) {
+ continue;
+ }
+ // Unless we're replaying edits to rebuild the index, we update the time stamp
+ // of the data table to prevent overlapping time stamps (which prevents index
+ // inconsistencies as this case isn't handled correctly currently).
+ setTimestamp(m, now);
+ if (m instanceof Put) {
+ ImmutableBytesPtr rowKeyPtr = new ImmutableBytesPtr(m.getRow());
+ Pair<Put, Put> dataRowState = context.dataRowStates.get(rowKeyPtr);
+ if (dataRowState == null) {
+ dataRowState = new Pair<Put, Put>(null, null);
+ context.dataRowStates.put(rowKeyPtr, dataRowState);
+ }
+ Put nextDataRowState = dataRowState.getSecond();
+ dataRowState.setSecond((nextDataRowState != null) ? applyNew((Put) m, nextDataRowState) : new Put((Put) m));
+ }
+ }
+ }
- public static void removeEmptyColumn(Mutation m, byte[] emptyCF, byte[] emptyCQ) {
- List<Cell> cellList = m.getFamilyCellMap().get(emptyCF);
- if (cellList == null) {
- return;
- }
- Iterator<Cell> cellIterator = cellList.iterator();
- while (cellIterator.hasNext()) {
- Cell cell = cellIterator.next();
- if (Bytes.compareTo(cell.getQualifierArray(), cell.getQualifierOffset(), cell.getQualifierLength(),
- emptyCQ, 0, emptyCQ.length) == 0) {
- cellIterator.remove();
- return;
- }
- }
- }
+ /**
+ * * Prepares data row current and next row states
+ */
+ private void prepareDataRowStates(ObserverContext<RegionCoprocessorEnvironment> c,
+ MiniBatchOperationInProgress<Mutation> miniBatchOp,
+ BatchMutateContext context,
+ long now) throws IOException {
+ if (context.rowsToLock.size() == 0) {
+ return;
+ }
+ // Retrieve the current row states from the data table
+ getCurrentRowStates(c, context);
+ applyPendingPutMutations(miniBatchOp, context, now);
+ applyPendingDeleteMutations(miniBatchOp, context);
+ }
- private void handleLocalIndexUpdates(ObserverContext<RegionCoprocessorEnvironment> c,
- MiniBatchOperationInProgress<Mutation> miniBatchOp,
- ListMultimap<HTableInterfaceReference, Pair<Mutation, byte[]>> indexUpdates) {
- byte[] tableName = c.getEnvironment().getRegion().getTableDesc().getTableName().getName();
- HTableInterfaceReference hTableInterfaceReference =
- new HTableInterfaceReference(new ImmutableBytesPtr(tableName));
- List<Pair<Mutation, byte[]>> localIndexUpdates = indexUpdates.removeAll(hTableInterfaceReference);
- if (localIndexUpdates == null || localIndexUpdates.isEmpty()) {
- return;
- }
- List<Mutation> localUpdates = new ArrayList<Mutation>();
- Iterator<Pair<Mutation, byte[]>> indexUpdatesItr = localIndexUpdates.iterator();
- while (indexUpdatesItr.hasNext()) {
- Pair<Mutation, byte[]> next = indexUpdatesItr.next();
- localUpdates.add(next.getFirst());
- }
- if (!localUpdates.isEmpty()) {
- miniBatchOp.addOperationsFromCP(0, localUpdates.toArray(new Mutation[localUpdates.size()]));
- }
- }
+ public static void removeEmptyColumn(Mutation m, byte[] emptyCF, byte[] emptyCQ) {
+ List<Cell> cellList = m.getFamilyCellMap().get(emptyCF);
+ if (cellList == null) {
+ return;
+ }
+ Iterator<Cell> cellIterator = cellList.iterator();
+ while (cellIterator.hasNext()) {
+ Cell cell = cellIterator.next();
+ if (Bytes.compareTo(cell.getQualifierArray(), cell.getQualifierOffset(), cell.getQualifierLength(),
+ emptyCQ, 0, emptyCQ.length) == 0) {
+ cellIterator.remove();
+ return;
+ }
+ }
+ }
- private void prepareIndexMutations(
- ObserverContext<RegionCoprocessorEnvironment> c,
- MiniBatchOperationInProgress<Mutation> miniBatchOp,
- BatchMutateContext context,
- Collection<? extends Mutation> mutations,
- long now,
- PhoenixIndexMetaData indexMetaData) throws Throwable {
- List<IndexMaintainer> maintainers = indexMetaData.getIndexMaintainers();
- // get the current span, or just use a null-span to avoid a bunch of if statements
- try (TraceScope scope = Trace.startSpan("Starting to build index updates")) {
- Span current = scope.getSpan();
- if (current == null) {
- current = NullSpan.INSTANCE;
- }
- // get the index updates for all elements in this batch
- context.indexUpdates = ArrayListMultimap.<HTableInterfaceReference, Pair<Mutation, byte[]>>create();
- this.builder.getIndexUpdates(context.indexUpdates, miniBatchOp, mutations, indexMetaData);
- current.addTimelineAnnotation("Built index updates, doing preStep");
- handleLocalIndexUpdates(c, miniBatchOp, context.indexUpdates);
- context.preIndexUpdates = ArrayListMultimap.<HTableInterfaceReference, Mutation>create();
- int updateCount = 0;
- for (IndexMaintainer indexMaintainer : maintainers) {
- updateCount++;
- byte[] emptyCF = indexMaintainer.getEmptyKeyValueFamily().copyBytesIfNecessary();
- byte[] emptyCQ = indexMaintainer.getEmptyKeyValueQualifier();
- HTableInterfaceReference hTableInterfaceReference =
- new HTableInterfaceReference(new ImmutableBytesPtr(indexMaintainer.getIndexTableName()));
- List <Pair<Mutation, byte[]>> updates = context.indexUpdates.get(hTableInterfaceReference);
- for (Pair<Mutation, byte[]> update : updates) {
- // add the VERIFIED cell, which is the empty cell
- Mutation m = update.getFirst();
- if (context.rebuild) {
- if (m instanceof Put) {
- long ts = getMaxTimestamp(m);
- // Remove the empty column prepared by Index codec as we need to change its value
- removeEmptyColumn(m, emptyCF, emptyCQ);
- ((Put) m).addColumn(emptyCF, emptyCQ, ts, VERIFIED_BYTES);
- }
- context.preIndexUpdates.put(hTableInterfaceReference, m);
- } else {
- if (m instanceof Put) {
- // Remove the empty column prepared by Index codec as we need to change its value
- removeEmptyColumn(m, emptyCF, emptyCQ);
- // Set the status of the index row to "unverified"
- ((Put) m).addColumn(emptyCF, emptyCQ, now, UNVERIFIED_BYTES);
- // This will be done before the data table row is updated (i.e., in the first write phase)
- context.preIndexUpdates.put(hTableInterfaceReference, m);
- }
- else {
- // Set the status of the index row to "unverified"
- Put unverifiedPut = new Put(m.getRow());
- unverifiedPut.addColumn(emptyCF, emptyCQ, now, UNVERIFIED_BYTES);
- // This will be done before the data table row is updated (i.e., in the first write phase)
- context.preIndexUpdates.put(hTableInterfaceReference, unverifiedPut);
- }
- }
- }
- }
- TracingUtils.addAnnotation(current, "index update count", updateCount);
- }
- }
+ /**
+ * The index update generation for local indexes uses the existing index update generation code (i.e.,
+ * the {@link IndexBuilder} implementation).
+ */
+ private void handleLocalIndexUpdates(ObserverContext<RegionCoprocessorEnvironment> c,
+ MiniBatchOperationInProgress<Mutation> miniBatchOp,
+ Collection<? extends Mutation> pendingMutations,
+ PhoenixIndexMetaData indexMetaData) throws Throwable {
+ ListMultimap<HTableInterfaceReference, Pair<Mutation, byte[]>> indexUpdates = ArrayListMultimap.<HTableInterfaceReference, Pair<Mutation, byte[]>>create();
+ this.builder.getIndexUpdates(indexUpdates, miniBatchOp, pendingMutations, indexMetaData);
+ byte[] tableName = c.getEnvironment().getRegion().getTableDesc().getTableName().getName();
+ HTableInterfaceReference hTableInterfaceReference =
+ new HTableInterfaceReference(new ImmutableBytesPtr(tableName));
+ List<Pair<Mutation, byte[]>> localIndexUpdates = indexUpdates.removeAll(hTableInterfaceReference);
+ if (localIndexUpdates == null || localIndexUpdates.isEmpty()) {
+ return;
+ }
+ List<Mutation> localUpdates = new ArrayList<Mutation>();
+ Iterator<Pair<Mutation, byte[]>> indexUpdatesItr = localIndexUpdates.iterator();
+ while (indexUpdatesItr.hasNext()) {
+ Pair<Mutation, byte[]> next = indexUpdatesItr.next();
+ localUpdates.add(next.getFirst());
+ }
+ if (!localUpdates.isEmpty()) {
+ miniBatchOp.addOperationsFromCP(0, localUpdates.toArray(new Mutation[localUpdates.size()]));
+ }
+ }
+ /**
+ * Retrieve the the last committed data row state. This method is called only for regular data mutations since for
+ * rebuild (i.e., index replay) mutations include all row versions.
+ */
+ private void getCurrentRowStates(ObserverContext<RegionCoprocessorEnvironment> c,
+ BatchMutateContext context) throws IOException {
+ Set<KeyRange> keys = new HashSet<KeyRange>(context.rowsToLock.size());
+ for (ImmutableBytesPtr rowKeyPtr : context.rowsToLock) {
+ keys.add(PVarbinary.INSTANCE.getKeyRange(rowKeyPtr.get()));
+ }
+ Scan scan = new Scan();
+ ScanRanges scanRanges = ScanRanges.createPointLookup(new ArrayList<KeyRange>(keys));
+ scanRanges.initializeScan(scan);
+ SkipScanFilter skipScanFilter = scanRanges.getSkipScanFilter();
+ scan.setFilter(skipScanFilter);
+ context.dataRowStates = new HashMap<ImmutableBytesPtr, Pair<Put, Put>>(context.rowsToLock.size());
+ try (RegionScanner scanner = c.getEnvironment().getRegion().getScanner(scan)) {
+ boolean more = true;
+ while(more) {
+ List<Cell> cells = new ArrayList<Cell>();
+ more = scanner.next(cells);
+ if (cells.isEmpty()) {
+ continue;
+ }
+ byte[] rowKey = CellUtil.cloneRow(cells.get(0));
+ Put put = new Put(rowKey);
+ for (Cell cell : cells) {
+ put.add(cell);
+ }
+ context.dataRowStates.put(new ImmutableBytesPtr(rowKey), new Pair<Put, Put>(put, new Put(put)));
+ }
+ }
+ }
- protected PhoenixIndexMetaData getPhoenixIndexMetaData(
- ObserverContext<RegionCoprocessorEnvironment> observerContext,
- MiniBatchOperationInProgress<Mutation> miniBatchOp) throws IOException {
- IndexMetaData indexMetaData = this.builder.getIndexMetaData(miniBatchOp);
- if (!(indexMetaData instanceof PhoenixIndexMetaData)) {
- throw new DoNotRetryIOException(
- "preBatchMutateWithExceptions: indexMetaData is not an instance of "+PhoenixIndexMetaData.class.getName() +
- ", current table is:" +
- observerContext.getEnvironment().getRegion().getRegionInfo().getTable().getNameAsString());
- }
- return (PhoenixIndexMetaData)indexMetaData;
- }
+ /**
+ * Generate the index update for a data row from the mutation that are obtained by merging the previous data row
+ * state with the pending row mutation.
+ */
+ private void prepareIndexMutations(BatchMutateContext context, List<IndexMaintainer> maintainers, long ts)
+ throws IOException {
+ List<Pair<IndexMaintainer, HTableInterfaceReference>> indexTables = new ArrayList<>(maintainers.size());
+ for (IndexMaintainer indexMaintainer : maintainers) {
+ if (indexMaintainer.isLocalIndex()) {
+ continue;
+ }
+ HTableInterfaceReference hTableInterfaceReference =
+ new HTableInterfaceReference(new ImmutableBytesPtr(indexMaintainer.getIndexTableName()));
+ indexTables.add(new Pair<>(indexMaintainer, hTableInterfaceReference));
+ }
+ for (Map.Entry<ImmutableBytesPtr, Pair<Put, Put>> entry : context.dataRowStates.entrySet()) {
+ ImmutableBytesPtr rowKeyPtr = entry.getKey();
+ Pair<Put, Put> dataRowState = entry.getValue();
+ Put currentDataRowState = dataRowState.getFirst();
+ Put nextDataRowState = dataRowState.getSecond();
+ if (currentDataRowState == null && nextDataRowState == null) {
+ continue;
+ }
+ for (Pair<IndexMaintainer, HTableInterfaceReference> pair : indexTables) {
+ IndexMaintainer indexMaintainer = pair.getFirst();
+ HTableInterfaceReference hTableInterfaceReference = pair.getSecond();
+ if (nextDataRowState != null) {
+ ValueGetter nextDataRowVG = new IndexRebuildRegionScanner.SimpleValueGetter(nextDataRowState);
+ Put indexPut = indexMaintainer.buildUpdateMutation(GenericKeyValueBuilder.INSTANCE,
+ nextDataRowVG, rowKeyPtr, ts, null, null);
+ if (indexPut == null) {
+ // No covered column. Just prepare an index row with the empty column
+ byte[] indexRowKey = indexMaintainer.buildRowKey(nextDataRowVG, rowKeyPtr,
+ null, null, HConstants.LATEST_TIMESTAMP);
+ indexPut = new Put(indexRowKey);
+ }
+ indexPut.addColumn(indexMaintainer.getEmptyKeyValueFamily().copyBytesIfNecessary(),
+ indexMaintainer.getEmptyKeyValueQualifier(), ts, UNVERIFIED_BYTES);
+ context.indexUpdates.put(hTableInterfaceReference,
+ new Pair<Mutation, byte[]>(indexPut, rowKeyPtr.get()));
+ // Delete the current index row if the new index key is different than the current one
+ if (currentDataRowState != null) {
+ ValueGetter currentDataRowVG = new IndexRebuildRegionScanner.SimpleValueGetter(currentDataRowState);
+ byte[] indexRowKeyForCurrentDataRow = indexMaintainer.buildRowKey(currentDataRowVG, rowKeyPtr,
+ null, null, HConstants.LATEST_TIMESTAMP);
+ if (Bytes.compareTo(indexPut.getRow(), indexRowKeyForCurrentDataRow) != 0) {
+ Mutation del = indexMaintainer.buildRowDeleteMutation(indexRowKeyForCurrentDataRow,
+ IndexMaintainer.DeleteType.ALL_VERSIONS, ts);
+ context.indexUpdates.put(hTableInterfaceReference,
+ new Pair<Mutation, byte[]>(del, rowKeyPtr.get()));
+ }
+ }
+ } else if (currentDataRowState != null) {
+ ValueGetter currentDataRowVG = new IndexRebuildRegionScanner.SimpleValueGetter(currentDataRowState);
+ byte[] indexRowKeyForCurrentDataRow = indexMaintainer.buildRowKey(currentDataRowVG, rowKeyPtr,
+ null, null, HConstants.LATEST_TIMESTAMP);
+ Mutation del = indexMaintainer.buildRowDeleteMutation(indexRowKeyForCurrentDataRow,
+ IndexMaintainer.DeleteType.ALL_VERSIONS, ts);
+ context.indexUpdates.put(hTableInterfaceReference,
+ new Pair<Mutation, byte[]>(del, rowKeyPtr.get()));
+ }
+ }
+ }
+ }
- private void preparePostIndexMutations(
- BatchMutateContext context,
- long now,
- PhoenixIndexMetaData indexMetaData) throws Throwable {
- context.postIndexUpdates = ArrayListMultimap.<HTableInterfaceReference, Mutation>create();
- List<IndexMaintainer> maintainers = indexMetaData.getIndexMaintainers();
- // Check if we need to skip post index update for any of the rows
- for (IndexMaintainer indexMaintainer : maintainers) {
- byte[] emptyCF = indexMaintainer.getEmptyKeyValueFamily().copyBytesIfNecessary();
- byte[] emptyCQ = indexMaintainer.getEmptyKeyValueQualifier();
- HTableInterfaceReference hTableInterfaceReference =
- new HTableInterfaceReference(new ImmutableBytesPtr(indexMaintainer.getIndexTableName()));
- List <Pair<Mutation, byte[]>> updates = context.indexUpdates.get(hTableInterfaceReference);
- for (Pair<Mutation, byte[]> update : updates) {
- // Are there concurrent updates on the data table row? if so, skip post index updates
- // and let read repair resolve conflicts
- ImmutableBytesPtr rowKey = new ImmutableBytesPtr(update.getSecond());
- PendingRow pendingRow = pendingRows.get(rowKey);
- if (!pendingRow.isConcurrent()) {
- Mutation m = update.getFirst();
- if (m instanceof Put) {
- Put verifiedPut = new Put(m.getRow());
- // Set the status of the index row to "verified"
- verifiedPut.addColumn(emptyCF, emptyCQ, now, VERIFIED_BYTES);
- context.postIndexUpdates.put(hTableInterfaceReference, verifiedPut);
- }
- else {
- context.postIndexUpdates.put(hTableInterfaceReference, m);
- }
+ /**
+ * This method prepares unverified index mutations which are applied to index tables before the data table is
+ * updated. In the three phase update approach, in phase 1, the status of existing index rows is set to "unverified"
+ * (these rows will be deleted from the index table in phase 3), and/or new put mutations are added with the
+ * unverified status. In phase 2, data table mutations are applied. In phase 3, the status for an index table row is
+ * either set to "verified" or the row is deleted.
+ */
+ private void preparePreIndexMutations(ObserverContext<RegionCoprocessorEnvironment> c,
+ MiniBatchOperationInProgress<Mutation> miniBatchOp,
+ BatchMutateContext context,
+ Collection<? extends Mutation> pendingMutations,
+ long now,
+ PhoenixIndexMetaData indexMetaData) throws Throwable {
+ List<IndexMaintainer> maintainers = indexMetaData.getIndexMaintainers();
+ // get the current span, or just use a null-span to avoid a bunch of if statements
+ try (TraceScope scope = Trace.startSpan("Starting to build index updates")) {
+ Span current = scope.getSpan();
+ if (current == null) {
+ current = NullSpan.INSTANCE;
+ }
+ current.addTimelineAnnotation("Built index updates, doing preStep");
+ // Handle local index updates
+ for (IndexMaintainer indexMaintainer : maintainers) {
+ if (indexMaintainer.isLocalIndex()) {
+ handleLocalIndexUpdates(c, miniBatchOp, pendingMutations, indexMetaData);
+ break;
+ }
+ }
+ // The rest of this method is for handling global index updates
+ context.indexUpdates = ArrayListMultimap.<HTableInterfaceReference, Pair<Mutation, byte[]>>create();
+ prepareIndexMutations(context, maintainers, now);
+
+ context.preIndexUpdates = ArrayListMultimap.<HTableInterfaceReference, Mutation>create();
+ int updateCount = 0;
+ for (IndexMaintainer indexMaintainer : maintainers) {
+ updateCount++;
+ byte[] emptyCF = indexMaintainer.getEmptyKeyValueFamily().copyBytesIfNecessary();
+ byte[] emptyCQ = indexMaintainer.getEmptyKeyValueQualifier();
+ HTableInterfaceReference hTableInterfaceReference =
+ new HTableInterfaceReference(new ImmutableBytesPtr(indexMaintainer.getIndexTableName()));
+ List <Pair<Mutation, byte[]>> updates = context.indexUpdates.get(hTableInterfaceReference);
+ for (Pair<Mutation, byte[]> update : updates) {
+ Mutation m = update.getFirst();
+ if (m instanceof Put) {
+ // This will be done before the data table row is updated (i.e., in the first write phase)
+ context.preIndexUpdates.put(hTableInterfaceReference, m);
+ } else {
+ // Set the status of the index row to "unverified"
+ Put unverifiedPut = new Put(m.getRow());
+ unverifiedPut.addColumn(emptyCF, emptyCQ, now, UNVERIFIED_BYTES);
+ // This will be done before the data table row is updated (i.e., in the first write phase)
+ context.preIndexUpdates.put(hTableInterfaceReference, unverifiedPut);
+ }
+ }
+ }
+ TracingUtils.addAnnotation(current, "index update count", updateCount);
+ }
+ }
- }
- }
- }
- // We are done with handling concurrent mutations. So we can remove the rows of this batch from
- // the collection of pending rows
- removePendingRows(context);
- context.indexUpdates.clear();
- }
+ protected PhoenixIndexMetaData getPhoenixIndexMetaData(ObserverContext<RegionCoprocessorEnvironment> observerContext,
+ MiniBatchOperationInProgress<Mutation> miniBatchOp)
+ throws IOException {
+ IndexMetaData indexMetaData = this.builder.getIndexMetaData(miniBatchOp);
+ if (!(indexMetaData instanceof PhoenixIndexMetaData)) {
+ throw new DoNotRetryIOException(
+ "preBatchMutateWithExceptions: indexMetaData is not an instance of "+PhoenixIndexMetaData.class.getName() +
+ ", current table is:" +
+ observerContext.getEnvironment().getRegion().getRegionInfo().getTable().getNameAsString());
+ }
+ return (PhoenixIndexMetaData)indexMetaData;
+ }
- public void preBatchMutateWithExceptions(ObserverContext<RegionCoprocessorEnvironment> c,
- MiniBatchOperationInProgress<Mutation> miniBatchOp) throws Throwable {
- ignoreAtomicOperations(miniBatchOp);
- PhoenixIndexMetaData indexMetaData = getPhoenixIndexMetaData(c, miniBatchOp);
- BatchMutateContext context = new BatchMutateContext(indexMetaData.getClientVersion());
- setBatchMutateContext(c, context);
- Mutation firstMutation = miniBatchOp.getOperation(0);
- ReplayWrite replayWrite = this.builder.getReplayWrite(firstMutation);
- context.rebuild = replayWrite != null;
- /*
- * Exclusively lock all rows so we get a consistent read
- * while determining the index updates
- */
- long now;
- if (!context.rebuild) {
- populateRowsToLock(miniBatchOp, context);
- lockRows(context);
- now = EnvironmentEdgeManager.currentTimeMillis();
- // Add the table rows in the mini batch to the collection of pending rows. This will be used to detect
- // concurrent updates
- populatePendingRows(context);
- }
- else {
- now = EnvironmentEdgeManager.currentTimeMillis();
- }
- // First group all the updates for a single row into a single update to be processed
- Collection<? extends Mutation> mutations = groupMutations(miniBatchOp, now, replayWrite);
- // early exit if it turns out we don't have any edits
- if (mutations == null) {
- return;
- }
+ private void preparePostIndexMutations(BatchMutateContext context, long now, PhoenixIndexMetaData indexMetaData)
+ throws Throwable {
+ context.postIndexUpdates = ArrayListMultimap.<HTableInterfaceReference, Mutation>create();
+ List<IndexMaintainer> maintainers = indexMetaData.getIndexMaintainers();
+ // Check if we need to skip post index update for any of the rows
+ for (IndexMaintainer indexMaintainer : maintainers) {
+ byte[] emptyCF = indexMaintainer.getEmptyKeyValueFamily().copyBytesIfNecessary();
+ byte[] emptyCQ = indexMaintainer.getEmptyKeyValueQualifier();
+ HTableInterfaceReference hTableInterfaceReference =
+ new HTableInterfaceReference(new ImmutableBytesPtr(indexMaintainer.getIndexTableName()));
+ List <Pair<Mutation, byte[]>> updates = context.indexUpdates.get(hTableInterfaceReference);
+ for (Pair<Mutation, byte[]> update : updates) {
+ // Are there concurrent updates on the data table row? if so, skip post index updates
+ // and let read repair resolve conflicts
+ ImmutableBytesPtr rowKey = new ImmutableBytesPtr(update.getSecond());
+ PendingRow pendingRow = pendingRows.get(rowKey);
+ if (!pendingRow.isConcurrent()) {
+ Mutation m = update.getFirst();
+ if (m instanceof Put) {
+ Put verifiedPut = new Put(m.getRow());
+ // Set the status of the index row to "verified"
+ verifiedPut.addColumn(emptyCF, emptyCQ, now, VERIFIED_BYTES);
+ context.postIndexUpdates.put(hTableInterfaceReference, verifiedPut);
+ }
+ else {
+ context.postIndexUpdates.put(hTableInterfaceReference, m);
+ }
+ }
+ }
+ }
+ // We are done with handling concurrent mutations. So we can remove the rows of this batch from
+ // the collection of pending rows
+ removePendingRows(context);
+ context.indexUpdates.clear();
+ }
- long start = EnvironmentEdgeManager.currentTimeMillis();
- prepareIndexMutations(c, miniBatchOp, context, mutations, now, indexMetaData);
- metricSource.updateIndexPrepareTime(EnvironmentEdgeManager.currentTimeMillis() - start);
-
- // Sleep for one millisecond if we have prepared the index updates in less than 1 ms. The sleep is necessary to
- // get different timestamps for concurrent batches that share common rows. It is very rare that the index updates
- // can be prepared in less than one millisecond
- if (!context.rowLocks.isEmpty() && now == EnvironmentEdgeManager.currentTimeMillis()) {
- Thread.sleep(1);
- LOG.debug("slept 1ms for " + c.getEnvironment().getRegion().getRegionInfo().getTable().getNameAsString());
- }
- // Release the locks before making RPC calls for index updates
- for (RowLock rowLock : context.rowLocks) {
- rowLock.release();
- }
- // Do the first phase index updates
- doPre(c, context, miniBatchOp);
- if (!context.rebuild) {
- // Acquire the locks again before letting the region proceed with data table updates
- List<RowLock> rowLocks = Lists.newArrayListWithExpectedSize(context.rowLocks.size());
- for (RowLock rowLock : context.rowLocks) {
- rowLocks.add(lockManager.lockRow(rowLock.getRowKey(), rowLockWaitDuration));
- }
- context.rowLocks.clear();
- context.rowLocks = rowLocks;
- preparePostIndexMutations(context, now, indexMetaData);
- }
- if (failDataTableUpdatesForTesting) {
- throw new DoNotRetryIOException("Simulating the data table write failure");
- }
- }
+ /**
+ * There are at most two rebuild mutation for every row, one put and one delete. They are listed in indexMutations
+ * next to each other such that put comes before delete by {@link IndexRebuildRegionScanner}. This method is called
+ * only for global indexes.
+ */
+ private void preBatchMutateWithExceptionsForRebuild(ObserverContext<RegionCoprocessorEnvironment> c,
+ MiniBatchOperationInProgress<Mutation> miniBatchOp,
+ BatchMutateContext context,
+ IndexMaintainer indexMaintainer) throws Throwable {
+ Put put = null;
+ List <Mutation> indexMutations = new ArrayList<>();
+ for (int i = 0; i < miniBatchOp.size(); i++) {
+ if (miniBatchOp.getOperationStatus(i) == IGNORE) {
+ continue;
+ }
+ Mutation m = miniBatchOp.getOperation(i);
+ if (!this.builder.isEnabled(m)) {
+ continue;
+ }
+ if (m instanceof Put) {
+ if (put != null) {
+ indexMutations.addAll(prepareIndexMutationsForRebuild(indexMaintainer, put, null));
+ }
+ put = (Put)m;
+ } else {
+ indexMutations.addAll(prepareIndexMutationsForRebuild(indexMaintainer, put, (Delete)m));
+ put = null;
+ }
+ miniBatchOp.setOperationStatus(i, NOWRITE);
+ }
+ if (put != null) {
+ indexMutations.addAll(prepareIndexMutationsForRebuild(indexMaintainer, put, null));
+ }
+ HTableInterfaceReference hTableInterfaceReference =
+ new HTableInterfaceReference(new ImmutableBytesPtr(indexMaintainer.getIndexTableName()));
+ context.preIndexUpdates = ArrayListMultimap.<HTableInterfaceReference, Mutation>create();
+ for (Mutation m : indexMutations) {
+ context.preIndexUpdates.put(hTableInterfaceReference, m);
+ }
+ doPre(c, context, miniBatchOp);
+ // For rebuild updates, no post index update is prepared. Just create an empty list.
+ context.postIndexUpdates = ArrayListMultimap.<HTableInterfaceReference, Mutation>create();
+ }
+ public void preBatchMutateWithExceptions(ObserverContext<RegionCoprocessorEnvironment> c,
+ MiniBatchOperationInProgress<Mutation> miniBatchOp) throws Throwable {
+ ignoreAtomicOperations(miniBatchOp);
+ PhoenixIndexMetaData indexMetaData = getPhoenixIndexMetaData(c, miniBatchOp);
+ BatchMutateContext context = new BatchMutateContext(indexMetaData.getClientVersion());
+ setBatchMutateContext(c, context);
+ Mutation firstMutation = miniBatchOp.getOperation(0);
+ ReplayWrite replayWrite = this.builder.getReplayWrite(firstMutation);
+ context.rebuild = replayWrite != null;
+ if (context.rebuild) {
+ preBatchMutateWithExceptionsForRebuild(c, miniBatchOp, context, indexMetaData.getIndexMaintainers().get(0));
+ return;
+ }
+ /*
+ * Exclusively lock all rows so we get a consistent read
+ * while determining the index updates
+ */
+ long now;
+ populateRowsToLock(miniBatchOp, context);
+ lockRows(context);
+ now = EnvironmentEdgeManager.currentTimeMillis();
+ // Add the table rows in the mini batch to the collection of pending rows. This will be used to detect
+ // concurrent updates
+ populatePendingRows(context);
+ // Prepare current and next data rows states for pending mutations (for global indexes)
+ prepareDataRowStates(c, miniBatchOp, context, now);
+ // Group all the updates for a single row into a single update to be processed (for local indexes)
+ Collection<? extends Mutation> mutations = groupMutations(miniBatchOp, now);
+ // early exit if it turns out we don't have any edits
+ if (mutations == null || mutations.isEmpty()) {
+ return;
+ }
+ long start = EnvironmentEdgeManager.currentTimeMillis();
+ preparePreIndexMutations(c, miniBatchOp, context, mutations, now, indexMetaData);
+ metricSource.updateIndexPrepareTime(EnvironmentEdgeManager.currentTimeMillis() - start);
+ // Sleep for one millisecond if we have prepared the index updates in less than 1 ms. The sleep is necessary to
+ // get different timestamps for concurrent batches that share common rows. It is very rare that the index updates
+ // can be prepared in less than one millisecond
+ if (!context.rowLocks.isEmpty() && now == EnvironmentEdgeManager.currentTimeMillis()) {
+ Thread.sleep(1);
+ LOG.debug("slept 1ms for " + c.getEnvironment().getRegion().getRegionInfo().getTable().getNameAsString());
+ }
+ // Release the locks before making RPC calls for index updates
+ for (RowLock rowLock : context.rowLocks) {
+ rowLock.release();
+ }
+ // Do the first phase index updates
+ doPre(c, context, miniBatchOp);
+ // Acquire the locks again before letting the region proceed with data table updates
+ List<RowLock> rowLocks = Lists.newArrayListWithExpectedSize(context.rowLocks.size());
+ for (RowLock rowLock : context.rowLocks) {
+ rowLocks.add(lockManager.lockRow(rowLock.getRowKey(), rowLockWaitDuration));
+ }
+ context.rowLocks.clear();
+ context.rowLocks = rowLocks;
+ preparePostIndexMutations(context, now, indexMetaData);
+ if (failDataTableUpdatesForTesting) {
+ throw new DoNotRetryIOException("Simulating the data table write failure");
+ }
+ }
private void setBatchMutateContext(ObserverContext<RegionCoprocessorEnvironment> c, BatchMutateContext context) {
this.batchMutateContext.set(context);
}
-
+
private BatchMutateContext getBatchMutateContext(ObserverContext<RegionCoprocessorEnvironment> c) {
return this.batchMutateContext.get();
}
-
+
private void removeBatchMutateContext(ObserverContext<RegionCoprocessorEnvironment> c) {
this.batchMutateContext.remove();
}
@@ -822,14 +1053,6 @@ public class IndexRegionObserver extends BaseRegionObserver {
}
/**
- * Exposed for testing!
- * @return the currently instantiated index builder
- */
- public IndexBuilder getBuilderForTesting() {
- return this.builder.getBuilderForTesting();
- }
-
- /**
* Enable indexing on the given table
* @param desc {@link HTableDescriptor} for the table on which indexing should be enabled
* @param builder class to use when building the index for this table
diff --git a/phoenix-core/src/main/java/org/apache/phoenix/index/IndexMaintainer.java b/phoenix-core/src/main/java/org/apache/phoenix/index/IndexMaintainer.java
index dba165b..3723388 100644
--- a/phoenix-core/src/main/java/org/apache/phoenix/index/IndexMaintainer.java
+++ b/phoenix-core/src/main/java/org/apache/phoenix/index/IndexMaintainer.java
@@ -1593,7 +1593,7 @@ public class IndexMaintainer implements Writable, Iterable<ColumnReference> {
*/
private void initCachedState() {
byte[] emptyKvQualifier = EncodedColumnsUtil.getEmptyKeyValueInfo(encodingScheme).getFirst();
- dataEmptyKeyValueRef = new ColumnReference(emptyKeyValueCFPtr.copyBytesIfNecessary(), emptyKvQualifier);
+ dataEmptyKeyValueRef = new ColumnReference(dataEmptyKeyValueCF, emptyKvQualifier);
this.allColumns = Sets.newLinkedHashSetWithExpectedSize(indexedExpressions.size() + coveredColumnsMap.size());
// columns that are required to evaluate all expressions in indexedExpressions (not including columns in data row key)
this.indexedColumns = Sets.newLinkedHashSetWithExpectedSize(indexedExpressions.size());