HubSpot Backport: HBASE-28568 Incremental backup set does not correct…

…ly shrink (apache#5876) (#97) * HubSpot Backport: HBASE-28568 Incremental backup set does not correctly shrink (apache#5876) The incremental backup set is the set of tables included when an incremental backup is created, it is managed per backup root dir and contains all tables that are present in at least one backup (in that root dir). The incremental backup set can only shrink when backups are deleted. However, the implementation was incorrect, causing this set to never be able to shrink. Reviewed-by: Ray Mattingly <[email protected]> Signed-off-by: Nick Dimiduk <[email protected]> * HubSpot Backport: HBASE-28568 Incremental backup set does not correctly shrink (addendum) (apache#5917) Import the correct shaded Guava and run spotless:apply. Signed-off-by: Duo Zhang <[email protected]> --------- Signed-off-by: Nick Dimiduk <[email protected]> Signed-off-by: Duo Zhang <[email protected]> Co-authored-by: DieterDP <[email protected]>
HubSpot · Jul 10, 2024 · 4187f4b · 4187f4b
1 parent a6eaaff
commit 4187f4b
Show file tree

Hide file tree

Showing 3 changed files with 58 additions and 39 deletions.
diff --git a/hbase-backup/src/main/java/org/apache/hadoop/hbase/backup/impl/BackupAdminImpl.java b/hbase-backup/src/main/java/org/apache/hadoop/hbase/backup/impl/BackupAdminImpl.java
@@ -21,7 +21,6 @@
 import java.util.ArrayList;
 import java.util.Arrays;
 import java.util.Collections;
-import java.util.HashMap;
 import java.util.HashSet;
 import java.util.List;
 import java.util.Map;
@@ -94,7 +93,6 @@ public BackupInfo getBackupInfo(String backupId) throws IOException {
   public int deleteBackups(String[] backupIds) throws IOException {
 
     int totalDeleted = 0;
-    Map<String, HashSet<TableName>> allTablesMap = new HashMap<>();
 
     boolean deleteSessionStarted;
     boolean snapshotDone;
@@ -130,20 +128,16 @@ public int deleteBackups(String[] backupIds) throws IOException {
       }
       snapshotDone = true;
       try {
+        List<String> affectedBackupRootDirs = new ArrayList<>();
         for (int i = 0; i < backupIds.length; i++) {
           BackupInfo info = sysTable.readBackupInfo(backupIds[i]);
-          if (info != null) {
-            String rootDir = info.getBackupRootDir();
-            HashSet<TableName> allTables = allTablesMap.get(rootDir);
-            if (allTables == null) {
-              allTables = new HashSet<>();
-              allTablesMap.put(rootDir, allTables);
-            }
-            allTables.addAll(info.getTableNames());
-            totalDeleted += deleteBackup(backupIds[i], sysTable);
+          if (info == null) {
+            continue;
           }
+          affectedBackupRootDirs.add(info.getBackupRootDir());
+          totalDeleted += deleteBackup(backupIds[i], sysTable);
         }
-        finalizeDelete(allTablesMap, sysTable);
+        finalizeDelete(affectedBackupRootDirs, sysTable);
         // Finish
         sysTable.finishDeleteOperation();
         // delete snapshot
@@ -176,26 +170,23 @@ public int deleteBackups(String[] backupIds) throws IOException {
 
   /**
    * Updates incremental backup set for every backupRoot
-   * @param tablesMap map [backupRoot: {@code Set<TableName>}]
-   * @param table     backup system table
+   * @param backupRoots backupRoots for which to revise the incremental backup set
+   * @param table       backup system table
    * @throws IOException if a table operation fails
    */
-  private void finalizeDelete(Map<String, HashSet<TableName>> tablesMap, BackupSystemTable table)
+  private void finalizeDelete(List<String> backupRoots, BackupSystemTable table)
     throws IOException {
-    for (String backupRoot : tablesMap.keySet()) {
+    for (String backupRoot : backupRoots) {
       Set<TableName> incrTableSet = table.getIncrementalBackupTableSet(backupRoot);
-      Map<TableName, ArrayList<BackupInfo>> tableMap =
+      Map<TableName, List<BackupInfo>> tableMap =
         table.getBackupHistoryForTableSet(incrTableSet, backupRoot);
-      for (Map.Entry<TableName, ArrayList<BackupInfo>> entry : tableMap.entrySet()) {
-        if (entry.getValue() == null) {
-          // No more backups for a table
-          incrTableSet.remove(entry.getKey());
-        }
-      }
+
+      // Keep only the tables that are present in other backups
+      incrTableSet.retainAll(tableMap.keySet());
+
+      table.deleteIncrementalBackupTableSet(backupRoot);
       if (!incrTableSet.isEmpty()) {
         table.addIncrementalBackupTableSet(incrTableSet, backupRoot);
-      } else { // empty
-        table.deleteIncrementalBackupTableSet(backupRoot);
       }
     }
   }

diff --git a/hbase-backup/src/main/java/org/apache/hadoop/hbase/backup/impl/BackupSystemTable.java b/hbase-backup/src/main/java/org/apache/hadoop/hbase/backup/impl/BackupSystemTable.java
@@ -87,8 +87,9 @@
  * <ul>
  * <li>1. Backup sessions rowkey= "session:"+backupId; value =serialized BackupInfo</li>
  * <li>2. Backup start code rowkey = "startcode:"+backupRoot; value = startcode</li>
- * <li>3. Incremental backup set rowkey="incrbackupset:"+backupRoot; value=[list of tables]</li>
- * <li>4. Table-RS-timestamp map rowkey="trslm:"+backupRoot+table_name; value = map[RS-%3E last WAL
+ * <li>3. Incremental backup set rowkey="incrbackupset:"+backupRoot; table="meta:"+tablename of
+ * include table; value=empty</li>
+ * <li>4. Table-RS-timestamp map rowkey="trslm:"+backupRoot+table_name; value = map[RS-> last WAL
  * timestamp]</li>
  * <li>5. RS - WAL ts map rowkey="rslogts:"+backupRoot +server; value = last WAL timestamp</li>
  * <li>6. WALs recorded rowkey="wals:"+WAL unique file name; value = backupId and full WAL file
@@ -842,23 +843,25 @@ public List<BackupInfo> getBackupHistoryForTable(TableName name) throws IOExcept
     return tableHistory;
   }
 
-  public Map<TableName, ArrayList<BackupInfo>> getBackupHistoryForTableSet(Set<TableName> set,
+  /**
+   * Goes through all backup history corresponding to the provided root folder, and collects all
+   * backup info mentioning each of the provided tables.
+   * @param set        the tables for which to collect the {@code BackupInfo}
+   * @param backupRoot backup destination path to retrieve backup history for
+   * @return a map containing (a subset of) the provided {@code TableName}s, mapped to a list of at
+   *         least one {@code BackupInfo}
+   * @throws IOException if getting the backup history fails
+   */
+  public Map<TableName, List<BackupInfo>> getBackupHistoryForTableSet(Set<TableName> set,
     String backupRoot) throws IOException {
     List<BackupInfo> history = getBackupHistory(backupRoot);
-    Map<TableName, ArrayList<BackupInfo>> tableHistoryMap = new HashMap<>();
-    for (Iterator<BackupInfo> iterator = history.iterator(); iterator.hasNext();) {
-      BackupInfo info = iterator.next();
-      if (!backupRoot.equals(info.getBackupRootDir())) {
-        continue;
-      }
+    Map<TableName, List<BackupInfo>> tableHistoryMap = new HashMap<>();
+    for (BackupInfo info : history) {
       List<TableName> tables = info.getTableNames();
       for (TableName tableName : tables) {
         if (set.contains(tableName)) {
-          ArrayList<BackupInfo> list = tableHistoryMap.get(tableName);
-          if (list == null) {
-            list = new ArrayList<>();
-            tableHistoryMap.put(tableName, list);
-          }
+          List<BackupInfo> list =
+            tableHistoryMap.computeIfAbsent(tableName, k -> new ArrayList<>());
           list.add(info);
         }
       }

diff --git a/hbase-backup/src/test/java/org/apache/hadoop/hbase/backup/TestBackupDelete.java b/hbase-backup/src/test/java/org/apache/hadoop/hbase/backup/TestBackupDelete.java
@@ -17,6 +17,7 @@
  */
 package org.apache.hadoop.hbase.backup;
 
+import static org.junit.Assert.assertEquals;
 import static org.junit.Assert.assertTrue;
 
 import java.io.ByteArrayOutputStream;
@@ -39,6 +40,7 @@
 import org.slf4j.LoggerFactory;
 
 import org.apache.hbase.thirdparty.com.google.common.collect.Lists;
+import org.apache.hbase.thirdparty.com.google.common.collect.Sets;
 
 @Category(LargeTests.class)
 public class TestBackupDelete extends TestBackupBase {
@@ -158,4 +160,27 @@ public long currentTime() {
     LOG.info(baos.toString());
     assertTrue(output.indexOf("Deleted 1 backups") >= 0);
   }
+
+  /**
+   * Verify that backup deletion updates the incremental-backup-set.
+   */
+  @Test
+  public void testBackupDeleteUpdatesIncrementalBackupSet() throws Exception {
+    LOG.info("Test backup delete updates the incremental backup set");
+    BackupSystemTable backupSystemTable = new BackupSystemTable(TEST_UTIL.getConnection());
+
+    String backupId1 = fullTableBackup(Lists.newArrayList(table1, table2));
+    assertTrue(checkSucceeded(backupId1));
+    assertEquals(Sets.newHashSet(table1, table2),
+      backupSystemTable.getIncrementalBackupTableSet(BACKUP_ROOT_DIR));
+
+    String backupId2 = fullTableBackup(Lists.newArrayList(table3));
+    assertTrue(checkSucceeded(backupId2));
+    assertEquals(Sets.newHashSet(table1, table2, table3),
+      backupSystemTable.getIncrementalBackupTableSet(BACKUP_ROOT_DIR));
+
+    getBackupAdmin().deleteBackups(new String[] { backupId1 });
+    assertEquals(Sets.newHashSet(table3),
+      backupSystemTable.getIncrementalBackupTableSet(BACKUP_ROOT_DIR));
+  }
 }