HBASE-12015 Not cleaning Mob data when Mob CF is removed from table.(Pankaj Kumar)

2015-07-13 14:36:44 +05:30 · 2015-07-13 14:36:44 +05:30 · d1f7bcbff7
parent a1060d1390
commit d1f7bcbff7
8 changed files with 105 additions and 10 deletions
--- a/hbase-server/src/main/java/org/apache/hadoop/hbase/master/MasterFileSystem.java
+++ b/hbase-server/src/main/java/org/apache/hadoop/hbase/master/MasterFileSystem.java
@ -46,9 +46,10 @@ import org.apache.hadoop.hbase.ServerName;
 import org.apache.hadoop.hbase.TableDescriptor;
 import org.apache.hadoop.hbase.TableName;
 import org.apache.hadoop.hbase.backup.HFileArchiver;
-import org.apache.hadoop.hbase.client.TableState;
 import org.apache.hadoop.hbase.exceptions.DeserializationException;
 import org.apache.hadoop.hbase.fs.HFileSystem;
+import org.apache.hadoop.hbase.mob.MobConstants;
+import org.apache.hadoop.hbase.mob.MobUtils;
 import org.apache.hadoop.hbase.protobuf.generated.ZooKeeperProtos.SplitLogTask.RecoveryMode;
 import org.apache.hadoop.hbase.regionserver.HRegion;
 import org.apache.hadoop.hbase.wal.DefaultWALProvider;
@ -573,7 +574,7 @@ public class MasterFileSystem {
    //      @see HRegion.checkRegioninfoOnFilesystem()
  }

-  public void deleteFamilyFromFS(HRegionInfo region, byte[] familyName)
+  public void deleteFamilyFromFS(HRegionInfo region, byte[] familyName, boolean hasMob)
      throws IOException {
    // archive family store files
    Path tableDir = FSUtils.getTableDir(rootdir, region.getTable());
@ -590,6 +591,22 @@ public class MasterFileSystem {
            + ")");
      }
    }
+    // archive and delete mob files
+    if (hasMob) {
+      Path mobTableDir =
+          FSUtils.getTableDir(new Path(getRootDir(), MobConstants.MOB_DIR_NAME), region.getTable());
+      HRegionInfo mobRegionInfo = MobUtils.getMobRegionInfo(region.getTable());
+      Path mobFamilyDir =
+          new Path(mobTableDir,
+              new Path(mobRegionInfo.getEncodedName(), Bytes.toString(familyName)));
+      // archive mob family store files
+      MobUtils.archiveMobStoreFiles(conf, fs, mobRegionInfo, mobFamilyDir, familyName);
+      if (!fs.delete(mobFamilyDir, true)) {
+        throw new IOException("Could not delete mob store files for family "
+            + Bytes.toString(familyName) + " from FileSystem region "
+            + mobRegionInfo.getRegionNameAsString() + "(" + mobRegionInfo.getEncodedName() + ")");
+      }
+    }
  }

  public void stop() {
--- a/hbase-server/src/main/java/org/apache/hadoop/hbase/master/procedure/AddColumnFamilyProcedure.java
+++ b/hbase-server/src/main/java/org/apache/hadoop/hbase/master/procedure/AddColumnFamilyProcedure.java
@ -312,7 +312,7 @@ public class AddColumnFamilyProcedure
      // Remove the column family from file system and update the table descriptor to
      // the before-add-column-family-state
      MasterDDLOperationHelper.deleteColumnFamilyFromFileSystem(env, tableName,
-        getRegionInfoList(env), cfDescriptor.getName());
+        getRegionInfoList(env), cfDescriptor.getName(), cfDescriptor.isMobEnabled());

      env.getMasterServices().getTableDescriptors().add(unmodifiedHTableDescriptor);

--- a/hbase-server/src/main/java/org/apache/hadoop/hbase/master/procedure/DeleteColumnFamilyProcedure.java
+++ b/hbase-server/src/main/java/org/apache/hadoop/hbase/master/procedure/DeleteColumnFamilyProcedure.java
@ -57,6 +57,7 @@ public class DeleteColumnFamilyProcedure
  private HTableDescriptor unmodifiedHTableDescriptor;
  private TableName tableName;
  private byte [] familyName;
+  private boolean hasMob;
  private UserGroupInformation user;

  private List<HRegionInfo> regionInfoList;
@ -283,6 +284,8 @@ public class DeleteColumnFamilyProcedure
      throw new InvalidFamilyOperationException("Family '" + getColumnFamilyName()
          + "' does not exist, so it cannot be deleted");
    }
+    // whether mob family
+    hasMob = unmodifiedHTableDescriptor.getFamily(familyName).isMobEnabled();
  }

  /**
@ -334,7 +337,7 @@ public class DeleteColumnFamilyProcedure
   **/
  private void deleteFromFs(final MasterProcedureEnv env) throws IOException {
    MasterDDLOperationHelper.deleteColumnFamilyFromFileSystem(env, tableName,
-      getRegionInfoList(env), familyName);
+      getRegionInfoList(env), familyName, hasMob);
  }

  /**
--- a/hbase-server/src/main/java/org/apache/hadoop/hbase/master/procedure/MasterDDLOperationHelper.java
+++ b/hbase-server/src/main/java/org/apache/hadoop/hbase/master/procedure/MasterDDLOperationHelper.java
@ -91,7 +91,8 @@ public final class MasterDDLOperationHelper {
      final MasterProcedureEnv env,
      final TableName tableName,
      List<HRegionInfo> regionInfoList,
-      final byte[] familyName) throws IOException {
+      final byte[] familyName,
+      boolean hasMob) throws IOException {
    final MasterFileSystem mfs = env.getMasterServices().getMasterFileSystem();
    if (LOG.isDebugEnabled()) {
      LOG.debug("Removing family=" + Bytes.toString(familyName) + " from table=" + tableName);
@ -101,7 +102,7 @@ public final class MasterDDLOperationHelper {
    }
    for (HRegionInfo hri : regionInfoList) {
      // Delete the family directory in FS for all the regions one by one
-      mfs.deleteFamilyFromFS(hri, familyName);
+      mfs.deleteFamilyFromFS(hri, familyName, hasMob);
    }
  }

--- a/hbase-server/src/main/java/org/apache/hadoop/hbase/master/procedure/ModifyTableProcedure.java
+++ b/hbase-server/src/main/java/org/apache/hadoop/hbase/master/procedure/ModifyTableProcedure.java
@ -368,7 +368,8 @@ public class ModifyTableProcedure
          env,
          getTableName(),
          getRegionInfoList(env),
-          familyName);
+          familyName,
+          oldHTableDescriptor.getFamily(familyName).isMobEnabled());
      }
    }
  }
--- a/hbase-server/src/main/java/org/apache/hadoop/hbase/mob/MobUtils.java
+++ b/hbase-server/src/main/java/org/apache/hadoop/hbase/mob/MobUtils.java
@ -45,6 +45,7 @@ import org.apache.hadoop.fs.FileSystem;
 import org.apache.hadoop.fs.Path;
 import org.apache.hadoop.hbase.Cell;
 import org.apache.hadoop.hbase.CellComparator;
+import org.apache.hadoop.hbase.HBaseConfiguration;
 import org.apache.hadoop.hbase.HColumnDescriptor;
 import org.apache.hadoop.hbase.HConstants;
 import org.apache.hadoop.hbase.HRegionInfo;
@ -870,4 +871,27 @@ public class MobUtils {
      return false;
    }
  }
+  
+  /**
+   * Archive mob store files
+   * @param conf The current configuration.
+   * @param fs The current file system.
+   * @param mobRegionInfo The mob family region info.
+   * @param mobFamilyDir The mob family directory.
+   * @param family The name of the column family.
+   * @throws IOException
+   */
+  public static void archiveMobStoreFiles(Configuration conf, FileSystem fs,
+      HRegionInfo mobRegionInfo, Path mobFamilyDir, byte[] family) throws IOException {
+    // disable the block cache.
+    Configuration copyOfConf = HBaseConfiguration.create(conf);
+    copyOfConf.setFloat(HConstants.HFILE_BLOCK_CACHE_SIZE_KEY, 0f);
+    CacheConfig cacheConfig = new CacheConfig(copyOfConf);
+    FileStatus[] fileStatus = FSUtils.listStatus(fs, mobFamilyDir);
+    List<StoreFile> storeFileList = new ArrayList<StoreFile>();
+    for (FileStatus file : fileStatus) {
+      storeFileList.add(new StoreFile(fs, file.getPath(), conf, cacheConfig, BloomType.NONE));
+    }
+    HFileArchiver.archiveStoreFiles(conf, fs, mobRegionInfo, mobFamilyDir, family, storeFileList);
+  }
 }
--- a/hbase-server/src/main/java/org/apache/hadoop/hbase/util/HFileArchiveUtil.java
+++ b/hbase-server/src/main/java/org/apache/hadoop/hbase/util/HFileArchiveUtil.java
@ -64,10 +64,8 @@ public class HFileArchiveUtil {
                                         HRegionInfo region,
                                         Path tabledir,
      byte[] family) throws IOException {
-    TableName tableName =
-        FSUtils.getTableName(tabledir);
    Path rootDir = FSUtils.getRootDir(conf);
-    Path tableArchiveDir = getTableArchivePath(rootDir, tableName);
+    Path tableArchiveDir = getTableArchivePath(rootDir, region.getTable());
    return HStore.getStoreHomedir(tableArchiveDir, region, family);
  }

--- a/hbase-server/src/test/java/org/apache/hadoop/hbase/regionserver/TestDeleteMobTable.java
+++ b/hbase-server/src/test/java/org/apache/hadoop/hbase/regionserver/TestDeleteMobTable.java
@ -24,6 +24,7 @@ import org.apache.hadoop.fs.FileSystem;
 import org.apache.hadoop.fs.Path;
 import org.apache.hadoop.hbase.HBaseTestingUtility;
 import org.apache.hadoop.hbase.HColumnDescriptor;
+import org.apache.hadoop.hbase.HRegionInfo;
 import org.apache.hadoop.hbase.HTableDescriptor;
 import org.apache.hadoop.hbase.TableName;
 import org.apache.hadoop.hbase.client.*;
@ -161,6 +162,48 @@ public class TestDeleteMobTable {
      }
    }
  }
+  
+  @Test
+  public void testMobFamilyDelete() throws Exception {
+    byte[] tableName = Bytes.toBytes("testMobFamilyDelete");
+    TableName tn = TableName.valueOf(tableName);
+    HTableDescriptor htd = new HTableDescriptor(tn);
+    HColumnDescriptor hcd = new HColumnDescriptor(FAMILY);
+    hcd.setMobEnabled(true);
+    hcd.setMobThreshold(0);
+    htd.addFamily(hcd);
+    htd.addFamily(new HColumnDescriptor(Bytes.toBytes("family2")));
+    HBaseAdmin admin = null;
+    Table table = null;
+    try {
+      admin = TEST_UTIL.getHBaseAdmin();
+      admin.createTable(htd);
+      table = ConnectionFactory.createConnection(TEST_UTIL.getConfiguration()).getTable(tn);
+      byte[] value = generateMobValue(10);
+      byte[] row = Bytes.toBytes("row");
+      Put put = new Put(row);
+      put.addColumn(FAMILY, QF, EnvironmentEdgeManager.currentTime(), value);
+      table.put(put);
+      admin.flush(tn);
+      // the mob file exists
+      Assert.assertEquals(1, countMobFiles(tn, hcd.getNameAsString()));
+      Assert.assertEquals(0, countArchiveMobFiles(tn, hcd.getNameAsString()));
+      String fileName = assertHasOneMobRow(table, tn, hcd.getNameAsString());
+      Assert.assertFalse(mobArchiveExist(tn, hcd.getNameAsString(), fileName));
+      Assert.assertTrue(mobTableDirExist(tn));
+      admin.deleteColumnFamily(tn, FAMILY);
+      Assert.assertEquals(0, countMobFiles(tn, hcd.getNameAsString()));
+      Assert.assertEquals(1, countArchiveMobFiles(tn, hcd.getNameAsString()));
+      Assert.assertTrue(mobArchiveExist(tn, hcd.getNameAsString(), fileName));
+      Assert.assertFalse(mobColumnFamilyDirExist(tn));
+    } finally {
+      table.close();
+      if (admin != null) {
+        admin.close();
+      }
+      TEST_UTIL.deleteTable(tableName);
+    }
+  }

  private int countMobFiles(TableName tn, String familyName) throws IOException {
    FileSystem fs = TEST_UTIL.getTestFileSystem();
@ -189,6 +232,14 @@ public class TestDeleteMobTable {
    Path tableDir = FSUtils.getTableDir(MobUtils.getMobHome(TEST_UTIL.getConfiguration()), tn);
    return fs.exists(tableDir);
  }
+  
+  private boolean mobColumnFamilyDirExist(TableName tn) throws IOException {
+    FileSystem fs = TEST_UTIL.getTestFileSystem();
+    Path tableDir = FSUtils.getTableDir(MobUtils.getMobHome(TEST_UTIL.getConfiguration()), tn);
+    HRegionInfo mobRegionInfo = MobUtils.getMobRegionInfo(tn);
+    Path mobFamilyDir = new Path(tableDir, new Path(mobRegionInfo.getEncodedName(), Bytes.toString(FAMILY)));
+    return fs.exists(mobFamilyDir);
+  }

  private boolean mobArchiveExist(TableName tn, String familyName, String fileName)
      throws IOException {