001/*
002 * Licensed to the Apache Software Foundation (ASF) under one
003 * or more contributor license agreements.  See the NOTICE file
004 * distributed with this work for additional information
005 * regarding copyright ownership.  The ASF licenses this file
006 * to you under the Apache License, Version 2.0 (the
007 * "License"); you may not use this file except in compliance
008 * with the License.  You may obtain a copy of the License at
009 *
010 *     http://www.apache.org/licenses/LICENSE-2.0
011 *
012 * Unless required by applicable law or agreed to in writing, software
013 * distributed under the License is distributed on an "AS IS" BASIS,
014 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
015 * See the License for the specific language governing permissions and
016 * limitations under the License.
017 */
018package org.apache.hadoop.hbase.mob;
019
020import java.io.IOException;
021import org.apache.hadoop.conf.Configuration;
022import org.apache.hadoop.conf.Configured;
023import org.apache.hadoop.fs.FileSystem;
024import org.apache.hadoop.hbase.HBaseConfiguration;
025import org.apache.hadoop.hbase.HConstants;
026import org.apache.hadoop.hbase.TableName;
027import org.apache.hadoop.hbase.client.Admin;
028import org.apache.hadoop.hbase.client.ColumnFamilyDescriptor;
029import org.apache.hadoop.hbase.client.Connection;
030import org.apache.hadoop.hbase.client.ConnectionFactory;
031import org.apache.hadoop.hbase.client.TableDescriptor;
032import org.apache.hadoop.hbase.io.hfile.CacheConfig;
033import org.apache.hadoop.hbase.util.Bytes;
034import org.apache.hadoop.hbase.util.EnvironmentEdgeManager;
035import org.apache.hadoop.util.Tool;
036import org.apache.hadoop.util.ToolRunner;
037import org.apache.yetus.audience.InterfaceAudience;
038import org.slf4j.Logger;
039import org.slf4j.LoggerFactory;
040
041/**
042 * The cleaner to delete the expired MOB files.
043 */
044@InterfaceAudience.Private
045public class ExpiredMobFileCleaner extends Configured implements Tool {
046
047  private static final Logger LOG = LoggerFactory.getLogger(ExpiredMobFileCleaner.class);
048
049  /**
050   * Cleans the MOB files when they're expired and their min versions are 0. If the latest timestamp
051   * of Cells in a MOB file is older than the TTL in the column family, it's regarded as expired.
052   * This cleaner deletes them. At a time T0, the cells in a mob file M0 are expired. If a user
053   * starts a scan before T0, those mob cells are visible, this scan still runs after T0. At that
054   * time T1, this mob file M0 is expired, meanwhile a cleaner starts, the M0 is archived and can be
055   * read in the archive directory.
056   * @param tableName The current table name.
057   * @param family    The current family.
058   */
059  public void cleanExpiredMobFiles(String tableName, ColumnFamilyDescriptor family)
060    throws IOException {
061    Configuration conf = getConf();
062    TableName tn = TableName.valueOf(tableName);
063    FileSystem fs = FileSystem.get(conf);
064    LOG.info("Cleaning the expired MOB files of " + family.getNameAsString() + " in " + tableName);
065    // disable the block cache.
066    Configuration copyOfConf = new Configuration(conf);
067    copyOfConf.setFloat(HConstants.HFILE_BLOCK_CACHE_SIZE_KEY, 0f);
068    CacheConfig cacheConfig = new CacheConfig(copyOfConf);
069    MobUtils.cleanExpiredMobFiles(fs, conf, tn, family, cacheConfig,
070      EnvironmentEdgeManager.currentTime());
071  }
072
073  public static void main(String[] args) throws Exception {
074    Configuration conf = HBaseConfiguration.create();
075    ToolRunner.run(conf, new ExpiredMobFileCleaner(), args);
076  }
077
078  private void printUsage() {
079    System.err.println("Usage:\n" + "--------------------------\n"
080      + ExpiredMobFileCleaner.class.getName() + " tableName familyName");
081    System.err.println(" tableName        The table name");
082    System.err.println(" familyName       The column family name");
083  }
084
085  @edu.umd.cs.findbugs.annotations.SuppressWarnings(value = "REC_CATCH_EXCEPTION",
086      justification = "Intentional")
087  @Override
088  public int run(String[] args) throws Exception {
089    if (args.length != 2) {
090      printUsage();
091      return 1;
092    }
093    String tableName = args[0];
094    String familyName = args[1];
095    TableName tn = TableName.valueOf(tableName);
096    Connection connection = ConnectionFactory.createConnection(getConf());
097    Admin admin = connection.getAdmin();
098    try {
099      TableDescriptor htd = admin.getDescriptor(tn);
100      ColumnFamilyDescriptor family = htd.getColumnFamily(Bytes.toBytes(familyName));
101      if (family == null || !family.isMobEnabled()) {
102        throw new IOException("Column family " + familyName + " is not a MOB column family");
103      }
104      if (family.getMinVersions() > 0) {
105        throw new IOException(
106          "The minVersions of the column family is not 0, could not be handled by this cleaner");
107      }
108      cleanExpiredMobFiles(tableName, family);
109      return 0;
110    } finally {
111      admin.close();
112      try {
113        connection.close();
114      } catch (IOException e) {
115        LOG.error("Failed to close the connection.", e);
116      }
117    }
118  }
119}