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.replication.regionserver; 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.fs.Path; 025import org.apache.hadoop.hbase.Abortable; 026import org.apache.hadoop.hbase.ChoreService; 027import org.apache.hadoop.hbase.CoordinatedStateManager; 028import org.apache.hadoop.hbase.HBaseConfiguration; 029import org.apache.hadoop.hbase.HConstants; 030import org.apache.hadoop.hbase.Server; 031import org.apache.hadoop.hbase.ServerName; 032import org.apache.hadoop.hbase.client.ClusterConnection; 033import org.apache.hadoop.hbase.client.Connection; 034import org.apache.hadoop.hbase.util.FSUtils; 035import org.apache.hadoop.hbase.zookeeper.MetaTableLocator; 036import org.apache.hadoop.hbase.zookeeper.ZKWatcher; 037import org.apache.hadoop.util.Tool; 038import org.apache.hadoop.util.ToolRunner; 039import org.apache.yetus.audience.InterfaceAudience; 040 041/** 042 * In a scenario of Replication based Disaster/Recovery, when hbase Master-Cluster crashes, this 043 * tool is used to sync-up the delta from Master to Slave using the info from ZooKeeper. The tool 044 * will run on Master-Cluser, and assume ZK, Filesystem and NetWork still available after hbase 045 * crashes 046 * 047 * <pre> 048 * hbase org.apache.hadoop.hbase.replication.regionserver.ReplicationSyncUp 049 * </pre> 050 */ 051@InterfaceAudience.Private 052public class ReplicationSyncUp extends Configured implements Tool { 053 054 private static final long SLEEP_TIME = 10000; 055 056 /** 057 * Main program 058 */ 059 public static void main(String[] args) throws Exception { 060 int ret = ToolRunner.run(HBaseConfiguration.create(), new ReplicationSyncUp(), args); 061 System.exit(ret); 062 } 063 064 @Override 065 public int run(String[] args) throws Exception { 066 Abortable abortable = new Abortable() { 067 @Override 068 public void abort(String why, Throwable e) { 069 } 070 071 @Override 072 public boolean isAborted() { 073 return false; 074 } 075 }; 076 Configuration conf = getConf(); 077 try (ZKWatcher zkw = 078 new ZKWatcher(conf, "syncupReplication" + System.currentTimeMillis(), abortable, true)) { 079 Path walRootDir = FSUtils.getWALRootDir(conf); 080 FileSystem fs = FSUtils.getWALFileSystem(conf); 081 Path oldLogDir = new Path(walRootDir, HConstants.HREGION_OLDLOGDIR_NAME); 082 Path logDir = new Path(walRootDir, HConstants.HREGION_LOGDIR_NAME); 083 084 System.out.println("Start Replication Server start"); 085 Replication replication = new Replication(); 086 replication.initialize(new DummyServer(zkw), fs, logDir, oldLogDir, null); 087 ReplicationSourceManager manager = replication.getReplicationManager(); 088 manager.init().get(); 089 while (manager.activeFailoverTaskCount() > 0) { 090 Thread.sleep(SLEEP_TIME); 091 } 092 while (manager.getOldSources().size() > 0) { 093 Thread.sleep(SLEEP_TIME); 094 } 095 manager.join(); 096 } catch (InterruptedException e) { 097 System.err.println("didn't wait long enough:" + e); 098 return -1; 099 } 100 return 0; 101 } 102 103 class DummyServer implements Server { 104 String hostname; 105 ZKWatcher zkw; 106 107 DummyServer(ZKWatcher zkw) { 108 // a unique name in case the first run fails 109 hostname = System.currentTimeMillis() + ".SyncUpTool.replication.org"; 110 this.zkw = zkw; 111 } 112 113 DummyServer(String hostname) { 114 this.hostname = hostname; 115 } 116 117 @Override 118 public Configuration getConfiguration() { 119 return getConf(); 120 } 121 122 @Override 123 public ZKWatcher getZooKeeper() { 124 return zkw; 125 } 126 127 @Override 128 public CoordinatedStateManager getCoordinatedStateManager() { 129 return null; 130 } 131 132 @Override 133 public MetaTableLocator getMetaTableLocator() { 134 return null; 135 } 136 137 @Override 138 public ServerName getServerName() { 139 return ServerName.valueOf(hostname, 1234, 1L); 140 } 141 142 @Override 143 public void abort(String why, Throwable e) { 144 } 145 146 @Override 147 public boolean isAborted() { 148 return false; 149 } 150 151 @Override 152 public void stop(String why) { 153 } 154 155 @Override 156 public boolean isStopped() { 157 return false; 158 } 159 160 @Override 161 public ClusterConnection getConnection() { 162 return null; 163 } 164 165 @Override 166 public ChoreService getChoreService() { 167 return null; 168 } 169 170 @Override 171 public ClusterConnection getClusterConnection() { 172 return null; 173 } 174 175 @Override 176 public FileSystem getFileSystem() { 177 return null; 178 } 179 180 @Override 181 public boolean isStopping() { 182 return false; 183 } 184 185 @Override 186 public Connection createConnection(Configuration conf) throws IOException { 187 return null; 188 } 189 } 190}