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 */ 018 019package org.apache.hadoop.hbase.chaos.actions; 020 021import java.io.IOException; 022import java.util.Collection; 023import java.util.Collections; 024import java.util.EnumSet; 025import java.util.List; 026import org.apache.commons.lang3.RandomUtils; 027import org.apache.hadoop.hbase.ClusterMetrics.Option; 028import org.apache.hadoop.hbase.HRegionInfo; 029import org.apache.hadoop.hbase.ServerName; 030import org.apache.hadoop.hbase.TableName; 031import org.apache.hadoop.hbase.chaos.factories.MonkeyConstants; 032import org.apache.hadoop.hbase.client.Admin; 033import org.apache.hadoop.hbase.client.RegionInfo; 034import org.apache.hadoop.hbase.util.Bytes; 035 036/** 037* Action that tries to move every region of a table. 038*/ 039public class MoveRegionsOfTableAction extends Action { 040 private final long sleepTime; 041 private final TableName tableName; 042 private final long maxTime; 043 044 public MoveRegionsOfTableAction(TableName tableName) { 045 this(-1, MonkeyConstants.DEFAULT_MOVE_REGIONS_MAX_TIME, tableName); 046 } 047 048 public MoveRegionsOfTableAction(long sleepTime, long maxSleepTime, TableName tableName) { 049 this.sleepTime = sleepTime; 050 this.tableName = tableName; 051 this.maxTime = maxSleepTime; 052 } 053 054 @Override 055 public void perform() throws Exception { 056 if (sleepTime > 0) { 057 Thread.sleep(sleepTime); 058 } 059 060 Admin admin = this.context.getHBaseIntegrationTestingUtility().getAdmin(); 061 ServerName[] servers = getServers(admin); 062 063 LOG.info("Performing action: Move regions of table {}", tableName); 064 List<HRegionInfo> regions = admin.getTableRegions(tableName); 065 if (regions == null || regions.isEmpty()) { 066 LOG.info("Table {} doesn't have regions to move", tableName); 067 return; 068 } 069 070 Collections.shuffle(regions); 071 072 long start = System.currentTimeMillis(); 073 for (HRegionInfo regionInfo:regions) { 074 075 // Don't try the move if we're stopping 076 if (context.isStopping()) { 077 return; 078 } 079 080 moveRegion(admin, servers, regionInfo); 081 if (sleepTime > 0) { 082 Thread.sleep(sleepTime); 083 } 084 085 // put a limit on max num regions. Otherwise, this won't finish 086 // with a sleep time of 10sec, 100 regions will finish in 16min 087 if (System.currentTimeMillis() - start > maxTime) { 088 break; 089 } 090 } 091 } 092 093 static ServerName [] getServers(Admin admin) throws IOException { 094 Collection<ServerName> serversList = 095 admin.getClusterMetrics(EnumSet.of(Option.LIVE_SERVERS)).getLiveServerMetrics().keySet(); 096 return serversList.toArray(new ServerName[serversList.size()]); 097 } 098 099 static void moveRegion(Admin admin, ServerName [] servers, RegionInfo regionInfo) { 100 try { 101 String destServerName = servers[RandomUtils.nextInt(0, servers.length)].getServerName(); 102 LOG.debug("Moving {} to {}", regionInfo.getRegionNameAsString(), destServerName); 103 admin.move(regionInfo.getEncodedNameAsBytes(), Bytes.toBytes(destServerName)); 104 } catch (Exception ex) { 105 LOG.warn("Move failed, might be caused by other chaos: {}", ex.getMessage()); 106 } 107 } 108}