Hadoop2 namenode HA+联邦+Resource Manager HA实验分析

本篇内容介绍了"Hadoop2 namenode HA+联邦+Resource Manager HA实验分析"的有关知识，在实际案例的操作过程中，不少人都会遇到这样的困境，接下来就让小编带领大家学习一下如何处理这些情况吧！希望大家仔细阅读，能够学有所成！

实验的Hadoop版本为2.5.2，硬件环境是5台虚拟机，使用的均是CentOS6.6操作系统，虚拟机IP和hostname分别为：
192.168.63.171 node1.zhch
192.168.63.172 node2.zhch
192.168.63.173 node3.zhch
192.168.63.174 node4.zhch
192.168.63.175 node5.zhch

ssh免密码、防火墙、JDK这里就不在赘述了。虚拟机的角色分配是：
node1为主namenode1、主resource manager、zookeeper、journalnode
node2为备namendoe1、zookeeper、journalnode
node3为主namenode2、备resource manager、zookeeper、journalnode、datanode
node4为备namenode2、datanode
node5为datanode

步骤和 Namenode HA的安装配置基本相同，需要先 安装zookeeper集群，主要的不同在于core-site.xml、hdfs-site.xml、yarn-site.xml配置文件，其余文件的配置和Namenode HA安装配置基本一致。

一、配置Hadoop

## 解压[yyl@node1 program]$ tar -zxf hadoop-2.5.2.tar.gz## 创建文件夹[yyl@node1 program]$ mkdir hadoop-2.5.2/name[yyl@node1 program]$ mkdir hadoop-2.5.2/data[yyl@node1 program]$ mkdir hadoop-2.5.2/journal[yyl@node1 program]$ mkdir hadoop-2.5.2/tmp## 配置hadoop-env.sh[yyl@node1 program]$ cd hadoop-2.5.2/etc/hadoop/[yyl@node1 hadoop]$ vim hadoop-env.shexport JAVA_HOME=/usr/lib/java/jdk1.7.0_80## 配置yarn-env.sh[yyl@node1 hadoop]$ vim yarn-env.shexport JAVA_HOME=/usr/lib/java/jdk1.7.0_80## 配置slaves[yyl@node1 hadoop]$ vim slavesnode3.zhchnode4.zhchnode5.zhch## 配置mapred-site.xml[yyl@node1 hadoop]$ cp mapred-site.xml.template mapred-site.xml[yyl@node1 hadoop]$ vim mapred-site.xml  mapreduce.framework.name  yarn   mapreduce.jobhistory.address   node2.zhch:10020     mapreduce.jobhistory.webapp.address   node2.zhch:19888 ## 配置core-site.xml[yyl@node1 hadoop]$ vim core-site.xml  fs.defaultFS  hdfs://mycluster  io.file.buffer.size  131072  hadoop.tmp.dir  file:/home/yyl/program/hadoop-2.5.2/tmp  hadoop.proxyuser.hduser.hosts  *  hadoop.proxyuser.hduser.groups  *  ha.zookeeper.quorum  node1.zhch:2181,node2.zhch:2181,node3.zhch:2181  ha.zookeeper.session-timeout.ms  1000## 配置hdfs-site.xml[yyl@node1 hadoop]$ vim hdfs-site.xml  dfs.namenode.name.dir  file:/home/yyl/program/hadoop-2.5.2/name  dfs.datanode.data.dir  file:/home/yyl/program/hadoop-2.5.2/data  dfs.replication  1  dfs.webhdfs.enabled  true  dfs.permissions  false  dfs.permissions.enabled  false  dfs.nameservices  mycluster,yourcluster  dfs.ha.namenodes.mycluster  nn1,nn2  dfs.namenode.rpc-address.mycluster.nn1  node1.zhch:9000  dfs.namenode.rpc-address.mycluster.nn2  node2.zhch:9000  dfs.namenode.servicerpc-address.mycluster.nn1  node1.zhch:53310  dfs.namenode.servicerpc-address.mycluster.nn2  node2.zhch:53310  dfs.namenode.http-address.mycluster.nn1  node1.zhch:50070  dfs.namenode.http-address.mycluster.nn2  node2.zhch:50070  dfs.ha.namenodes.yourcluster  nn1,nn2  dfs.namenode.rpc-address.yourcluster.nn1  node3.zhch:9000  dfs.namenode.rpc-address.yourcluster.nn2  node4.zhch:9000  dfs.namenode.servicerpc-address.yourcluster.nn1  node3.zhch:53310  dfs.namenode.servicerpc-address.yourcluster.nn2  node4.zhch:53310  dfs.namenode.http-address.yourcluster.nn1  node3.zhch:50070  dfs.namenode.http-address.yourcluster.nn2  node4.zhch:50070  dfs.namenode.shared.edits.dir  qjournal://node1.zhch:8485;node2.zhch:8485;node3.zhch:8485/mycluster  dfs.client.failover.proxy.provider.mycluster  org.apache.hadoop.hdfs.server.namenode.ha.ConfiguredFailoverProxyProvider  dfs.client.failover.proxy.provider.yourcluster  org.apache.hadoop.hdfs.server.namenode.ha.ConfiguredFailoverProxyProvider  dfs.ha.fencing.methods  sshfence  dfs.ha.fencing.ssh.private-key-files  /home/yyl/.ssh/id_rsa  dfs.ha.fencing.ssh.connect-timeout  30000  dfs.journalnode.edits.dir  /home/yyl/program/hadoop-2.5.2/journal  dfs.ha.automatic-failover.enabled.mycluster  true  dfs.ha.automatic-failover.enabled.yourcluster  true  ha.failover-controller.cli-check.rpc-timeout.ms  60000  ipc.client.connect.timeout  60000  dfs.image.transfer.bandwidthPerSec  4194304## 配置yarn-site.xml[yyl@node1 hadoop]$ vim yarn-site.xml  yarn.nodemanager.aux-services  mapreduce_shuffle  yarn.nodemanager.aux-services.mapreduce.shuffle.class  org.apache.hadoop.mapred.ShuffleHandler  yarn.resourcemanager.connect.retry-interval.ms  2000  yarn.resourcemanager.ha.enabled  true  yarn.resourcemanager.ha.automatic-failover.enabled  true  yarn.resourcemanager.ha.automatic-failover.embedded  true  yarn.resourcemanager.cluster-id  yarn-cluster  yarn.resourcemanager.ha.rm-ids  rm1,rm2  yarn.resourcemanager.ha.id  rm1  yarn.resourcemanager.scheduler.class  org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.FairScheduler  yarn.resourcemanager.recovery.enabled  true  yarn.app.mapreduce.am.scheduler.connection.wait.interval-ms  5000  yarn.resourcemanager.store.class  org.apache.hadoop.yarn.server.resourcemanager.recovery.ZKRMStateStore  yarn.resourcemanager.zk-address  node1.zhch:2181,node2.zhch:2181,node3.zhch:2181  yarn.resourcemanager.zk.state-store.address  node1.zhch:2181,node2.zhch:2181,node3.zhch:2181  yarn.resourcemanager.address.rm1  node1.zhch:23140  yarn.resourcemanager.address.rm2  node3.zhch:23140  yarn.resourcemanager.scheduler.address.rm1  node1.zhch:23130  yarn.resourcemanager.scheduler.address.rm2  node3.zhch:23130  yarn.resourcemanager.admin.address.rm1  node1.zhch:23141  yarn.resourcemanager.admin.address.rm2  node3.zhch:23141  yarn.resourcemanager.resource-tracker.address.rm1  node1.zhch:23125  yarn.resourcemanager.resource-tracker.address.rm2  node3.zhch:23125  yarn.resourcemanager.webapp.address.rm1  node1.zhch:23188  yarn.resourcemanager.webapp.address.rm2  node3.zhch:23188  yarn.resourcemanager.webapp.https.address.rm1  node1.zhch:23189  yarn.resourcemanager.webapp.https.address.rm2  node3.zhch:23189## 分发到各个节点[yyl@node1 hadoop]$ cd /home/yyl/program/[yyl@node1 program]$ scp -rp hadoop-2.5.2 yyl@node2.zhch:/home/yyl/program/[yyl@node1 program]$ scp -rp hadoop-2.5.2 yyl@node3.zhch:/home/yyl/program/[yyl@node1 program]$ scp -rp hadoop-2.5.2 yyl@node4.zhch:/home/yyl/program/[yyl@node1 program]$ scp -rp hadoop-2.5.2 yyl@node5.zhch:/home/yyl/program/## 修改主namenode2(node3.zhch)和备namenode2(node4.zhch)的 hdfs-site.xml 配置文件中 dfs.namenode.shared.edits.dir 的值为 qjournal://node1.zhch:8485;node2.zhch:8485;node3.zhch:8485/yourcluster ，其余属性值不变。## 修改备resource manager(node3.zhch)的 yarn-site.xml 配置文件中 yarn.resourcemanager.ha.id 的值为 rm2 ，其余属性值不变。## 在各个节点上设置hadoop环境变量[yyl@node1 ~]$ vim .bash_profile export HADOOP_PREFIX=/home/yyl/program/hadoop-2.5.2export HADOOP_COMMON_HOME=$HADOOP_PREFIXexport HADOOP_HDFS_HOME=$HADOOP_PREFIXexport HADOOP_MAPRED_HOME=$HADOOP_PREFIXexport HADOOP_YARN_HOME=$HADOOP_PREFIXexport HADOOP_CONF_DIR=$HADOOP_PREFIX/etc/hadoopexport PATH=$PATH:$HADOOP_PREFIX/bin:$HADOOP_PREFIX/sbin

二、格式化与启动

## 启动Zookeeper集群## 在主namenode1(node1.zhch)、主namenode2(node3.zhch)上执行命令： $HADOOP_HOME/bin/hdfs zkfc -formatZK[yyl@node1 ~]$ hdfs zkfc -formatZK[yyl@node3 ~]$ hdfs zkfc -formatZK[yyl@node2 ~]$ zkCli.sh[zk: localhost:2181(CONNECTED) 0] ls /[hadoop-ha, zookeeper][zk: localhost:2181(CONNECTED) 1] ls /hadoop-ha[mycluster, yourcluster]## 在node1.zhch node2.zhch node3.zhch上启动journalnode：[yyl@node1 ~]$ hadoop-daemon.sh start journalnodestarting journalnode, logging to /home/yyl/program/hadoop-2.5.2/logs/hadoop-yyl-journalnode-node1.zhch.out[yyl@node1 ~]$ jps1985 QuorumPeerMain2222 Jps2176 JournalNode[yyl@node2 ~]$ hadoop-daemon.sh start journalnodestarting journalnode, logging to /home/yyl/program/hadoop-2.5.2/logs/hadoop-yyl-journalnode-node2.zhch.out[yyl@node2 ~]$ jps1783 Jps1737 JournalNode1638 QuorumPeerMain[yyl@node3 ~]$ hadoop-daemon.sh start journalnodestarting journalnode, logging to /home/yyl/program/hadoop-2.5.2/logs/hadoop-yyl-journalnode-node3.zhch.out[yyl@node3 ~]$ jps1658 JournalNode1495 QuorumPeerMain1704 Jps## 在主namenode1(node1.zhch)上格式化namenode[yyl@node1 ~]$ hdfs namenode -format -clusterId c1## 在主namenode1(node1.zhch)上启动namenode进程[yyl@node1 ~]$ hadoop-daemon.sh start namenodestarting namenode, logging to /home/yyl/program/hadoop-2.5.2/logs/hadoop-yyl-namenode-node1.zhch.out[yyl@node1 ~]$ jps2286 NameNode1985 QuorumPeerMain2369 Jps2176 JournalNode## 在备namenode1(node2.zhch)上同步元数据[yyl@node2 ~]$ hdfs namenode -bootstrapStandby## 在备namenode1(node2.zhch)上启动namenode进程[yyl@node2 ~]$ hadoop-daemon.sh start namenodestarting namenode, logging to /home/yyl/program/hadoop-2.5.2/logs/hadoop-yyl-namenode-node2.zhch.out[yyl@node2 ~]$ jps1923 Jps1737 JournalNode1638 QuorumPeerMain1840 NameNode## 在主namenode2(node3.zhch)上格式化namenode[yyl@node3 ~]$ hdfs namenode -format -clusterId c1## 在主namenode2(node3.zhch)上启动namenode进程[yyl@node3 ~]$ hadoop-daemon.sh start namenodestarting namenode, logging to /home/yyl/program/hadoop-2.5.2/logs/hadoop-yyl-namenode-node3.zhch.out[yyl@node3 ~]$ jps1658 JournalNode1495 QuorumPeerMain1767 NameNode1850 Jps## 在备namenode2(node4.zhch)上同步元数据[yyl@node4 ~]$ hdfs namenode -bootstrapStandby## 在备namenode2(node4.zhch)上启动namenode进程[yyl@node4 ~]$ hadoop-daemon.sh start namenodestarting namenode, logging to /home/yyl/program/hadoop-2.5.2/logs/hadoop-yyl-namenode-node4.zhch.out[yyl@node4 ~]$ jps1602 Jps1519 NameNode## 在所有的namenode上启动ZooKeeperFailoverController[yyl@node1 ~]$ hadoop-daemon.sh start zkfcstarting zkfc, logging to /home/yyl/program/hadoop-2.5.2/logs/hadoop-yyl-zkfc-node1.zhch.out[yyl@node2 ~]$ hadoop-daemon.sh start zkfcstarting zkfc, logging to /home/yyl/program/hadoop-2.5.2/logs/hadoop-yyl-zkfc-node2.zhch.out[yyl@node3 ~]$ hadoop-daemon.sh start zkfcstarting zkfc, logging to /home/yyl/program/hadoop-2.5.2/logs/hadoop-yyl-zkfc-node3.zhch.out[yyl@node4 ~]$ hadoop-daemon.sh start zkfcstarting zkfc, logging to /home/yyl/program/hadoop-2.5.2/logs/hadoop-yyl-zkfc-node4.zhch.out## 启动DataNode[yyl@node1 ~]$ hadoop-daemons.sh start datanodenode4.zhch: starting datanode, logging to /home/yyl/program/hadoop-2.5.2/logs/hadoop-yyl-datanode-node4.zhch.outnode5.zhch: starting datanode, logging to /home/yyl/program/hadoop-2.5.2/logs/hadoop-yyl-datanode-node5.zhch.outnode3.zhch: starting datanode, logging to /home/yyl/program/hadoop-2.5.2/logs/hadoop-yyl-datanode-node3.zhch.out## 启动Yarn[yyl@node1 ~]$ start-yarn.shstarting yarn daemonsstarting resourcemanager, logging to /home/yyl/program/hadoop-2.5.2/logs/yarn-yyl-resourcemanager-node1.zhch.outnode3.zhch: starting nodemanager, logging to /home/yyl/program/hadoop-2.5.2/logs/yarn-yyl-nodemanager-node3.zhch.outnode4.zhch: starting nodemanager, logging to /home/yyl/program/hadoop-2.5.2/logs/yarn-yyl-nodemanager-node4.zhch.outnode5.zhch: starting nodemanager, logging to /home/yyl/program/hadoop-2.5.2/logs/yarn-yyl-nodemanager-node5.zhch.out## 在备resource manager(node3.zhch)上启动resource manager[yyl@node3 ~]$ yarn-daemon.sh start resourcemanagerstarting resourcemanager, logging to /home/yyl/program/hadoop-2.5.2/logs/yarn-yyl-resourcemanager-node3.zhch.out## 查看resource manager状态[yyl@node1 ~]$ yarn rmadmin -getServiceState rm1active[yyl@node1 ~]$ yarn rmadmin -getServiceState rm2standby

三、验证

开两个终端，都连接到主resource manager，在终端A中运行jps命令查看resource manager进程ID，在终端B中运行MapReduce程序；然后再到终端A中kill掉resource manager进程；最后观察在主resource manager进程挂掉后，MapReduce任务是否还能正常执行完毕。

"Hadoop2 namenode HA+联邦+Resource Manager HA实验分析"的内容就介绍到这里了，感谢大家的阅读。如果想了解更多行业相关的知识可以关注网站，小编将为大家输出更多高质量的实用文章！