三个节点 OS：64位RHEL5及以上或者64位CentOS6.0及以上 JVM：预装64位JDK 1.8及以上版本
准备三个节点的虚拟机
在各个节点执行以下操作来修改主机名，使集群下的主机有格式一个统一的主机名，以便后续的操作和维护。 修改主机名 vi /etc/sysconfig/network 192.168.xx.210 ha01 （其它俩台分别修改自己的ha02 ha03） 修改host映射：
vi /etc/hosts
192.168.xx.210 ha01 192.168.xx.220 ha02 192.168.xx.230 ha03service iptables stop chkconfig iptables off
配置时间同步
配置免密登陆
jdk安装步骤
安装zookeeper 安装hadoop
hadoop部分的配置分为两部分hdfs和yarn。
修改配置文件 修改core-site.xml（如果文件不存在，但是core-site.xml.template文件存在，则先修改文件名，执行mv core-site.xml.template core-site.xml）
vi /usr /local/hadoop-2.7.3/etc/hadoop/core-site.xml
修改为以下内容：
<configuration> <property> <name>fs.defaultFS</name> <value>hdfs://beh</value> <final>false</final> </property> <property> <name>hadoop.tmp.dir</name> <value>/usr/local/hadoopdata</value> <final>false</final> </property> <property> <name>ha.zookeeper.quorum</name> <value>ha01:2181,ha02:2181,ha03:2181</value> <final>false</final> </property> </configuration>修改hdfs-site.xml
vi /usr/local/hadoop-2.7.3/etc/hadoop/hdfs-site.xml
修改为以下内容：
<configuration> <property> <name>dfs.nameservices</name> <value>beh</value> <final>false</final> </property> <property> <name>dfs.ha.namenodes.beh</name> <value>nn1,nn2</value> <final>false</final> </property> <property> <name>dfs.namenode.rpc-address.beh.nn1</name> <value>ha01:9000</value> <final>false</final> </property> <property> <name>dfs.namenode.http-address.beh.nn1</name> <value>ha01:50070</value> <final>false</final> </property> <property> <name>dfs.namenode.rpc-address.beh.nn2</name> <value>ha02:9000</value> <final>false</final> </property> <property> <name>dfs.namenode.http-address.beh.nn2</name> <value>ha02:50070</value> <final>false</final> </property> <property> <name>dfs.namenode.shared.edits.dir</name> <value>qjournal://ha01:8485;ha02:8485;ha03:8485/beh</value> <final>false</final> </property> <property> <name>dfs.ha.automatic-failover.enabled.beh</name> <value>true</value> <final>false</final> </property> <property> <name>dfs.client.failover.proxy.provider.beh</name> <value>org.apache.hadoop.hdfs.server.namenode.ha.ConfiguredFailoverProxyProvider</value> <final>false</final> </property> <property> <name>dfs.journalnode.edits .dir</name> <value>/usr/local/metadata/journal</value> <final>false</final> </property> <property> <name>dfs.ha.fencing.methods</name> <value>sshfence shell(/bin/true) </value> <final>false</final> </property> <property> <name>dfs.ha.fencing.ssh.private-key-files</name> <value>/usr/local/.ssh/id_rsa</value> <final>true</final> </property> <property> <name>dfs.replication</name> <value>2</value> <final>false</final> </property> <configuration>修改slaves
vi /usr/local/hadoop-2.7.3/etc/hadoop/slaves
修改为以下内容：
ha02 ha03修改mapred-site.xml
vi /usr/local/hadoop2.7.3/etc/hadoop/mapred-site.xml
修改为以下内容：
<configuration> <property> <name>mapreduce.framework.name</name> <value>yarn</value> </property> <property> <name>mapreduce.jobhistory.address</name> <value>ha02:10020</value> </property> <property> <name>mapreduce.jobhistory.webapp.address</name> <value>ha03:19888</value> </property> <property> <name>yarn.app.mapreduce.am.staging-dir</name> <value>/usr/local/metadata/hadoop-yarn/staging</value> </property> </configuration>修改yarn-site.xml
vi /usr/local/hadoop2.7.3/etc/hadoop/yarn-site.xml
修改为以下内容：
<configuration> <property> <name>yarn.nodemanager.aux-services</name> <value>mapreduce_shuffle</value> </property> <property> <name>yarn.nodemanager.aux-services.mapreduce.shuffle.class</name> <value>org.apache.hadoop.mapred.ShuffleHandler</value> </property> <property> <name>yarn.nodemanager.local-dirs</name> <value>/usr/local/metadata/yarn</value> </property> <property> <name>yarn.nodemanager.log-dirs</name> <value>/usr/local/logs/yarn/userlogs</value> </property> <property> <name>yarn.log-aggregation-enable</name> <value>true</value> </property> <property> <description>Where to aggregate logs</description> <name>yarn.nodemanager.remote-app-log-dir</name> <value>hdfs://beh/var/log/hadoop-yarn/apps</value> </property> <!-- Resource Manager Configs --> <property> <name>yarn.resourcemanager.connect.retry-interval.ms</name> <value>2000</value> </property> <property> <name>yarn.resourcemanager.ha.enabled</name> <value>true</value> </property> <property> <name>yarn.resourcemanager.ha.automatic-failover.enabled</name> <value>true</value> </property> <property> <name>yarn.resourcemanager.cluster-id</name> <value>beh</value> </property> <property> <name>yarn.resourcemanager.ha.rm-ids</name> <value>rm1,rm2</value> </property> <!--RM1 RM2 is different--> <property> <name>yarn.resourcemanager.ha.id</name> <value>rm1</value> </property> <property> <name>yarn.resourcemanager.scheduler.class</name> <value>org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.FairScheduler</value> </property> <property> <name>yarn.resourcemanager.recovery.enabled</name> <value>true</value> </property> <property> <name>yarn.resourcemanager.store.class</name> <value>org.apache.hadoop.yarn.server.resourcemanager.recovery.ZKRMStateStore</value> </property> <property> <name>yarn.resourcemanager.zk.state-store.address</name> <value>ha01:2181,ha02:2181,ha03:2181</value> </property> <property> <name>yarn.app.mapreduce.am.scheduler.connection.wait.interval-ms</name> <value>5000</value> </property> <!-- RM1 configs --> <property> <name>yarn.resourcemanager.address.rm1</name> <value>ha01:23140</value> </property> <property> <name>yarn.resourcemanager.scheduler.address.rm1</name> <value>ha01:23130</value> </property> <property> <name>yarn.resourcemanager.webapp.https.address.rm1</name> <value>ha01:23189</value> </property> <property> <name>yarn.resourcemanager.webapp.address.rm1</name> <value>ha01:23188</value> </property> <property> <name>yarn.resourcemanager.resource-tracker.address.rm1</name> <value>ha01:23125</value> </property> <property> <name>yarn.resourcemanager.admin.address.rm1</name> <value>ha01:23141</value> </property> <!-- RM2 configs --> <property> <name>yarn.resourcemanager.address.rm2</name> <value>ha02:23140</value> </property> <property> <name>yarn.resourcemanager.scheduler.address.rm2</name> <value>ha02:23130</value> </property> <property> <name>yarn.resourcemanager.webapp.https.address.rm2</name> <value>ha02:23189</value> </property> <property> <name>yarn.resourcemanager.webapp.address.rm2</name> <value>ha02:23188</value> </property> <property> <name>yarn.resourcemanager.resource-tracker.address.rm2</name> <value>ha02:23125</value> </property> <property> <name>yarn.resourcemanager.admin.address.rm2</name> <value>ha02:23141</value> </property> <!-- Node Manager Configs --> <property> <name>mapreduce.shuffle.port</name> <value>23080</value> </property> <property> <name>yarn.resourcemanager.zk-address</name> <value>ha01:2181,ha02:2181,ha03:2181</value> </property> </configuration>修改环境变量
vim /usr/local/hadoop-2.7.3/etc/hadoop/hadoop-env.sh vim /usr/local/hadoop-2.7.3/etc/hadoop/yarn-env.sh
修改为以下内容：
export JAVA_HOME=/usr/local/jdk1.8.0_102scp -r /usr/local/hadoop2.7.3 ha02:/usr/local scp -r /usr/local/hadoop2.7.3 ha03:/usr/local
注：将以上配置复制到所有节点
启动journalnode（进程名：JournalNode）
sbin/hadoop-daemon.sh start journalnode
格式化zookeeper，在ha01上执行
hdfs zkfc -formatZK
对ha01节点进行格式化和启动启动namenode（进程名：NameNode）：
hdfs namenode -format sbin/hadoop-daemon.sh start namenode
对ha02节点进行格式化和启动
hdfs namenode -bootstrapStandby sbin/hadoop-daemon.sh start namenode
在ha01和ha02上启动zkfc服务（zkfc服务进程名：DFSZKFailoverController）：此时ha01和ha02就会有一个节点变为active状态
sbin/hadoop-daemon.sh start zkfc
启动datanode（进程名：DataNode）：在ha01上执行
sbin/hadoop-daemons.sh start datanode
打开浏览器，访问 hadoop1:50070 以及 hadoop2:50070，你将会看到两个namenode一个是active而另一个是standby。 然后kill掉其中active的namenode进程，另一个standby的naemnode将会自动转换为active状态 hadoop01:50070或hadoop01的ip:50070 hadoop02:50070或hadoop02的ip:50070
