YARN的HA搭建 | 每天，遇到更好的你

1 2	选择192.168.230.10(node1),192.168.230.14(node5)两台主机作为ResourceManager datanode与nodemanager一定要放在一个机器上

YARN的体系架构

1.配置/opt/modules/hadoop-2.5.1/etc/hadoop/mapred-site.xml:

<configuration>
	   <property>
	      <name>mapreduce.framework.name</name>
	      <value>yarn</value>
	  </property>
</configuration>

2.配置/opt/modules/hadoop-2.5.1/etc/hadoop/yarn-site.xml:

<configuration>
	  <property>
	      <name>yarn.nodemanager.aux-services</name>
	      <value>mapreduce_shuffle</value>
	  </property>
</configuration>

<!-- 启用 -->
<property>
  <name>yarn.resourcemanager.ha.enabled</name>
  <value>true</value>
</property>
<!-- 集群ID号 -->
<property>
  <name>yarn.resourcemanager.cluster-id</name>
  <value>cluster1</value>
</property>
<property>
  <name>yarn.resourcemanager.ha.rm-ids</name>
  <value>rm1,rm2</value>
</property>
<property>
  <name>yarn.resourcemanager.hostname.rm1</name>
  <value>node1</value>
</property>
<property>
  <name>yarn.resourcemanager.hostname.rm2</name>
  <value>node5</value>
</property>
<property>
  <name>yarn.resourcemanager.zk-address</name>
  <value>node2:2181,node3:2181,node5:2181</value>
</property>

3.将在node1主机上配置好的yarn-site.xml，mapred-site.xml文件发送到node2.node3.node5主机上

scp /opt/modules/hadoop-2.5.1/etc/hadoop/yarn-site.xml root@node2:/opt/modules/hadoop-2.5.1/etc/hadoop/
scp /opt/modules/hadoop-2.5.1/etc/hadoop/mapred-site.xml root@node2:/opt/modules/hadoop-2.5.1/etc/hadoop/

scp /opt/modules/hadoop-2.5.1/etc/hadoop/yarn-site.xml root@node3:/opt/modules/hadoop-2.5.1/etc/hadoop/
scp /opt/modules/hadoop-2.5.1/etc/hadoop/mapred-site.xml root@node3:/opt/modules/hadoop-2.5.1/etc/hadoop/

scp /opt/modules/hadoop-2.5.1/etc/hadoop/yarn-site.xml root@node5:/opt/modules/hadoop-2.5.1/etc/hadoop/
scp /opt/modules/hadoop-2.5.1/etc/hadoop/mapred-site.xml root@node5:/opt/modules/hadoop-2.5.1/etc/hadoop/

4.在node1主机上启动yarn

1	[root@node1 hadoop-2.5.1]# ./sbin/start-yarn.sh

1	[root@node1 hadoop-2.5.1]# jps

5.启动单节点ResourceManager

1	[root@node5 hadoop-2.5.1]# ./sbin/yarn-daemon.sh start resourcemanager

1	[root@node5 hadoop-2.5.1]# jps

6.在浏览器地址栏上访问

1	http://node1:8088/cluster/nodes

1	http://node5:8088/cluster/nodes，页面会直接跳转到node1的页面

6.1.杀死node1主机的resourcemanager进程，http://node1:8088/cluster访问不成功

1	[root@node1 hadoop-2.5.1]# kill -9 8145

6.2.查看http://node5:8088/cluster发现node5节点的ResourceManager成为Active

6.3.在node1主机重新单节点启动ResourceManager，发现node1的ResourceManager变成了Standby

1	[root@node1 hadoop-2.5.1]# ./sbin/yarn-daemon.sh start resourcemanager