hadoop集群中 退役旧数据节点

环境说明

192.168.100.101 node01
192.168.100.102 node02
192.168.100.103 node03
192.168.100.104 node04

四台Vmware Workstatuin中搭建的centos6.9主机，hadoop版本为2.6.0

①创建etc/hadoop/dfs.hosts.exclude配置文件，并添加需要退役的节点

[root@node01 hadoop]# cat dfs.hosts.exclude 
node04

②修改hdfs-site.xml，添加修改hdfs-site.xml属性

<property>
<name>dfs.hosts.exclude</name>
<value>/export/servers/hadoop-2.6.0-cdh5.14.0/etc/hadoop/dfs.hosts.exclude</value>
</property>

③namenode节点刷新namenode，刷新resourceManager

hdfs dfsadmin -refreshNodes
yarn rmadmin -refreshNodes

④节点退役完成，停止该节点进程

[root@node01 hadoop]# hadoop-daemon.sh stop datanode
stopping datanode
[root@node01 hadoop]# yarn-daemon.sh stop nodemanager
stopping nodemanager

⑤从dfs.hosts文件中删除退役节点

[root@node01 hadoop]# cat dfs.hosts
node01
node02
node03

⑥namenode节点刷新namenode，刷新resourceManager

hdfs dfsadmin -refreshNodes
yarn rmadmin -refreshNodes

⑦从namenode的slave文件中删除退役节点

[root@node01 hadoop]# cat slaves 
node01
node02
node03

⑧负载均衡

start-balancer.sh