服役新结点&退役旧结点

2019-07-28  本文已影响0人  Map_Reduce

服役新结点

  1. 环境准备
    克隆一台和原来一模一样的机器(环境要相同)
    注意:
    ①修IP地址和主机名称
    ②删除原来HDFS文件系统留存的文件(/opt/module/hadoop-2.7.2/data和log)
  2. 服役新节点具体步骤
    (1)直接启动DataNode,即可关联到集群
[zby@hadoop105 hadoop-2.7.2]$ sbin/hadoop-daemon.sh start datanode
[zby@hadoop105 hadoop-2.7.2]$ sbin/yarn-daemon.sh start nodemanager

(2)如果数据不均衡,可以用命令实现集群的再平衡

[zby@hadoop102 sbin]$ ./start-balancer.sh
starting balancer, logging to /opt/module/hadoop-2.7.2/logs/hadoop-atguigu-balancer-hadoop102.out
Time Stamp               Iteration#  Bytes Already Moved  Bytes Left To Move  Bytes Being Moved

退役旧结点

方法一:添加白名单

添加到白名单的主机节点,都允许访问NameNode,不在白名单的主机节点,都会被退出。
配置白名单的具体步骤如下:
(1)在NameNode的/opt/module/hadoop-2.7.2/etc/hadoop目录下创建dfs.hosts文件

[zby@hadoop102 hadoop]$ pwd
/opt/module/hadoop-2.7.2/etc/hadoop
[zby@hadoop102 hadoop]$ touch dfs.hosts
[zby@hadoop102 hadoop]$ vi dfs.hosts

添加如下主机名称(不添加hadoop105)

hadoop102
hadoop103
hadoop104

(2)在NameNode的hdfs-site.xml配置文件中增加dfs.hosts属性

<property>
<name>dfs.hosts</name>
<value>/opt/module/hadoop-2.7.2/etc/hadoop/dfs.hosts</value>
</property>

(3)配置文件分发

[zby@hadoop102 hadoop]$ xsync hdfs-site.xml

(4)刷新NameNode

[zby@hadoop102 hadoop-2.7.2]$ hdfs dfsadmin -refreshNodes
Refresh nodes successful

(5)更新ResourceManager节点

[zby@hadoop102 hadoop-2.7.2]$ yarn rmadmin -refreshNodes
17/06/24 14:17:11 INFO client.RMProxy: Connecting to ResourceManager at hadoop103/192.168.1.103:8033

(6)如果数据不均衡,可以用命令实现集群的再平衡

[zby@hadoop102 sbin]$ ./start-balancer.sh
starting balancer, logging to /opt/module/hadoop-2.7.2/logs/hadoop-atguigu-balancer-hadoop102.out
Time Stamp               Iteration#  Bytes Already Moved  Bytes Left To Move  Bytes Being Moved

方法二:添加黑名单

在黑名单上面的主机都会被强制退出。

  1. 在NameNode的/opt/module/hadoop-2.7.2/etc/hadoop目录下创建dfs.hosts.exclude文件
[zby@hadoop102 hadoop]$ pwd
/opt/module/hadoop-2.7.2/etc/hadoop
[zby@hadoop102 hadoop]$ touch dfs.hosts.exclude
[zby@hadoop102 hadoop]$ vi dfs.hosts.exclude

添加如下主机名称(要退役的节点)

hadoop105
  1. 在NameNode的hdfs-site.xml配置文件中增加dfs.hosts.exclude属性
<property>
<name>dfs.hosts.exclude</name
 <value>/opt/module/hadoop-2.7.2/etc/hadoop/dfs.hosts.exclude</value>
</property>
  1. 刷新NameNode、刷新ResourceManager
[zby@hadoop102 hadoop-2.7.2]$ hdfs dfsadmin -refreshNodes
Refresh nodes successful
[zby@hadoop102 hadoop-2.7.2]$ yarn rmadmin -refreshNodes
17/06/24 14:55:56 INFO client.RMProxy: Connecting to ResourceManager at hadoop103/192.168.1.103:8033
  1. 检查Web浏览器,退役节点的状态为decommission in progress(退役中),说明数据节点正在复制块到其他节点,如下图所示


    退役中.jpg
  2. 等待退役节点状态为decommissioned(所有块已经复制完成),停止该节点及节点资源管理器。注意:如果副本数是3,服役的节点小于等于3,是不能退役成功的,需要修改副本数后才能退役,如下图所示


    已退役.jpg
  3. 如果数据不均衡,可以用命令实现集群的再平衡
[zby@hadoop102 hadoop-2.7.2]$ sbin/start-balancer.sh 
starting balancer, logging to /opt/module/hadoop-2.7.2/logs/hadoop-atguigu-balancer-hadoop102.out
Time Stamp               Iteration#  Bytes Already Moved  Bytes Left To Move  Bytes Being Moved

注意:不允许白名单和黑名单中同时出现同一个主机名称。最好用添加黑名单的方式来退役旧结点!

上一篇 下一篇

猜你喜欢

热点阅读