`假设新增一台slave5:172.16.0.8` #### 准备工作 `a.所有机器添加主机名slave5的解析` ```bash cat /etc/hosts ``` ```bash 172.16.0.3 master 172.16.0.4 slave1 172.16.0.5 slave2 172.16.0.6 slave3 172.16.0.7 slave4 172.16.0.8 slave5 ``` `b.master添加用户hadoop免密登陆` ```bash su - hadoop ssh-copy-id slave4 scp .bash_profile slave4:/opt/hadoop ``` &emsp; #### 主节点slaves添加新机器信息 `master` ```bash cat > etc/hadoop/slaves << EOF master slave1 slave2 slave3 slave4 slave5 EOF ``` &emsp; #### 安装java和hadoop `同上操作即可` &emsp; #### 拷贝hadoop配置到新增节点 ```bash su - hadoop scp etc/hadoop/* slave5:/opt/hadoop/etc/hadoop/ scp sbin/* slave5:/opt/hadoop/sbin/ ``` #### 启动节点 ```bash ./sbin/hadoop-daemon.sh start datanode ./sbin/yarn-daemon.sh start nodemanager ``` #### 执行数据负载均衡 ```bash start-balancer.sh -threshold 1 ```