5、linux上安装zookeeper

1.基本信息

版本	3.4.6
安装机器	三台机器
账号	hadoop
源路径	/opt/software/zookeeper-3.4.6.tar.gz
目标路径	/opt/zookeeper -> /opt/zookeeper-3.4.6/
依赖关系	无

2.安装过程

1).使用hadoop账户，通过tar -xxvf命令将zookeeper-${version}.tar.gz解压缩至目的安装目录,并设置软连接。

[root@bgs-5p173-wangwenting opt]# su hadoop

[hadoop@bgs-5p173-wangwenting opt]$ cd /opt/software

[hadoop@bgs-5p173-wangwenting software]$ tar -zxvf zookeeper-${version}.tar.gz -C /opt/

[hadoop@bgs-5p173-wangwenting software]$  cd /opt/

[hadoop@bgs-5p173-wangwenting opt]$  ln -s /opt/zookeeper-${version}   /opt/zookeeper

修改zoo.cfg配置文件。

[hadoop@bgs-5p173-wangwenting zookeeper]# cd /opt/zookeeper/conf

[hadoop@bgs-5p173-wangwenting conf]$ cp zoo_sample.cfg   zoo.cfg

[hadoop@bgs-5p173-wangwenting conf]$ vim zoo.cfg

2).在zoo.cfg文件中添加如下配置：

dataDir=/opt/zookeeper/data

dataLogDir=/opt/zookeeper/log

server.1=bgs-5p173-wangwenting:2888:3888

server.2=bgs-5p174-wangwenting:2888:3888

server.3=bgs-5p175-wangwenting:2888:3888

3).创建data目录和log目录，并在data目录下创建文件，命名为myid：

[hadoop@bgs-5p173-wangwenting conf]$ mkdir /opt/zookeeper/data

[hadoop@bgs-5p173-wangwenting conf]$ mkdir /opt/zookeeper/log

[hadoop@bgs-5p173-wangwenting conf]$ vim  /opt/zookeeper/data/myid

在myid文件中写入1（同理在另外两台机器中写2和3）

1

4).配置zookeeper运行的环境变量。

[hadoop@bgs-5p173-wangwenting conf]$ vim /etc/profile

添加下面的内容：

export ZK_HOME=/opt/hadoop/zookeeper

export PATH=$ZK_HOME/bin:$PATH

配置成功后，执行source /etc/profile使配置生效

[hadoop@bgs-5p173-wangwenting conf]$ source /etc/profile

5).将安装包从bgs-5p173-wangwenting服务器复制到bgs-5p174-wangwenting，bgs-5p175-wangwenting两台机器，并配置环境变量：

[hadoop@bgs-5p173-wangwenting conf]$  scp -r /opt/zookeeper-3.4.6  hadoop@bgs-5p174-wangwenting:/opt

[hadoop@bgs-5p173-wangwenting conf]$  scp -r /opt/zookeeper-3.4.6  hadoop@bgs-5p175-wangwenting:/opt

在bgs-5p174-wangwenting和bgs-5p175-wangwenting上分别执行

[hadoop@bgs-5p174-wangwenting conf]$ ln -s /opt/zookeeper-3.4.6 /opt/zookeeper

[hadoop@bgs-5p175-wangwenting conf]$ ln -s /opt/zookeeper-3.4.6 /opt/zookeeper

在bgs-5p174-wangwenting和bgs-5p175-wangwenting上分别修改/opt/zookeeper/data/myid文件，bgs-5p174-wangwenting中的myid改为2bgs-5p175-wangwenting中的改为3

[hadoop@bgs-5p174-wangwenting conf]$ vim /opt/zookeeper/data/myid

2

[hadoop@bgs-5p175-wangwenting conf]$ vim /opt/zookeeper/data/myid

3

在bgs-5p174-wangwenting和bgs-5p175-wangwenting上分别步骤4配置zookeeper运行的环境变量

6).在三台机器上分别执行zookeeper命令：$ zkServer.sh start

[hadoop@bgs-5p173-wangwenting conf]$  zkServer.sh start

[hadoop@bgs-5p174-wangwenting conf]$  zkServer.sh start

[hadoop@bgs-5p175-wangwenting conf]$  zkServer.sh start

7).检查每台机器的zookeeper状态：$ zkServer.sh status

若输出结果是leader 或者follower, 三台机器中有一台是leader 和2 个follower, 说明安装成功了。

[hadoop@bgs-5p173-wangwenting conf]$  zkServer.sh status

[hadoop@bgs-5p174-wangwenting conf]$  zkServer.sh status

[hadoop@bgs-5p175-wangwenting conf]$  zkServer.sh status

8).检查zkCli命令是否能够执行成功：

[hadoop@bgs-5p173-wangwenting conf]$ zkCli.sh -server bgs-5p173-wangwenting:2181,bgs-5p174-wangwenting:2181,bgs-5p175-wangwenting:2181

若能成功连接，说明安装启动成功了