Zookeeper集群部署

准备

三台Linux虚拟机(这里是centos7.5，我将其分别命名为hadoop001、hadoop002、hadoop003)

jdk1.8.0_151

zookeeper-3.4.6

配置hosts文件及hostname

#在所有节点的hosts文件中最加集群ip。注意一定不要动hosts文件中的前两行，这是个巨坑！ echo "192.168.56.101 hadoop001" >> /etc/hosts echo "192.168.56.102 hadoop002" >> /etc/hosts echo "192.168.56.103 hadoop003" >> /etc/hosts #在hadoop001节点上执行 echo "hadoop001" > /etc/hostname hostname hadoop001 #在hadoop002节点上执行 echo "hadoop002" > /etc/hostname hostname hadoop002 #在hadoop003节点上执行 echo "hadoop003" > /etc/hostname hostname hadoop003

在所有节点上部署JDK

安装所有节点上的Zookeeper

---------------------------------在所有节点上安装操作------------------------------- mkdir /opt/app/ tar -zxvf zookeeper-3.4.13.tar.gz /opt/app/ cd /opt/app/ #创建一个软链接 ln -s /opt/app/zookeeper-3.4.13 zookeeper cd zookeeper #配置zookeeper环境变量 echo 'export ZOOKEEPER_HOME=/opt/app/zookeeper/' >> ~/.bash_profile echo 'export PATH=${ZOOKEEPER_HOME}/bin:$PATH' >> ~/.bash_profile source ~/.bash_profile #修改文件配置 mkdir data cp conf/zoo_sample.cfg conf/zoo.cfg dataDir=/opt/app/zookeeper/data echo "server.1=hadoop001:2888:3888" >> conf/zoo.cfg echo "server.2=hadoop002:2888:3888" >> conf/zoo.cfg echo "server.3=hadoop003:2888:3888" >> conf/zoo.cfg vi conf/zoo.cfg #修改dataDir路径：dataDir=/opt/app/zookeeper/data ---------------------------------单独操作------------------------------------ echo 1 > data/myid #在hadoop001中执行该命令 echo 2 > data/myid #在hadoop002中执行该命令 echo 3 > data/myid #在hadoop003中执行该命令

启动所有节点上的zk

zkServer.sh start #启动zk zkServer.sh status #查看zk是否成功启动，注意：是在所有zk都启动成功后，执行此命令查看，出现如下信息则启动成功

  或