Zookeeper study notes (two) installation automation configuration and common commands

Single machine or distributed installation

Decompress zookeeper to the corresponding paths on different machines

For stand-alone installation, change the for loop to the IP of your own host

#!/bin/bash

zootmp=/root/liutao/zookeeper/tmp		# zk的目录下的数据目录，需要自己创建,此路径用于存储zookeeper中数据的内存快照、及事物日志文件
zoocfg=/root/liutao/zookeeper/conf/zoo.cfg	# zk目录下的配置文件的路径

myid=1
for h7_node in h71 h72 h73;do
ssh -Tq root@$h7_node << EOF
echo $myid
if test -e $zootmp
then 
        rm -rf $zootmp
        mkdir $zootmp
        touch $zootmp/myid
        echo $myid >> ~/liutao/zookeeper/tmp/myid		// zk主机号
else
        mkdir $zootmp
        touch $zootmp/myid
        echo $myid >> ~/liutao/zookeeper/tmp/myid		// // zk主机号
fi

if test -e $zoocfg
then
        echo "tickTime=2000" > $zoocfg
        echo "initLimit=10" >> $zoocfg
        echo "syncLimit=5"  >> $zoocfg
        echo "dataDir=/root/liutao/zookeeper/tmp" >> $zoocfg
        echo  "clientPort=2181" >> $zoocfg		// 2181：对cline端提供服务
        echo "server.1=h71:2888:3888" >> $zoocfg 		// 3888：选举leader使用 , 2888：集群内机器通讯使用（Leader监听此端口）
        echo "server.2=h72:2888:3888" >> $zoocfg 
        echo "server.3=h73:2888:3888" >> $zoocfg 
         

else
        touch $zoocfg
        echo "tickTime=2000" > $zoocfg
        echo "initLimit=10" >> $zoocfg
        echo "syncLimit=5"  >> $zoocfg
        echo "dataDir=/root/liutao/zookeeper/tmp" >> $zoocfg
        echo  "clientPort=2181" >> $zoocfg
        echo "server.1=h71:2888:3888" >> $zoocfg 
        echo "server.2=h72:2888:3888" >> $zoocfg 
        echo "server.3=h73:2888:3888" >> $zoocfg 
         
fi
exit
EOF

myid=`expr $myid + 1`
done

Environment variable

export ZOOKEEPER_HOME=/root/liutao/zookeeper
export PATH=$PATH:$ZOOKEEPER_HOME/bin

Common commands

start stop

// 启动zookeeper		zkServer.sh start
// 停止zookeeper	    zkServer.sh stop
// 查看状态:zkServer.sh status

Create node

zkCli.sh  链接本地zk服务器，退出使用quit
create [-s] [-e] path data #其中-s 为有序节点,-e 临时节点，-s会记录每个子节点创建的先后顺序

Create a persistent node and write data, get command to get data

Persistent ordered nodes and write data, the node name will be changed to a combination of name and number, the role is to create a unique ID

After the temporary node is created, quit exits the session, the node is automatically deleted, and the temporary node cannot have child nodes.

Update node

The version number will be updated after the node is updated

zkCli.sh  链接本地zk服务器，退出使用quit
set 路径 更新的数据 【当前dataversion，可选】		#

Delete node

delete path	【当前dataversion，可选】		# 结点有子结点，是不允许删除的
rmr path		# 删除该结点以及所有的子结点

View node

get path		# 返回数据和当前结点的属性
stat path		# 返回当前结点属性

Property description:

Return to node list

ls path		# 返回当前路径下的所有子结点
ls2 path	# 返回当前路径下的所有子结点和当前子结点的属性