Prerequisites: JDK and Hadoop have been configured, and a Hadoop cluster (three hosts only) has been set up.

1. Unpack the zookeeper – 3.4.10. Tar. Gz

Tar -zxvf zookeeper-3.4.10.tar.gz -c /opt/module/Copy the code

2. CD Go to the Decompressed Zookeeper directory and create a zkdata folder (/opt/module/zookeeper-3.4.10/ zkdata directory).



3. Go to the conf file and modify the zoo_sample. CFG



Rename zoo_sample. CFG to zoo.cfg(nice look)



4. Configuration zoo. CFG

DataDir =/opt/module/zookeeper-3.4.10/zkData

(2) add :(zzf111 is the hostname)

server.1=zzf111:2888:3888      

server.2=zzf112:2888:3888

server.3=zzf113:2888:3888

5. Add environment variable /opt/module/zookeeper-3.4.10

vi /etc/profileCopy the code

At the end of the profile file add:

exportZK_HOME = / opt/module/zookeeper - 3.4.10export PATH=TH=$PATH:$ZK_HOME/binCopy the code



then

source /etc/profileCopy the code

6. Go to the zkData folder and create a myID file



And then the vi myid

By adding a number to myID (the number after server), zzf111 users add the number 1



7. Go to the directory on zooKeeper and upload ZooKeeper-3.4.10 to another host



8. For zzf112, change the myid file under zkData to 2.



Zzf113 host same method

9. Start the ZooKeeper cluster. Run the script to start the ZooKeeper service on each node of the ZooKeeper cluster, as shown in the following figure

Bin/zkserver. sh start (Check the status of all three hosts.) bin/ zkserver. sh statusCopy the code





At this point, the distribution of ZooKeeper is complete