Install hbase cluster (cdh5.0.2) in the form of tarball -- how to configure zookeeper in the distributed Cluster

For cluster installation overview, see here

Zookeeper Configuration

1. Add ZK path settings to/etc/profile. See the background description above.

2. Go ~ /ZK/conf directory, copy zoo_sample.cfg to zoo. cfg

Vim zoo. conf

Ticktime = 2000 initlimit = 10 synclimit = 5 datadir =/home/hadoop/HDFS/ZK/datadatalogdir =/home/hadoop/HDFS/ZK/logclientport = 2181server. 21 = zk1: 2888: 3888server. 22 = ZK2: 2888: 3888server. 23 = zk3: 2888: 3888

Make sure to create ~ /HDFS/ZK/data and log directories. In the data directory, You need to manually create a myid file with numbers corresponding to X in server. X.

Start zookeeper and run the following command on each ZK node:

Zkserver. Sh start

View the zookeeper status and use

Zkserver. Sh status

If any problem occurs, you can start zookeeper the day before yesterday and find the problem by searching the console output:

Zkserver. Sh start-foreground