Monitoring Data - Big Data Made Easy: A Working Guide to the Complete Hadoop Toolset - page 193

Next, I modify the HBase configuration files on all HBase cluster nodes under /etc/hbase/conf. In the
regionservers file used in the Linux Cat command, I insert the name of the HBase master node:
[root@hc1r1m1 conf]# cat regionservers
hc1r1m1
In the HBase environment file hbase-env.sh, I make sure that logging is set up and I ensure that HBase is
configured to not manage ZooKeeper. This is because the ZooKeeper servers are already running and are being used
by multiple Hadoop components. HBase can use them, but does not need to manage them, thereby avoiding any
impact on their availability for other components:
export HBASE_LOG_DIR=/var/log/hbase
export HBASE_MANAGES_ZK=false
I create a link in the HBase configuration directory to the Hadoop core configuration file so that HBase has access
to the Hadoop configuration:
[root@hc1r1m1 conf]# ln -s /etc/hadoop/conf/core-site.xml core-site.xml
[root@hc1r1m1 conf]# ls -l
total 28
lrwxrwxrwx. 1 root root 30 Aug 9 12:07 core-site.xml -> /etc/hadoop/conf/core-site.xml
I next set up the hbase-site.xml file by adding the following property entries between the file's configuration tags:
<property>
<name>hbase.zookeeper.quorum</name>
<value>hc1r1m1,hc1r1m2,hc1r1m3</value>
<description>
Comma separated list of Zookeeper servers (match to what is specified
in zoo.cfg but without portnumbers)
</description>
</property>
<property>
<name>hbase.cluster.distributed</name>
<value>true</value>
</property>
<property>
<name>hbase.master.wait.on.regionservers.mintostart</name>
<value>1</value>
</property>
<property>
<name>hbase.rootdir</name>
<value>hdfs://hc1nn:8020/hbase</value>
</property>
<property>
<name>hbase.rest.port</name>
<value>60050</value>
</property>