Environment='JAVA_OPTS=-Djava.awt.headless=true -Djava.security.egd=file:/dev/urandom -Djava.io.tmpdir=/opt/datagaps/DataOpsServer/temp'

3.Need relocate postgresql data directory.

   https://help.datagaps.com/articles/#!dataops-suite/red-hat-linux/a/h3_747174585

Prerequisites 

1.Need passwordless communication between nodes and within nodes.

2.datagaps user required in all nodes make home directory /opt/datagaps

Deployment

1.extract deployment files in /opt/datagaps/

2.edit the below files according to the environment.

   a./opt/datagaps/hadoop/etc/hadoop/core-site.xml

         hdfs://<hostname>:9000

   b./opt/datagaps/hadoop/etc/hadoop/yarn-site.xml

        updatehost and update values from qa environment

   c./opt/datagaps/livy7/conf/livy.conf

        livy.server.host <hostname>

   d./opt/datagaps/spark-3.0.1-bin-without-hadoop/conf/spark-defaults.conf

        update host name from 72 to 76th line

   e.add below lines to  /opt/datagaps/ .bashrc

       export HADOOP_HOME=/opt/datagaps/hadoop

      export PATH=${PATH}:${HADOOP_HOME}/bin:${HADOOP_HOME}/sbin
export JAVA_HOME=/opt/datagaps/hadoop/jre
export PATH=${PATH}:${JAVA_HOME}/bin
export HADOOP_CONF_DIR=/opt/datagaps/hadoop/etc/hadoop
f. add below line to /opt/datagaps/.profile
PATH=/opt/datagaps/hadoop/bin:/opt/datagaps/hadoop/sbin:$PATH

 3.send hadoop folder to other two nodes /opt/datagaps/ make sure the owner should be datagaps.

 4.Run below commands as datagaps user.  in master

      a.hdfs namenode -format

      b.start-dfs.sh

      c.start-yarn.sh

      d.hdfs dfs -mkdir -p hdfs:///var/log/spark/apps

      e.hdfs dfs -mkdir -p hdfs:///user/spark/warehouse

 5.copy livyserver.service to /etc/systemd/system/

 6.systemctl daemon-reload

 7.systemctl start livyserver.service