Sequence of Red Hat Cluster Services to be stopped during maintenance

Consider you have 2 node Red Hat cluster server with an an Oracle RAC DB whose storage /u02 is mounted in a GFS2 filesystem.When you wanted to do patching in one of the node in the cluster you normally follow the below sequence steps:

• Take backup of DB.

Activity in Node1

• Note down the pmon process "ps -aef | grep pmon".
• Stop the DB instances.
• Stop the crs service (ie., ASM pmon) as root user.
• Stop the GFS2 service "/etc/init.d/gfs2 stop"
• Stop the Cluster Resource Manager service "/etc/init.d/rgmanager stop"
• Stop the Cluster Volume Manager service "/etc/init.d/clvmd stop"
• Stop the Cluster manager service "/etc/init.d/cman stop"
• Now node1  has been separated from Red Hat and Oracle DB cluster.
• Patch the node1 server and reboot.

Follow similar steps as above in Node2, once the Node1 DB connectivity is up and running.