Node Eviction in RAC

The node eviction/reboot for I/O fencing to ensure that writes from I/O capable clients can be cleared avoiding potential corruption scenarios in the event of a network split, node hang, or some other fatal event in clustered environment.                                                                          Evictions occur when CSS detects a heartbeat problem that must be dealt with.  Why it occur?

-  Lost network communication(30 sec) with another node(s) or lost disk heartbeat (200 sec) information from another node(s)

Diagnosis:

OS log messages (/var/log/messages) If you notice  "kernel: SysRq : Resetting". This shows this node is evicted by some other node from cluster.

CSSD logs  <ORA_CRS_HOME>/log/<hostname>/cssd

/etc/oracle/oprocd/<hostname>. oprocd.log

This file exists in <CRS_HOME>/log/<hostname>/cssd/oclsomon dir for errors