Other Node Reboots When Rebooting One Node in SFRAC as dgfailpolicy Parameter is Set Wrong (Doc ID 1328779.1)

Last updated on SEPTEMBER 06, 2012

Applies to:

Oracle Server - Enterprise Edition - Version 10.2.0.4 and later
Information in this document applies to any platform.
Platform: All platforms where SFRAC (Veritas Cluster File System) and Oracle RAC can be installed


Symptoms

The installation involves Oracle CRS/Grid Infrastructure on top of Symantec SFRAC (Veritas Cluster and Veritas Cluster File System from version 4.0 until latest 5.1).

During high availability tests to reset, reboot or power off one of the node, the surviving node hangs when accessing the shared disks (OS level commands like "df -k", "ls -l" also hang) and possibly reboots.

In this case, node2 was rebooted, shortly node1 went down on it's own:

Jun  6 14:44:26 racnode1 vmunix: Phase 0 - /dev/vx/dsk/ocrvotedg03/ocrvotevol - Blocking buffer reads for recovery. gencnt 0 primary 1 leavers: 2 0 0 0
..
Jun  6 14:44:37 racnode1 syslog: Oracle CRS failure.  Rebooting for cluster integrity.

Cause

Sign In with your My Oracle Support account

Don't have a My Oracle Support account? Click to get started

My Oracle Support provides customers with access to over a
Million Knowledge Articles and hundreds of Community platforms