Machine suffered from unexpected reboot after Cluster Server was stopped on a node

book

Article ID: 100003707

calendar_today

Updated On:

Description

Error Message

Nov 10 14:13:50 xxxxxxxxx root: [ID 702911 user.alert] Oracle CSSD failure.  Rebooting for cluster integrity.
 

Cause

At the time, it wasn't apparent that the  cssd resource in the main.cf had been removed. This resource is concerned with Clusterware operation. Since the resource had been removed, any attempt at stopping VCS would lead to an Oracle Clusterware-induced panic since the shutdown of VCS brought down the Oracle Clusterware vote device before Oracle Clusterware daemons had been stopped.

Resolution

Ensure that the cssd resource is not removed from the main.cf configuration, and that it is made dependent on the vote device. Please refer to the appropriate Installation and Configuration guide for examples of recommended main.cf configurations.

 

Applies To

Solaris 10

Storage Foundation for Oracle RAC 5.1

Issue/Introduction

This particular issue is only seen in a Storage Foundation Oracle RAC environment. The decision was taken to shut down Cluster Server (VCS) on a node to help remedy another problem. Almost immediately, a system reboot unexpectedly took place.
Powered by Wolken Software