ADMIN_WAIT state reported after one or more Veritas Cluster Server(VCS) nodes are rebooted.

book

Article ID: 100002878

calendar_today

Updated On:

Description

Error Message

The cluster will report an error as ADMIN_WAIT status for the rebooted node.

 

Cause

This is caused by ServerA as it fired the command "hastop -local" before ServerB rebooted, but one or more resources fails to offline and get hung. This causes the ServerA to report "LEAVING" state

Owing to this issue, VCS will report ADMIN_WAIT status for node ServerB, after it is rebooted.

 

Resolution

Run the following commands step by step to resolve the issue

1) Run the following command on ServerA ( for which the VCS state is "Leaving" ).
# hastop -local -force


2) Run the following command on ServerB ((for which the VCS state is "admin_wait" ).
# hastop -local


3) Run the following command on both nodes.
# hastart

 

 

 

Issue/Introduction

The issue is seen while VCS is in abnormal state. For example, if there is a two nodes VCS cluster. The nodes are ServerA and ServerB. The node ServerB can't re-join the cluster after its system is rebooted. It can be found that the VCS state of ServerA is in "Leaving" state and ServerB is in "admin_wait" state via the following command.

# hastatus -sum

-- SYSTEM STATE
-- System State Frozen

A ServerA LEAVING 0
A ServerB ADMIN_WAIT 0
Powered by Wolken Software