Windows operating systems

Managing tape failover in a cluster

Edit online

As part of your regular routine, check the event log to ensure that the configuration is operating properly. If a server fails, the error is logged. The log provides you with information to understand why the failure took place.

Before you begin

To ensure that the server can identify or reset device names upon failover, set the SANDISCOVERY option to ON. By default, this option is set to OFF. For more information, see SANDISCOVERY.

About this task

Sometimes a node must rejoin the cluster, for example:
  • When a node failed
  • When a new Host Bus Adapter fiber card is added (equipment changes)

Procedure

Complete the following tasks in any order to ensure that a node can successfully join the cluster:

  • Update, if necessary, the drive and library that use the IBM Storage Protect cluster tool.
  • Take the IBM Storage Protect server offline until the failed node rejoins the cluster. This action helps ensure that the IBM Storage Protect server that is running on the other node is not affected.