General notes on clustered primary server administration
Consider the following operational issues to consider when you use NetBackup as a failover server:
If a NetBackup server fails over and backup jobs are running, one of the following things happen:
If checkpoint restart is enabled for backup jobs for the policy, any backups that were running are resumed at the last checkpoint. Note that the schedule window must permit the job to be resumed.
If checkpoint restart is not enabled for backup jobs, any backup jobs that were running fail. These jobs show up as failed jobs in the Activity Monitor or NetBackup reports. Scheduled backup jobs are retried according to the normal retry logic. (See the and the attributes). The schedule window and so on must permit the retry of the job. After a failover occurs, the user must restart user-directed backups and manual backups.
Refer to the NetBackup Administrator's Guide, Volume I, for more information on checkpoint restart , Job retry delay, and Scheduled backup attempts.
If a NetBackup primary server fails over, then any restore jobs that were running during that time will fail. These jobs appear as failed jobs in the Activity Monitor and in the NetBackup reports. Even if you try to resume the failed jobs after the server has failed over successfully, the jobs remain in a partial state.
To resolve this issue, you must trigger such restore jobs again after the primary server has failed over successfully.
After a failover occurs on a NetBackup failover server, it takes time for the tape devices and media servers to be reset. For example, a tape may have to be rewound before it is made available. Until the device is again ready for use, you may see errors. For example, you may see error 213 (no storage units available for use) or error 219 (the required storage unit is unavailable).
After a failover, NetBackup may change the status of a drive that is required for a backup or restore to PEND. More information is available about how to use the Device Monitor to resolve a pending request or action.
If you have the NetBackup Administration Console open when a failover occurs, you need to refresh the display. This refresh ensures that the services on the active node are displayed correctly. In the NetBackup Administration Console (from the View menu) click .
The backup and restore process is the same whether you are in a cluster or a non-cluster environment. See the NetBackup Troubleshooting Guide, for further information on the process for backups and restores.
True Image Restore (TIR) is not supported for clustered shared drives.