Upgrade NetBackup failover server on the PCS Linux
Before you upgrade any options, ensure that the NetBackup resource is online in the cluster. Also ensure that server processes are running on the active node.
To upgrade NetBackup failover server on PCS Linux
- Ensure the NetBackup resource is online in the cluster and that server processes are running on the active node before proceeding with any upgrades.
- Confirm the existence of a catalog backup for your cluster environment. For more details, refer to the Configuring NetBackup catalog backups in a cluster.
- Prior to the upgrade, take the PCS NetBackup cluster resource offline.
Note:
Ensure that shared disk and IP resources remain online on the active node during the upgrade process.
- Freeze the PCS cluster using the command: pcs property set maintenance-mode=true
- Disable the NetBackup cluster agent resource on all nodes of the NetBackup group: pcs resource disable <nbu_cluster_agent_resource_name>
- The upgrade procedure will vary depending on the current release you are upgrading from.
- Install the NetBackup server software on the active node:
Ensure to follow the upgrade instructions provided in the NetBackup Installation Guide.
Specify the virtual server name if required.
- Install the NetBackup server software on each inactive node that may potentially fail over.
- After completing the upgrade installation, verify if security certificates are deployed on all cluster nodes. Generate certificates if necessary. For detailed instructions, refer to the NetBackup Security and Encryption Guide.
- Enable the PCS NetBackup cluster agent resource on all nodes:pcs resource enable <nbu_cluster_agent_resource_name>
- Unfreeze the PCS cluster to resume normal operation:pcs property set maintenance-mode=false
- Confirm that all Pacemaker resources are online.
- Verify the proper failover functionality of NetBackup within the cluster. see, Verifying NetBackup configuration.
These steps ensure a systematic and controlled upgrade process for the NetBackup failover server while maintaining cluster integrity and service availability.