Clusters

Veeam Plug-in supports backup and restore of IBM Db2 databases that operate as part of a failover cluster.

Veeam Plug-in supports data backup and restore for the following types of clusters:

High availability disaster recovery (HADR)
Failover clusters

To back up data with Veeam Plug-in, clustered databases must be managed with the following cluster management software:

IBM PowerHA (for clusters running IBM AIX only)
IBM Tivoli System Automation (TSA)
Pacemaker (for clusters running Linux-based OS only)

Backup of Cluster

To back up databases that operate as part of a failover cluster, do the following:

Install and configure Veeam Plug-in for IBM Db2 on each node of the cluster. For more information, see Installing Veeam Plug-in for IBM Db2 and Configuring Plug-in.
Perform the backup on the active cluster node. For more information, see Performing Backup.

Veeam Plug-in will start the backup job on the primary cluster node.

In case of clustered database, Veeam Plug-in can store data from several cluster nodes in the same backup file on the backup repository. To do this, Veeam Plug-in needs the name of the cluster. Depending on the cluster management software that you use, Veeam Plug-in gets the cluster name in one of the following names:

In case of IBM PowerH and IBM TSA, Veeam Plug-in gets the cluster name automatically.
In case of Pacemaker, you must set the cluster name in the veeam_config.xml file. You must set the name with the customServerName parameter. The parameter value must be the same as the name of your cluster.

For example, if your cluster has the name db2 and the domain for the environment is customer1.local, you have to set the following entry:

Keep in mind that you must add the parameter to the existing line in the veeam_config.xml file. If you create a new line with the same name as the existing line, Veeam Plug-in will consider parameters only in the first detected line. Other parameters will be ignored.

Restore of Cluster

To restore a database that operates as part of a failover cluster, you must start the restore process on the primary cluster node.