DB2 pureScale topology changes and high availability disaster recovery (HADR)

In a DB2® pureScale® environment, making changes to the HADR primary cluster and HADR standby cluster can require an outage.

In general, the primary and standby clusters must have the same member topology; that is, each instance must have the same number of members with the same member IDs. The only exception is when you add members to the standby. You can add members when the database is either offline or online. If you drop a member from the primary cluster (dropping a member is not allowed on the standby), you must stop HADR, deactivate the primary, and reinitialize the standby.

If you add a cluster caching facility (CF), you also require an outage on the DB2 pureScale instance.