7.4.4 Process and procedure
This section only discusses a small part of the considerations required when implementing a disaster recovery strategy, inasmuch as System p can assist in the implementation of that strategy. Everything previously discussed regarding disaster recovery still applies. For a useful reference about making WebSphere Application Server highly available and implementing an overall disaster recovery plan, refer to WebSphere Application Server Network Deployment V6: High Availability Solutions, SG24-6688.
A key element of a successful DR plan is to implement known processes and procedures, and test them regularly; the time to test disaster recovery plans should not be during a disaster. In addition, it should become "business as usual" to update this disaster recovery environment when the production environment is updated, because there is little value in having a disaster recovery environment that is not identical to the production environment.
As part of this readiness, scheduled failovers to the disaster recovery environment should be run every so often, to confirm function and ensure that any recent changes have been implemented and updated. One strategy to ensure that this is the case in our example would be to treat Server 1 and Server 2 not as "primary" and "disaster recovery" environments, but rather for whichever is currently holding the production environment to be considered the real production environment, until the time comes to fail over to the other.