Our site is configured with bidirectional replication. We have a document managment and an e-mail archiving application reading and writing from seperate virtual pools. Is it a good practice to configure the IP addresses used by applications to include the access node IP addresses of both the primary and secondary clusters?. i.e. The configuration would have: IP address , IP address , IP address , IP address . The thinking behind this is that if the primary cluster is unavailable the application will write/read from the secondary becuase the secondary IP addresses are defined and (hopefully) zero down time.