IBM® Db2 active/passive using cluster failover software

When using Db2® active/passive using cluster failover software, the standby node takes over the primary node's resources (the database files, logs) and identity (IP address, SAN WWNN). The database service is then started on the standby node.

Conceptually, Db2 active/passive failover configurations using cluster failover software operates as described in Active/Passive Cluster Failover Configurations.

During the startup, Db2 goes through its normal crash recovery and ensures committed changes are made to the database and incomplete transactions are rolled back. When Db2 is finished with crash recovery, the database service is made available.

From the Sterling Order Management System Software perspective, you can expect the following to occur after the primary node fails (and the database server is unavailable).

• Transactions in the application, agent and integration servers that were actively processing throw a SQL error message. The changes from those transactions are correctly rolled back later when the database server comes up on the standby node.
• The Sterling Order Management System Software servers continually reissue the transactions until the database service is restored. You do not have to restart the Sterling Order Management System Software servers during the transition to the standby node.
• If the source of the work request (specifically for the agents and integration servers) came from message queues, the messages remain in the message queue. When the database service is restored, these messages are processed.

Setting up and testing an active/passive failover configuration can be tricky with many interdependencies and related parts. We strongly encourage you to contact the cluster failover vendors for assistance in planning and implementing your cluster failover.