Hitachi

Hitachi Advanced Database Setup and Operation Guide


16.5.4 When a node failure occurs

When a node failure occurs, the transactions that were executing on the node where the failure occurred are rolled back.

Recovery processing (rollback) performed on update processing that had been running on the master node you forcibly terminated, is executed on the slave node that becomes the new master node. During this recovery processing, if an application program running on a slave node accesses a target page of the recovery processing, that application program's processing might result in an error. At this time, the KFAA31733-E message is output. In this case, follow the steps given in the message.

Furthermore, application programs that were executing transactions on the node where the failure occurred are disconnected from the HADB servers in the multi-node configuration. To re-execute the transactions, connect again to the HADB servers in the multi-node configuration.

If a node failure occurs on the master node, application programs that were not executing transactions on any node at the time the failure occurred, are disconnected from the HADB servers in the multi-node configuration.