Automatic switchover

If the primary server in a HDR-availability system fails, the HDR secondary server can be automatically converted to either a standard or primary database server.

If automatic switchover occurs, the HDR secondary server to rolls back open transactions, and then switches to online mode as either a primary or standard database server. Automatic switchover does not redirect client applications the new primary or standard database server.

Automatic switchover can be a better option than manual failover, but using Connection Managers to control failover for a high-availability cluster has even more advantages.

Automatic switchover requires a very stable network, and works only with primary servers and HDR secondary servers. Connection Managers can be configured to be tolerant of network instability, and can perform failover to HDR, RS, and SD secondary servers. Connection Managers can also prioritize connections between application servers and the primary server if a network failure occurs.

Automatic switchover does not redirect client connections from the old primary server to the new standard or primary server. Connection Managers can redirect clients to whichever secondary server becomes the primary server.

If you use automatic switchover, you must ensure that logical-log files are backed up, and that the HDR server has enough logical-log disk space to allow processing to continue without backing up logical-log files.

To configure automatic switchover or Connection Manager failover, you must set the DRAUTO configuration parameter on all database servers of a high-availability cluster.