Manually triggering a failover

You might want to change which machine runs as the primary head node. You have the flexibility to switch that role between machines. You might also want to take down a machine for maintenance, or other temporary actions. In that case, you can switch the primary role to the other head node before you shut down.

Before you begin

Before manually triggering a failover, verify that the automatic failover controller is stopped by running the following command on both Db2® Big SQL head nodes:

bigsql status -failoverController

If the automatic failover controller is not stopped, run the following command on both Db2 Big SQL head nodes:

bigsql stop -failoverController

About this task

In some circumstances, the NEWLOGPATH database configuration parameter might be set to specify a different location for the transaction log files. As a result, the second failover that is triggered after this parameter is set might appear to hang. But that is not the case, and the command will complete within the value of the HADR peer window setting. No user intervention is required, and subsequent failovers will run normally.

The NEWLOGPATH parameter might be set to move the transaction logs to a new location due to space constraints. It is recommended that you enable automatic log pruning when HA is enabled.

Important: When a manually triggered or automatic failover occurs, any SQL queries that are running at the precise time of the failover event might fail (SQL1229N or SQL5105N). You can rerun these queries as soon as the failover is completed.

Procedure

To manually trigger a failover, specify the -triggerFailover option in the Db2 Big SQL administration utility.

For more information, see Db2 Big SQL cluster administration utility.