NetApp - Forcing a Controller Failover

If you're testing the configuration of a new NetApp storage appliance, preparing to perform maintenance on a controller (e.g. adding a new PCIe card), or preparing to perform an upgrade, you may need to perform a manual failover from one controller to another.  These are done primarily with the cf suite of commands.

To check on the status of services on each controller, issue the following command:

cf status


When you're prepard to begin the failover process, log onto the controller that is taking over for the other controller and issue this command:

cf takeover

 

This will cause LUNs and services to be moved from the opposing controller to the controller on which you executed the command.  Once services have been passed over, the opposing controller will reboot.  In my experience, the migration of services happens very quickly.  It takes 3-5 minutes for the opposing controller to fully reboot, although your mileage may vary on older systems.  Once you've completed your testing/maintenance activities, you can restore services by executing the following command on the controller that currently owns the services:

cf giveback