Storage Foundation and High Availability Solutions 7.4.2 Solutions Guide - Windows

To bring up the application on the secondary host

1. From the left pane in the VEA GUI console on the secondary host, right-click on the desired secondary RVG node inside the replication network. Select the Take Over option. The Take Over dialog box is displayed.

   • By default, the Enable Fast-Failback Logging option is selected if the data volumes have DCM logs associated with them. You can use this option to perform takeover with fast-failback logging.

     The DCM is activated for fast-failback logging and the new incoming writes are marked on the DCM of the New Primary. If the replication status of Secondary RVG was Inactive when the Primary failed, then the Enable Fast-Failback Logging option is unavailable for selection. In this case you can perform Take Over without using fast-failback logging.

   • Select the Synchronize Automatically option if you want the new Primary and the original Primary to get synchronized automatically, after the original Primary recovers.

     If you have not selected this option, the original Primary, after it recovers will be in the Acting as Secondary state. To synchronize this original Primary with the new Primary use the Resynchronize Secondaries option from new Primary RVG's right-click menu. When the resynchronization starts, the original Primary which was in the Acting as Secondary state is converted to a Secondary of the new Primary. The new Primary now starts replaying the DCM to update the Secondary with the writes that were written to the DCM.

2. If you do not want to use the Enable Fast-Failback Logging option, clear the checkbox, and click OK to perform Take Over without the fast-failback logging.

   After takeover is complete, to add the Secondary hosts of the original Primary as Secondary hosts of the new Primary, delete the existing RVGs of the original Secondary hosts and then add them as a part of the new Primary.

3. If you have chosen to perform the Take Over operation without using fast-failback logging and the original Primary becomes available again, convert it to a Secondary using the Make Secondary option. Then resynchronize the original Primary with the new Primary using the Synchronize Automatically option. Depending on the size of the data volume this may take quite a while.

   Only after the synchronization is complete can you migrate the Primary role back to the original Primary.

   After takeover, the existing Secondary becomes the new Primary.

4. Assign drive letters to the volumes on the new Primary. Make sure that these drive letters are the same as those of the original Primary.
5. Bring the application resource online.

   Now you can start using the application on the new Primary.