Recover From a Failed Switchover

Follow

Summary

This Knowledgebase article provides information on what to do when an error is encountered during a switchover.


More Information

Symptoms

A switchover operation fails and the following error message or a similar one can be found in the Neverfail logs:

ControllerSwitch](com.neverfail.controller.ControllerCmd) - AsynchCmdThread: ControllerSwitch failed:
com.neverfail.controller.ControlException: Switchover vetoed

Cause

During the switchover operation, at least one of Heartbeat components has reported that it cannot switch or there has been a channel disconnection. Because of this, the state of the components on the two servers could not be verified (e.g., roles may or may not have already switched). Currently, Heartbeat handles this behavior by starting the application on one of the servers but without replication.

Resolution

Neverfail recommends the administrator check the data integrity of the two servers before starting replication.
If no problems are discovered, it is then safe to start replication.

If any problems are observed, Neverfail recommends that the roles of the servers be changed back to their pre-switchover state. To do this, follow these steps:

  1. Stop Neverfail Heartbeat on both servers selecting "Shutdown Heartbeat and stop protected applications".

    Note: The protected application will be stopped immediately so the users will experience an interruption in the service until the Heartbeat is started on the new active server.
  2. On each server:
    1. Open the Configure Server wizard by selecting Start -> All Programs -> Neverfail -> Configure Server .
    2. From the 'Machine' tab, set the servers' roles as they were before the switchover was issued.
    3. Click the Finish button to save the changes.
  3. Restart Neverfail Heartbeat on both servers (start with the new passive).


Applies To

All versions


Related Information

None

KBID-1296

0 out of 0 found this helpful

Comments

0 comments

Please sign in to leave a comment.