Fault management considerations in ApplicationHA environments

A fault in an application running on an ApplicationHA node may not be detected depending on the VM.GracefulRebootPolicy attribute of the virtual machine.

If VM.GracefulRebootPolicy is enabled and App.RestartAttempts is set to 0, the VBS daemon does not get sufficient time to notify the update to other VBS daemons. If VM.GracefulRebootPolicy is disabled or if the ApplicationHA node is the top tier, fault propagation works correctly. Ensure that VM.GracefulRebootPolicy is disabled if the ApplicationHA tier has parent service groups in a virtual business service.

For information about ApplicationHA-initiated virtual machine restarts, see the ApplicationHA User's Guide.