When HVDB changes occur, they are not effective automatically since the role is not notifying "Restart AAC Runtime" handler.
On the other hand, if the role would notify "Restart AAC Runtime", it would always initiate a restart of the AAC Runtime even when other cluster changes occured unrelated to the DB occured.
The python code that will restart the AAC Runtime - will only do a restart if the process is flagged for a restart. So it maybe best to add the handler to the task making the change.
When HVDB changes occur, they are not effective automatically since the role is not notifying "Restart AAC Runtime" handler.
On the other hand, if the role would notify "Restart AAC Runtime", it would always initiate a restart of the AAC Runtime even when other cluster changes occured unrelated to the DB occured.
So I don't know what is best approach.