other rc scripts for services use daemon(8) with "-r" to recover from daemon crashes automatically without requiring admin intervention.
Today our unbound resolver (v1.10) crashed but since unbound does not make use of daemon it stayed down until we started it manually - which is worse than it could be.
Please add daemon to unbound's rc script to recover from crashes automatically.
for an example on how to use daemon:
ping for maintainer feedback.
(In reply to Li-Wen Hsu from comment #1)
Ah, forgot abut this feature request.
It would actually be more beneficial to find out why unbound-s crashing. I'm not sure whether it is a good idea to blindly restart a crashing process. Let me thonk about this.
I agree that it is important to find and solve the root cause of the crash but in the moment of the crash the most important thing is to get the service back up again. Using log monitoring we are not "blind" and notice nonetheless that the crash happened and can have a manual look at the crash logs.