That way, when most of the worker services (the ones ingesting) are stopped.
We can let the checker worker service run.

Providing a better feedback to the deposit's clients (when the downtime is solely about the obj/storage)