Some problems with the current webhook handling:

• You can only really tell if it's working by checking the Celery tasks (indicating a process_zaken_notification task result) and indirectly through the Timeline logger (it might list reasons for not successfully processing a notification, e.g. missing role) or the Yubin mail log (mail sent as a consequence of a webhook).
• You cannot easily tell to which subscription a Celery task result for a notification belongs, which is painful if you're trying to figure out if any have gone stale.
• You cannot at all see any sign in the admin for webhooks being received from stale subscriptions, because they'll be refused at the HTTP handler level, but not logged anywhere (the Celery task is only spawned for accepted notifications).

A lot of this could be improved by logging more to the Timeline logger for diagnostics, but that would significantly increase the volume, so #1899 would be needed first.