if we get initial errors from a notification, we retry several times to deliver the notification. if the notification keeps failing, we stop after a while sending this notification.
if notification to a specific endpoint keep failing for a longer period of time, we disable the endpoint for future notifications.
you can acquire/get the status of a webhook, or you can ping a webhook
if your function app goes down, you should check if your webhook is still active on restart. If it was down for an extensive length of time, you might need to check on the status of your running transcriptions.