re: JRA-43836, this isn't the first time a JIRACloud update has killed webhooks. For me specifically, this is a really bad outcome, as my cloud addon is dependent on them. All my customers got an outage because of this, some may still not know there is a problem. I don't know who has the problem because I can't guage status, which leads me to:

As a Cloud addon, I have no way to validate the 'heath' of the core Atlassian Connect infrastructure (of which webhooks are a part). Some potential options:

1. Atlassian could implement some post instance startup validation to sanity check things like 'is webhooks enabled / usable' (specifically talking of JIRACloud here, so Atlassian can 'know' about the problem at boot time, rather than customer feedback time )
2. Atlassian could create an Hourly Webhook 'Im alive' POST including a payload of internal healthheck status. This allows addons to track when a host was last 'healthy', potentially communicating to host owner about that. (possible, but Id rather not get auto pinged to death, if I got a webhook 5m ago, I dont need to know webhooks are healthy)
3. Atlassian could add a new REST service to divulge internal health check would make all manner of 3rd party monitoring possible, Pingdom : "webhooks : enabled" etc. I would find this very handy. (possibly valid for JIRAserver as well as JIRACloud, being able to locally in an addon, know I hadn't got a webhook for more than an hour, could start polling health to check status, taking notification action as necessary)