Don't retry webhook deliveries to obviously-invalid addresses

Bug #1879687 reported by Colin Watson on 2020-05-20
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Launchpad itself
Low
Unassigned

Bug Description

Occasionally somebody sets up webhook deliveries to target URLs that obviously can't work. For example, this happens if somebody sets up a local deployment of something like build.snapcraft.io, causing webhooks to be created for hosts like 0.0.0.0 or something.lxd. This causes lots of retry attempts to build up in our celery job queue; they clear eventually, but sometimes cause spurious alerts in the meantime.

I'm not sure whether we can outright refuse to deliver to such target URLs, as it may be difficult to work out exactly what "obviously can't work" immediately (and we need to consider legitimate cases of local test deployments too), but I think we can probably do some better error handling in the job itself and selectively decline to retry the job in some more cases.

To post a comment you must log in.
This report contains Public information  Edit
Everyone can see this information.

Other bug subscribers