We recently had some unusual CI failures here, see e.g. issue #754 and #724.
I'd like to ensure these are noticed quickly. For this we need a good way to monitor them. E.g.
- send an email if certain jobs fail
- Notify some Slack channel about failures
- ...
