Build+deployment workflow does not include Slack notifications #16788
Description
Description
When the new docker+GAR+ArgoCD pipeline was added, the workflow was mostly based on previous art from other teams, and their notifications or integration could have been different — but here most of the runs on sensitive branches send results to #www-notify, which this pipeline, one of the most important, does not. When it succeeds there's a lot of additional events that announce themselves, but if a failure occurs early on in the build+push CI, e.g. a timeout on docker or apt infra or some availability issue on pypa or npmjs side, such issue is not surfaced enough for more folks 👀 to be notified.
While adding yet another start+end notifications doesn't look that helpful, it could at least trigger in case of a failure.
Success Criteria
- Decide whether CI start+finish event notifications to Slack are valuable to add, given the second job has its own notifier, and the final job triggers yet another notifier, if successful…
- However jobs failed in the initial build steps before dockerhub are not notified to any other place besides the CI status check:/
- Maybe the best would be not notify on start and on success, but only notify on failure?
┆Issue is synchronized with this Jira Task
┆Priority: P2
┆Sprint: Backlog