We would greatly use kinda dangerous feature. Sometimes our long running tests just fail with a stupid error, e.g. suddenly an API provider is down for a while. We need to be able to mark the flow as “successful” so that the flow can continue and we can quickly deploy on our staging servers.
It steals lots of our precious time. For production, we wouldn’t of course use this feature as it is a madness.
This is our workflow:
workflows: version: 2 build_and_deploy: jobs: - build: filters: tags: only: /^v[0-9]+(\.[0-9]+)*/ - deploy_prod: requires: - build filters: tags: only: /^v[0-9]+(\.[0-9]+)*/ branches: ignore: /.*/ - deploy_stag: type: skippable <============================================================ requires: - build filters: branches: only: master
When the job build fails, we need to restart the job again. Using e.g.
type: skippable in the job deploy_stag could continue the flow from this very job if we say YES somewhere in your administration. (A similar thing to your Manual Approval.)
What do you say about it?