Stop asking the event for inputs it will never have #1087
No reviewers
Labels
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
jaylouisw/packetforge!1087
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "ci/auto-rebase-drop-inputs"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Twenty runs, none ever started. Not on schedule, not on
workflow_dispatchwith inputs supplied, not on push after #1086 removed the concurrency group.So #1086 was wrong, and here is the evidence rather than another hypothesis:
Meanwhile
ci.ymldispatches normally on the same runner, sameruns-on, same trigger types. The job row exists in both cases —status 5,runs_on ["ubuntu-latest"],task_id 0— so the difference is not in the job, it is in whatever decides to turn the job into a task.The one thing left in this file that
ci.ymldoes not do is evaluategithub.event.inputs. An expression that cannot be evaluated at job-preparation time produces exactly this shape: a job that exists, is pending, and never becomes a task.It bought nothing — the two variables fed
batchandsettle_seconds, both of which the script already defaults. They are gone, and the defaults are literals now.If the next push to main still never starts a run, the problem is not expressions either. At that point the honest move is to stop guessing at this file and read Forgejo's dispatcher, and I will say so rather than opening a fourth variation of the same guess.