Replies: 1 comment
|
💬 Your Product Feedback Has Been Submitted 🎉 Thank you for taking the time to share your insights with us! Your feedback is invaluable as we build a better GitHub experience for all our users. Here's what you can expect moving forward ⏩
Where to look to see what's shipping 👀
What you can do in the meantime 💻
As a member of the GitHub community, your participation is essential. While we can't promise that every suggestion will be implemented, we want to emphasize that your feedback is instrumental in guiding our decisions and priorities. Thank you once again for your contribution to making GitHub even better! We're grateful for your ongoing support and collaboration in shaping the future of our platform. ⭐ |
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
🏷️ Discussion Type
Bug
💬 Feature/Topic Area
Other
Discussion Details
A workflow run in a public repository has been
queuedsince 2026-08-19 — 22days — and cannot be reached by any API call. It is not a job waiting for a
runner: the run never expanded into jobs at all.
Run: https://github.com/PyDevices/pygraphics/actions/runs/32219727413
The API's answers contradict each other and the run's own record:
GET /runs/32219727413status: queuedPOST /runs/32219727413/cancelPOST /runs/32219727413/force-cancelPOST /runs/32219727413/rerunDELETE /runs/32219727413gh run cancelThe run reports itself as queued, cancel reports it was never queued, rerun
reports it is running, and delete requires a completed run — which it cannot
become. Those four cannot all be true, which is what makes this a backend state
inconsistency rather than a misconfiguration.
How it started: the run was orphaned when a commit removed the workflow
definition (
.github/workflows/docs.yml) while this run was already queued. Theremoval was deliberate — ReadTheDocs replaced MkDocs as the documentation
builder.
Already ruled out:
state: activebutdoes not release the run. Tried twice, 2026-08-20 and 2026-09-03, retrying
cancel / force-cancel / delete in between, reverting afterwards both times.
completed successfully.
gh run deleteuses the sameDELETEendpoint and gets the same 403.ever done this. It is not reproducible on demand.
This is not an isolated report. Searching before posting turned up the same
failure repeatedly, none of them resolved:
re-run, cancel returns 409
workflow stuck queued; cancel, force-cancel (409) and delete (403) all fail
stuck queued, cancellation returns HTTP 500
Pages workflows stuck queued, cannot be cancelled
cancel a workflow re-run that has not yet queued"
over 24h, cannot cancel
The common shape across all of them: a run whose public status disagrees with
the scheduler's, zero jobs, and every removal endpoint refusing for a different
and mutually exclusive reason.
What I am asking for: two things. Confirmation that this is a platform-side
inconsistency with no repository-side remedy — and, given how often it recurs,
whether there is any plan for a user-reachable way to clear a run in this state.
Right now the only exit appears to be manual intervention by GitHub, requested
one discussion at a time.
Impact: low but permanent. The run reads as perpetually active, so it
appears in every CI health sweep and cannot honestly be filtered out.
All reactions