Two Actions runs stuck queued with zero jobs; cancellation returns contradictory state errors

[Sunny091]

Type

Bug report — GitHub Actions workflow execution / scheduling

Summary

Two manually dispatched runs in our public repository remain queued with zero jobs. The first run cannot be cancelled: normal cancellation reports that it is completed, while force cancellation reports that it has not been queued yet. A fresh dispatch of the same commit is also queued with zero jobs.

Repository and affected runs

• Repository: https://github.com/NCHU-NLP-Lab/docker-ml-dev
• Workflow: .github/workflows/build.yml — Build and Publish
• Ref: main
• Commit: 15735db9653c255a0a10ce23712866d1c05b5c09
• Run 1: https://github.com/NCHU-NLP-Lab/docker-ml-dev/actions/runs/42979897223
  • Created: 2026-11-11 07:29:48 UTC
  • Check suite: 116393089894
• Run 2: https://github.com/NCHU-NLP-Lab/docker-ml-dev/actions/runs/42982107980
  • Created: 2026-11-11 07:47:47 UTC
• Both were still queued with zero jobs when checked at approximately 2026-11-11 12:20 UTC.

Observed API responses

Run queries return status: queued, conclusion: null; jobs endpoints return total_count: 0. The first run's check suite is also queued with zero check runs.

Normal cancellation of run 1:

gh run cancel 42979897223 --repo NCHU-NLP-Lab/docker-ml-dev
Cannot cancel a workflow run that is completed

Force cancellation of run 1:

POST /repos/NCHU-NLP-Lab/docker-ml-dev/actions/runs/42979897223/force-cancel
HTTP 409
Cannot cancel a workflow run that has not been queued yet.

After those cancellation failures, querying run 1 still returned queued. Dispatching the workflow once more created run 2, which also has not produced any jobs.

Checks performed

• Repository is public, not archived or disabled.
• Repository Actions permissions report enabled: true, allowed_actions: all.
• Workflow state is active.
• No pending deployments were reported for run 1.
• The workflow has no concurrency group or deployment environment.
• It defines three matrix jobs on ubuntu-latest and a Spark build job on ${{ vars.SPARK_BUILD_RUNNER || 'ubuntu-24.04-arm' }}.
• No repository-level SPARK_BUILD_RUNNER variable was found. Organization-level policy/variable queries returned 403 with the available credentials, so organization restrictions have not been ruled out.
• No Docker build or publishing step has started; there are no job logs to diagnose.
• A separate minimal runner-probe workflow has not been run.

Expected behavior and request

The runs should create jobs and execute, or reach a terminal status with an actionable error. Cancellation should consistently reflect the run's actual state.

Could GitHub staff investigate the scheduling/check-suite state for these run IDs, help clear any stuck records, and confirm whether an organization restriction or a backend issue is preventing job creation? This is blocking our container-image build and publishing workflow.

The documented symptoms under Troubleshooting workflows — Canceling Workflows appear similar, but we have not established the root cause.

[github-actions[bot]]

Thank you for your interest in contributing to our community! We currently only accept discussions created through the GitHub UI using our provided discussion templates. Please re-submit your discussion by navigating to the appropriate category and using the template provided.

This discussion has been closed because it was not submitted through the expected format. If you believe this was a mistake, please reach out to the maintainers.

[jamesyorke]

Also having this issue.

[x0craft]

I'm having this issue as well on a private repo.

[BlackPearlsDev]

Same here

[kteslenko-cpu]

same here. private repo

[MarkoKorenic]

same problem with private repo

[MrMcNab]

Same here.

Got my ticket closed by Virtual Assistant

[jvisker]

Same issue

[jvisker]

https://github.com/orgs/community/discussions/212803