Forgejo Actions runners offline, release pipelines stalled since 2026-05-27 #165
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
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?
Symptom
Both Forgejo Actions runners are
offline, so no workflow has executed since 2026-05-27 ~03:34 UTC. Every release pipeline (coily and siblings) is stalled.Evidence (via
/api/v1/admin/actions/runners):forgejo-runner-0(uuid 524f4035...) status=offline, v12.10.1, labels [docker]forgejo-runner-1(uuid ebd5fe3e...) status=offline, v12.10.1, labels [docker]Newest Actions task is id 134 (
windows-assets, headfeb7a45e) created 2026-05-27T03:34:17Z. Pushes to coily main since then (8be1ccd0,df5a17f3) produced no runs.Impact
sync-lockdownjob (coily#143) cannot run, so the canonical-bump fan-out is untested and not propagating.bump-formulahas been failing on every historical run anyway (empty token bug, fixed in coily df5a17f3 but needs a runner to verify).Likely fix
Runner deploy is
infrastructure/deploy/forgejo-runner.ymlon k3s (kai-server). Check the runner pod(s):kubectl -n forgejo get pods, restart the runner Deployment/DaemonSet, confirm re-registration (status flips to online). Note: this session's user cannot read/etc/rancher/k3s/k3s.yaml(permission denied), so kubectl-based diagnosis needs a context that can.