CI workflow: go test + go build on push #2
Loading…
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?
Workflow file is in the repo and the run is queued, but no runner picks it up: the existing runners are scoped elsewhere and a repo-scoped runner for Winnie/bankwars is being registered. In progress.
Workflow file is in the repo and correct (go vet + go test on push). Runner situation: Forgejo 16.0.3 accepts user-level runner registration (Register returns 200 and a runner record) but Declare is rejected as unregistered, so the job sits waiting. Existing instance runners serve other scopes. This needs either a Forgejo upgrade or an admin-scoped runner. Blocked on infra, not on the workflow.
Fixed, CI is green on the regular runners. The real problem was mine, not the runners: the workflow asked for the label docker:docker://debian:trixie while the instance runners declare docker, so every job waited forever. Changed to runs-on: docker, and added a setup step that installs git (the debian:trixie job image has no git, which broke checkout on the first picked-up run). The extra runner I built is deleted, container and registrations both. Run 18 on master: success.