fj workflow run: "could not find the triggered run" after every dispatch (server reports an empty event) #321
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?
fj workflow run <wf> --ref <branch>dispatches the workflow, then fails after about 30 seconds with:The run does exist and starts normally.
Cause: after dispatching, fj polls the run list for a new run whose event is
workflow_dispatch. The Forgejo on rasterhub.com reports an emptyeventfor dispatched runs (its payload carriesinputsandref, but theeventfield is""), so the filter never matches. Push runs reportpushand pull request runs reportpull_request, so only dispatch runs are affected. It hits every dispatch, on any branch, including workflows that take their ownrefinput (the run is on the default branch there, which is correct).Fix: accept an empty event as a dispatch candidate, keeping the workflow, branch and not-seen-before checks, and still reject runs that report another event.