pipelines-as-code
pipelines-as-code copied to clipboard
Rerun in github app is not working
clicking on rerun button give me this in the log:
∙ 09:53:32 CEL expression is not matching push-generate-coverage-releaseyaml-, skipping
∙ 09:53:32 CEL expression is not matching go-testing-, skipping
∙ 09:53:32 cannot match pipeline from payload to a pipelinerun in .tekton/ dir
∙ 09:53:32 payload target event is push with source branch main and target branch main
∙ 09:53:32 available configuration of the PipelineRuns annotations in .tekton/ dir
∙ 09:53:32 cannot match pipeline from payload to a pipelinerun in .tekton/ dir, event=push, branch=mai
/assign
good luck, i was sure i fixed it and it always somehow broke 🙃 there is a functional test for it but what github does and what we do does not seem to match
not able to reproduce this 🤔
https://github.com/openshift-pipelines/pipelines-as-code/assets/55777192/db39c649-aabf-43ea-9e58-70eac49956ee
although it should start only the failed one but it restarts all of them
humm weird what about clicking on the Rerun field on the left near the PR (for individual rerun)
it does restart all runs
🤔
https://github.com/openshift-pipelines/pipelines-as-code/assets/55777192/293b2cb8-fb2f-4b64-a3f8-a47ce98d1d87
ah i think it's on push to main! rerunning there when there is a failure
Can verify this problem in v0.19.4 (OCP Pipelines 1.11) with GHE:
{"level":"warn","ts":"xxx","logger":"pipelinesascode","caller":"matcher/annotation_matcher.go:170","msg":"CEL expression is not matching xxx-, skipping","provider":"github","event-id":"xxx","event-sha":"xxx","event-type":"push","namespace":"xxx"}
It indeed occurs, when I try to re-run a failed push pipeline on main.
This is still a problem. While it is possible to rerun the Tekton pipeline manually, this approach does not report back to GitHub appropriately. So other developers get the wrong information about a pipeline's status.