kbendick opened a new issue #3002: URL: https://github.com/apache/iceberg/issues/3002
I've noticed that when we push to a PR, the previous workflow run does not get cancelled. To me, this seems like a waste of resources, as in almost all cases, users want the results from their most recent workflow run (and don't really care about previous results enough to block up the limited CI results in the apache queue). I've experimented with creating a cancel work flow run action in my own fork, however: - GitHub Actions have special behaviors in forks - I'm not a committer (so don't necessarily have full access in this repo) I am going to copy the cancel duplicate workflow runs from Spark, and then will work with a committer (assuming I can find one, might tap @cwsteinbach on this as he has been doing a lot of repository related work). The one from Spark simply camels duplicate runs, but it skips on push event (which I believe is what we _do_ want to have it do). However, I think copying from Spark and then going from there and the documentation of that GitHub action is probably the best idea. Spark Cancel Duplicate Workflows GH Action: https://github.com/apache/spark/blob/master/.github/workflows/cancel_duplicate_workflow_runs.yml Documentation on Action used in Spark repo: https://github.com/potiuk/cancel-workflow-runs I will sync with people from my team, as well as Carl (potentially, depending on his availability), to see about implementing this. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
