wombatu-kun opened a new pull request, #16771:
URL: https://github.com/apache/iceberg/pull/16771
## What
Adds a `concurrency:` group with PR-scoped `cancel-in-progress` to the five
lightweight, PR-triggered workflows that did not have one: `docs-ci`,
`license-check`, `zizmor`, `codeql`, and `asf-allowlist-check`.
```yaml
concurrency:
group: ${{ github.workflow }}-${{ github.ref }}
cancel-in-progress: ${{ github.event_name == 'pull_request' }}
```
## Why
The ASF GitHub Actions pool is a shared, capacity-limited resource, and
infra guidance asks every project to cancel superseded PR runs so stale commits
stop holding runner slots. When several commits are pushed to a PR in quick
succession, runs triggered by earlier commits keep going even though their
results are no longer relevant. The heavy CI workflows (spark, flink, java,
hive, kafka-connect, delta-conversion, and others) already cancel superseded PR
runs with exactly this pattern; these five lightweight workflows were the
remaining gap. `docs-ci` is the most valuable to fix because each superseded
run holds a scarce macOS runner.
## Notes
The block is identical to the one already used across the heavy workflows,
including the conditional `cancel-in-progress: ${{ github.event_name ==
'pull_request' }}` so that pushes to `main` and scheduled runs (zizmor, codeql,
asf-allowlist-check) are never interrupted; only superseded `pull_request` runs
are cancelled. For the workflows that also run on `main` or on a schedule, two
overlapping non-PR runs now serialize within the group instead of running in
parallel, which is harmless for these short jobs.
--
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]
