jekeanyanwu commented on PR #66716: URL: https://github.com/apache/airflow/pull/66716#issuecomment-4425856963
Thanks for the heads-up @jscheffl — I had a look at #66705. The two PRs touch the same `trigger_reentry` block but address orthogonal failure modes: - **#66705** handles `event["status"] == "error"` when the pod is *still alive* (transient triggerer/API issues) by re-deferring up to `MAX_REDEFER_ATTEMPTS`. The `get_pod` call itself is left unguarded — moved *outside* the `try/finally` in that PR. - **#66716** (this one) handles the case where the pod is *already gone* by the time `trigger_reentry` runs — `get_pod` raises `ApiException(404)` and escapes uncaught. On success events the trigger already observed completion, so we return cleanly; otherwise we raise `PodNotFoundException`. So they're complementary rather than conflicting in intent, but there's a real textual conflict at the `get_pod` call site. Whichever lands second will need a small rebase. Happy to do that on this side if #66705 goes in first — the fix here collapses to a small `try/except ApiException` around the (now unguarded) `get_pod` call in that PR's structure. Also just pushed a ruff-format fix for the static-checks failure (function signatures had been wrapped where ruff wants them on one line). -- 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]
