pmcquighan-camus commented on issue #76871: URL: https://github.com/apache/airflow/issues/76871#issuecomment-6830893399
> working on this, will make the trigger rediscover the job's pods on each loop and wait on the base container instead of the Succeeded phase, so the retried pod gets its xcom extracted. adding tests for the retry case too that sounds great, thank you! as an aside I've noticed that it's possible when querying for matching pods that the old failed pod shows up and so extracting xcom tries to get data from both pods, so it's *maybe* possible that a check on containerStatus is also needed? but i'm not totally sure here. -- 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]
