johnhoran opened a new issue, #58490: URL: https://github.com/apache/airflow/issues/58490
### Apache Airflow Provider(s) cncf-kubernetes ### Versions of Apache Airflow Providers When using KubernetesPodOperator in deferred mode I recently ran into an issue where airflow wasn't able to hand back from the triggerer to the worker. The underlying pod had succeeded but the task failed, and subsequent retries failed for the same reason until the task was manually cleared. I suspect there is some underlying bug here, that I'm not currently able to reproduce, but I think it raises another point. KPO doesn't actually try recreate the pod on task failure, it tries to adopt the existing pod. If the existing pod has succeeded by the time we get to the retry, then why defer at all, can we just go straight to the reentry method? ### Apache Airflow version 2.11 ### Operating System linux ### Deployment Official Apache Airflow Helm Chart ### Deployment details _No response_ ### What happened _No response_ ### What you think should happen instead _No response_ ### How to reproduce - ### Anything else _No response_ ### Are you willing to submit PR? - [x] Yes I am willing to submit a PR! ### Code of Conduct - [x] I agree to follow this project's [Code of Conduct](https://github.com/apache/airflow/blob/main/CODE_OF_CONDUCT.md) -- 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]
