Subham-KRLX commented on PR #67569: URL: https://github.com/apache/airflow/pull/67569#issuecomment-4551541294
> > We had similar PR recently. I did not like it. > > > If a long running process has a problem then there should be other means. A rolling restart is for me only a workaround and should not be/get a permanent feature in Helm chart therefore. > > > See also https://airflow.apache.org/docs/apache-airflow/stable/faq.html#how-to-prevent-api-server-memory-growth > > +1. Handling of worker restarts should be implemented in the API server itself if it is needed. Agree a rolling restart is a workaround and should not be a permanent chart feature. This PR adds an opt-in CronJob disabled by default so teams can use it as a short term mitigation. -- 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]
