potiuk commented on issue #15496:
URL: https://github.com/apache/airflow/issues/15496#issuecomment-886167314


   I think it's good as it is. In case of unrecoverable error, when you try 
several times and cannot connect, shutting down the whole application is the 
best thing you can do. It protects against any kind of memory build-ups, 
internal errors etc. Usually -in all serious deployments, the deployment should 
be resilient to application failure and restart it - signalling appropriate 
error via monitoring. Whether you use K8S or Docker Compose or even upstart, 
automated shutdown (and subsequent restarts) of the application is the right 
things that should happen. Any exponential back-offs, retries, monitoring and 
eventual giving up with the restarts should be handled outside of the 
application by monitoring/deployment mechanisms.


-- 
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]


Reply via email to