gyfora commented on PR #227: URL: https://github.com/apache/flink-kubernetes-operator/pull/227#issuecomment-1152018399
@Aitozi thank you for raising these concerns. I think the main question is, why would anyone use autoscaling/HPA (and expect it to work) if they hardcode parallelism or have very fine grained resource requirements. I think if they have these problems the autoscaling logic is really not for them and I think it's very hard to solve it nicely. The general assumption for HPA is homogeneous resources. The hardcoding / overwriting of configuration from within your program is a concern in general when using the operator. Not just for scaling but any manual spec change too we should advise users to only use the spec to configure these settings. Other than this, since the current behaviour is not impacted by this change, people can decide whether to use HPA or not. :) -- 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]
