Hi, I am currently working shipping flink jobs to production and have two general queries/ concerns regarding the deployment that I wanted to list down
1) Currently we deploy jobs on Kubernetes on Application mode and want to do a blue-green deployment but Stopping a job with Savepoint terminates the deployment rather than suspend/ Finish the Job so that it can be later drained or Restarted, Is there any better way to do a safe upgrade for jobs running in Application mode on Kubernetes? 2) The pod-template used in Native deployment currently allows adding of new volume and configmaps but is there is a way we can override the existing volumes, configmaps used for flink-conf.yaml without depending on the cli arguments for more complex usecases like encoding secrets in configmaps and decoding them in our init-containers Thank you -- <https://grofers.com>
