Hi we just upgraded to flink 1.17.1 and want to try the autoscaling feature. I followed the configuration here. Our apache operator version is 1.4.0. Nothing seems to happen with the autoscaler, all tasks still run at the same parallelism set at

FlinkDeployment.spec.job.parallelism

. Are these versions reasonable? Where can I find more information about the autoscaler? Should the logs in JM pods?