After a subcluster is "shut down"(shutdown set to true), its pods will stay down and cannot get restarted by the operator. We go a step further and delete the pods by scaling the statefulset to zero. This is useful for realdb as we do not want to keep down pods around and the EKS instances will also be shut down.
When shutdown is set back to false, the statefulset will be scale up to its original size(subcluster size in spec).
After a subcluster is "shut down"(shutdown set to true), its pods will stay down and cannot get restarted by the operator. We go a step further and delete the pods by scaling the statefulset to zero. This is useful for realdb as we do not want to keep down pods around and the EKS instances will also be shut down. When shutdown is set back to false, the statefulset will be scale up to its original size(subcluster size in spec).