Pod scheduling is based on Requests so you can definitely go without setting a limit. I don't think the limit itself plays a role in Kubernetes scheduling (unle

Pod scheduling is based on Requests so you can definitely go without setting a limit. I don't think the limit itself plays a role in Kubernetes scheduling (unless you don't specify a request, then limit equals request). And you are definitely right: scheduling a pod without request/limit is like giving a blank check.