I have a Kubernetes cluster running within Azure AKS, which has a node pool dedicated to a service called orchestrator. The node pool only runs that specific service and has scaled UP 20 nodes. But I noticed one of the first nodes has 82 pods, while the others have around 30, some 13 others 40. I wonder why is Kubernetes putting 30 pods in two different nodes since it's clear that one single node could handle 82? Is there any setting where I can prevent that?
Here is a screenshot:
