Currently the image registry when using two replicas requires (requiredDuringSchedulingIgnoredDuringExecution) the pods to be scheduled in different nodes. If the number of replicas is higher than 2 then preferredDuringSchedulingIgnoredDuringExecution will cause them to run in different nodes but do not enforce it. This means on a multiple nodes(i.e. 3 infra) cluster, when set replicas to 3, the 3 image registry pods schedule the same nodes sometimes. 

This is very common for larger clusters to increase image registry replicas and currently requires the customer to set requiredDuringSchedulingIgnoredDuringExecution on the operator while this behaviour is not documented in terms of the difference from 2 replicas versus 3 replicas.

Can we enforce affinity for 3 replicas across 3 nodes as well?