I am learning to work with EKS and I have 3 clusters, 1 in ap-south-1 and 2 in us-east-1. All 3 clusters have been configured the same way, at least I have tried to and each have 3 nodes. But the node names in the ap-south-1 cluster looks different from the node names in the 2 clusters in us-east-1.
The nodes in the ap-south-1 cluster is of the form ip-<IP>.ap-south-1.compute.internal.
The nodes in the 2 east-us-1 clusters is of the form ip-<IP>.ec2.internal.
Now, I don't have enough information to know if this is a regional thing or not, but I tried to create new nodegroups in the clusters and they are following their respective naming schemes.
Don't misunderstand me, there is nothing wrong at all but this behaviour has thrown me off and I can't find any info anywhere. Some insight would be great!
This seems to just be an unimportant inconsistency between implementations in different regions.
The docs don't say why, so this sentence is a guess, but this is likely because us-east-1 usually gets new features before other regions, and by the time the naming convention was settled it was too late to change us-east-1 to match without potentially breaking things for users.