I've been using an EKS cluster with success for sometime. today we were going through and cleaning up stacks in cloud formation and I believe we may have deleted something important. Now the status of all the nodes in the EKS cluster is "Unknown" and we cannot connect to them. The same situation occurrs if we create a new node group. We can see that the EC2 instance are generated, but EKS never detects them.
In addtion, if I use Kubectl get pods, I recieve the following message 4 times in succession before the correct result is finally returned:
87482 memcache.go:106] couldn't get resource list for metrics.k8s.io/v1beta1: the server is currently unable to handle the request
New pods will fail with the following error:
Warning FailedScheduling 83s default-scheduler no nodes available to schedule pods
I can only assume the deletions we performed in CloudFormation removed an essntial piece for our EKS cluster. Could someone please helpt us determine what it is and how best to correct this?
