Hello all,
we had a working EKS 1.15 Cluster and did an update to 1.16. After the update I did an update for the managed worker nodes to use the latest 1.16 AMI. Unfortunately nodes with the available 1.16 AMI do not join the cluster now. For now we run the nodes with 1.15 AMI and that works fine. The nodes are in two managed node groups (one for 1.15 and the other for 1.16). So they are managed completely by AWS. They all run in a private subnet that has a NAT gateway attached.
When I do a kubectl describe node <WORKER_NODE> I get the following error message:
KubeletNotReady runtime network not ready: NetworkReady=false reason:NetworkPluginNotReady message:docker: network plugin is not ready: cni config uninitialized
My question is, what do I need to configure to get my managed worker nodes with 1.16 up and running? There must be some differences regarding 1.15 and the needed network setup that I do not know or could find in the documentation.
Some more details:
kubectl get nodes
NAME STATUS ROLES AGE VERSION
ip-192-168-112-139.eu-central-1.compute.internal Ready <none> 14d v1.15.10-eks-bac369
ip-192-168-151-23.eu-central-1.compute.internal Ready <none> 14d v1.15.10-eks-bac369
ip-192-168-170-31.eu-central-1.compute.internal Ready <none> 14d v1.15.10-eks-bac369
ip-192-168-173-63.eu-central-1.compute.internal Ready <none> 14d v1.15.10-eks-bac369
ip-192-168-188-165.eu-central-1.compute.internal NotReady <none> 4h1m v1.16.8-eks-fd1ea7
If you need more information please let me know. If someone can point me to the correct documentation link, that solves this issue, would be greatly appriciated.
THX for any help here because I am lost
