When EKS nodes fail to join the cluster, it’s often due to permissions or network config hiccups. Here are a few things to double-check:
IAM Policies: Make sure your node group’s role (example-node-group-role) has AmazonEKSWorkerNodePolicy, AmazonEC2ContainerRegistryReadOnly, and AmazonEKS_CNI_Policy attached. Without these, the nodes can’t communicate with EKS or pull necessary images.
Subnet Configuration: Are your subnets public, or do they at least have a route to an internet gateway? If nodes can’t reach the internet, they’ll struggle with necessary connections (like downloading container images).
Security Groups: Ensure that the control plane and node group security groups allow communication over essential ports—443 for API access, 10250 for worker node kubelet communication, etc.
Give these a try, and if you’re still stuck, I’d be happy to dive in further with you! You can reach me at my Profile.