EKS problem after upgrade from 1.23

Question

Hi, we upgraded eks from 1.23 to 1.30 and now the nodegroup creation fails to create nodes.

In the aws-node, we have this message:

(combined from similar events): Readiness probe failed: {"level":"info","ts":"2024-06-04T10:11:50.568Z","caller":"/usr/local/go/src/runtime/proc.go:267","msg":"timeout: failed to connect service \":50051\" within 5s"}

we installed CNI plugin and upgraded kube-proxy to v1.30.0-eksbuild.3 as mentioned in other posts.

Garre Sandeep · Answer

CNI Plugin Compatibility:

Double-check the specific CNI plugin version you installed for compatibility with Kubernetes 1.30. While you mentioned installing the CNI plugin, ensure it's the recommended version for your current Kubernetes distribution. Refer to the official AWS documentation for known compatible versions:

https://docs.aws.amazon.com/eks/latest/userguide/managing-vpc-cni.html

**kube-proxy Verification:**

Verify that kube-proxy is running correctly on each node. You can use the following kubectl command to check its pod status:

> kubectl get pods -n kube-system | grep kube-proxy

**Network Connectivity:**

Re-evaluate network connectivity between the control plane and the nodes. Security groups might have inadvertently blocked port 50051 after the upgrade. Ensure inbound traffic on port 50051 from the control plane to the nodes is explicitly allowed

For more please follow the link
https://aws.amazon.com/about-aws/whats-new/2023/12/amazon-eks-surfaces-cluster-health-status-details/

https://github.com/aws/amazon-vpc-cni-k8s

EKS problem after upgrade from 1.23

Add your answer

Relevant content