I've AWS EKS cluster in my organization for internal practice purpose.

Currently installed applications on this cluster are - Jenkins, Nexus, Spinnaker, Grafana, Prometheus

Now in order to save the cost and optimal usage, I used to scale up to 8 nodes every morning and down to 0 nodes every evening.

I've observed that sometimes when I wanted to see the kubectl logs or run kubectl exec command I used to get this error

Error from server: no preferred addresses found; known addresses: []

To quickly resolve it, I delete the pod and let it created automatically again, and then this error would gone.

However, today I'm running jenkins build, where Jenkins master spin up a jenkins-slave pod dynamically to run the build process, and everytime I'm getting the same error again in the jenkins logs

java.net.ProtocolException: Expected HTTP 101 response but was '500 Internal Server Error'


hudson.remoting.ProxyException: io.fabric8.kubernetes.client.KubernetesClientException: no preferred addresses found; known addresses: []

Please advise how to resolve this issue as its becoming frequent now.

I've also found that in worker nodes Internal IPs are not assigned, may be due to that, this issue is occurring.

NAME                                        STATUS   ROLES    AGE    VERSION               INTERNAL-IP   EXTERNAL-IP   OS-IMAGE         KERNEL-VERSION                  CONTAINER-RUNTIME
xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx   Ready    <none>   3h7m   v1.12.10-eks-aae39f   <none>        <none>        Amazon Linux 2   4.14.171-136.231.amzn2.x86_64   docker://18.9.9
xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx   Ready    <none>   3h7m   v1.12.10-eks-aae39f   <none>        <none>        Amazon Linux 2   4.14.171-136.231.amzn2.x86_64   docker://18.9.9
xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx   Ready    <none>   3h7m   v1.12.10-eks-aae39f   <none>        <none>        Amazon Linux 2   4.14.171-136.231.amzn2.x86_64   docker://18.9.9

Thanks