Not able to setup the cluster using v1.30.1

[nisarg2121]

I am trying to create an EKS-D cluster using kubeadm on an Ubuntu machine. During the setup, the cluster initialization failed with the following error:

Cluster setup failed with the error The API server is not healthy after 4m0.000160729s

Upon examining the logs, I found that the API server pod is failing when attempting to mount the volume for the certificates. The specific error is a read-only file system error.

Error Logs

kubelet[27121]: E0626 10:34:25.747755 27121 pod_workers.go:1298] "Error syncing pod, skipping" err="failed to \"StartContainer\" for \"kube-apiserver\" with RunContainerError: \"failed to start container \\"a0acc763c942f07021975dbf3a6b652b962aa2897eedad30ba7cca4ccc31eda4\\": Error response from daemon: failed to create task for container: failed to create shim task: OCI runtime create failed: runc create failed: unable to start container process: error during container init: error mounting \\"/etc/ssl/certs\\" to rootfs at \\"/etc/ssl/certs\\": mkdir /var/lib/docker/overlay2/6a8df9b9fde244dfc48086a12e2c1ebb1bbf3a1847ff46df289c609f3582f63f/merged/etc/pki/tls: read-only file system: unknown\"" pod="kube-system/kube-apiserver-k8s-master" podUID="070cb4f18a0fdef28cd17a5615cdcc01"

Steps to Reproduce

• Set up an Ubuntu machine.
• Install Docker version 25.0.3 and CRI-Dockerd version 0.3.10.
• Use kubeadm to create an EKS-D cluster with the specified release channel and release version.
• Observe the error during the cluster initialization process.

Environment:

• EKS Distro Release Channel: 1-30-8
• EKS Distro Release: 1-30-8
• OS: Ubuntu
• Docker Version: 25.0.3
• CRI-Dockerd Version: 0.3.10

[vineeth-bandi]

Hi, I was not able to recreate this issue using these versions. Can you share the flags you are passing with kubeadm init as well as Ubuntu version?