Unable to provision Custom ClusterClass with TKG Service 3.2.0 and Later "unable to retrieve kubeadm Configmap from the guest cluster: failed to get API group resources: unable to retrieve the complete list of server APIs: v1: Get https://ip_address:6443"
search cancel

Unable to provision Custom ClusterClass with TKG Service 3.2.0 and Later "unable to retrieve kubeadm Configmap from the guest cluster: failed to get API group resources: unable to retrieve the complete list of server APIs: v1: Get https://ip_address:6443"

book

Article ID: 401161

calendar_today

Updated On:

Products

VMware vSphere Kubernetes Service

Issue/Introduction

Guest cluster provisioning is failing. The cluster object appears under vCenter inventory, but the guest cluster is not reachable and does not become Ready. API requests to the guest cluster fail with timeout errors while provisioning the guest cluster, kubectl describe cluster says that API calls to the Kubernetes control plane fail.

unable to retrieve kubeadm Configmap from the guest cluster: failed to get API group resources: unable to retrieve the complete list of server APIs: v1: Get "https://ip_address:6443/api/v1?timeout=10s": net/http: request canceled while waiting for connection (Client.Timeout exceeded while awaiting headers)

unable to retrieve kube-proxy daemonset from the guest cluster: failed to get API group resources: unable to retrieve the complete list of server APIs: apps/v1: Get "https://ip_address:6443/apis/apps/v1?timeout=10s": net/http: request canceled while waiting for connection (Client.Timeout exceeded while awaiting headers).

failed to create ClusterRoleBinding: failed to get API group resources: unable to retrieve the complete list of server APIs: rbac.authorization.k8s.io/v1: Get "https://ip_address:6443/apis/rbac.authorization.k8s.io/v1?timeout=10s": context deadline exceeded (Client.Timeout exceeded while awaiting headers)

Environment

vSphere 8.x

Cause

This issue is typically seen when the guest cluster is not fully initialised or reachable. Additionally, the required annotation run.tanzu.vmware.com/resolve-tkr: was missing in the YAML, which is essential when using the builtin-generic-v3.1.0 ClusterClass. This annotation helps resolve the correct TKR version during cluster creation.

Resolution

Verify that the TKC cluster YAML includes the following annotation (required when using builtin-generic-v3.1.0 ClusterClass):
run.tanzu.vmware.com/resolve-tkr: ""

Retrieve the full cluster YAML using:
kubectl get cluster <cluster-name> -n <namespace> -o yaml

Confirm whether the official documentation steps were followed during the guest cluster provisioning process.