vKS Cluster creation stuck on creating virtual machine object in vCentrer for the cluster nodes

Products

• VMware vSphere Kubernetes Service

Issue/Introduction

• The workload clsuter  machine,vspheremachine,and vm k8s object are not getting created
  

  kubect get cluster,vm,ma,vspheremachine -n namespace01

NAME                                   CLUSTERCLASS             PHASE         AGE    VERSION
cluster.cluster.x-k8s.io/new-cluster   tanzukubernetescluster   Provisioned   36h    v1.29.4+vmware.3-fips.1
  
  

  NAME                                                                       POWER-STATE   AGE
virtualmachine.vmoperator.vmware.com/new-cluster-9jd7q-gmmtk                             49m
  
  

  NAME                                                           CLUSTER       NODENAME                              PROVIDERID                                       PHASE          AGE     VERSION
machine.cluster.x-k8s.io/new-cluster-9jd7q-gmmtk               new-cluster                                                                                          Provisioning   49m     v1.29.4+vmware.3-fips.1
  
  

  NAME                                                                                  ZONE                  PROVIDERID                                       IPADDR
vspheremachine.vmware.infrastructure.cluster.x-k8s.io/new-cluster-7kdqd-fnnll         clusters-vsphere-zone

• The vmware-system-vmop-controller-manager pods logs on the supervisor cluster show error similer to the following:
  
  E0702 04:21:22.132493       1 virtualmachine_controller.go:243] "Failed to reconcile VirtualMachine" err="VirtualMachineClass.vmoperator.vmware.com \"guaranteed-medium\" not found" logger="VirtualMachine" name="namespace01/new-cluster-9jd7q-dc867"
  
  E0702 04:21:22.132953       1 controller.go:329] "Reconciler error" err="VirtualMachineClass.vmoperator.vmware.com \"guaranteed-medium\" not found" controller="virtualmachine" controllerGroup="vmoperator.vmware.com" controllerKind="VirtualMachine" VirtualMachine="namespace01/new-cluster-9jd7q-dc867" namespace="namespace01" name="new-cluster-9jd7q-dc867" reconcileID="0879b032-0f79-49bd-8a00-124c77529f22"
  
  
  
  
• The  workload cluster k8s cluster object showing  clsuter  showing that cluster node creation is faling since the VirtualMachineClass Not Found
  kubectl get cluster  new-cluster -n namespace01 -o jsonpath={'.status'} | jq
  {
    "conditions": [
      {
        "lastTransitionTime": "2025-07-02T17:21:04Z",
        "message": "1 of 2 completed",
        "reason": "VirtualMachineClassNotFound @ /new-cluster-9jd7q-gmmtk",
        "severity": "Error",
        "status": "False",
        "type": "Ready"
      },
      {
        "lastTransitionTime": "2025-07-01T05:16:37Z",
        "message": "Waiting for control plane provider to indicate the control plane has been initialized",
        "reason": "WaitingForControlPlaneProviderInitialized",
        "severity": "Info",
        "status": "False",
        "type": "ControlPlaneInitialized"
      },
      {
        "lastTransitionTime": "2025-07-02T17:21:04Z",
        "message": "1 of 2 completed",
        "reason": "VirtualMachineClassNotFound @ /new-cluster-9jd7q-gmmtk",
        "severity": "Error",
        "status": "False",
        "type": "ControlPlaneReady"
      },
• The workload cluster is created using a vmClass EX: "guaranteed-medium"
  
  kubectl get cluster  new-cluster -n namespace01 -o jsonpath={'.spec.topology.variables'} | jq | grep vmClass -A1
      "name": "vmClass",
      "value": "guaranteed-medium"
  
  
• The "vmClass"  "guaranteed-medium" is not added to the namespce where the cluster is getting created.
  
  get virtualmachineclasses.vmoperator.vmware.com -n namespace01
  
  NAME                  CPU   MEMORY
  best-effort-2xlarge   8     64Gi
  best-effort-4xlarge   16    128Gi
  best-effort-8xlarge   32    128Gi
  best-effort-large     4     16Gi
  best-effort-medium    2     8Gi
  best-effort-small     2     4Gi
  best-effort-xlarge    4     32Gi
  best-effort-xsmall    2     2Gi
  guaranteed-2xlarge    8     64Gi
  guaranteed-4xlarge    16    128Gi
  guaranteed-xsmall     2     2Gi
  
  
• The Workload Management  >> The cluster namespace >> VM Service  >> Manage VM Classes tab not showing the vmClass name which the cluster node used to get created is not selected
  
  

Environment

VMware vSphere Kubernetes Service (vKS)

Cause

The vKS workload load cluster is stuck on creating the cluster node virtual machine since the vmClass name which used to create the  cluster nodes is not selected under Workload Management  >> The cluster namespace >> VM Service  >> Manage VM Classes tab.

Resolution

Add the  missing vmClass to the workload cluster namespce using the following steps:

1. From the vSphere Client home menu, select "Workload Management".
2. Click on the "workload cluster namespace".
3. On the "VM Service" section, click "Manage VM Classes".
4. Find the vmClass needed to create the cluster node.
5. Select the vmClass and click "OK".
   
6. The the required vmClass now will show up as available.
   
   kubectl get virtualmachineclasses -n namespace01
   
   NAME                  CPU   MEMORY
   best-effort-2xlarge   8     64Gi
   best-effort-4xlarge   16    128Gi
   best-effort-8xlarge   32    128Gi
   best-effort-large     4     16Gi
   best-effort-medium    2     8Gi
   best-effort-small     2     4Gi
   best-effort-xlarge    4     32Gi
   best-effort-xsmall    2     2Gi
   guaranteed-2xlarge    8     64Gi
   guaranteed-4xlarge    16    128Gi
   guaranteed-medium     2     8Gi
   guaranteed-xsmall     2     2Gi