cluster fails to deploy greater than k8s 1.23

[Vivian Rook]

Deploying k8s cluster on Bobcat I get a failure to deploy k8s clusters on versions greater than 1.23.

I can deploy a template changing only the kube_tag variable from v1.23.15-rancher1-linux-amd64 to v1.26.8-rancher1 (or versions of 1.24 or 1.25) the template builds, and the cluster starts to deploy, though only the master node deploys.

The heat log (/var/log/heat-config/heat-config-script/15fceb87-7fdf-40db-aa7e-0c35bcae1b6b-paws-dev-126-1-lurpkkeiwwva-kube_masters-jfbcxmblt4n4-0-o6nsp2f3habo-master_config-3im4cpl67mdb.log) gives:
```
Trying to label master node with node-role.kubernetes.io/master=""
++ kubectl get --raw=/healthz
+ '[' ok = ok ']'
+ kubectl patch node paws-dev-126-1-lurpkkeiwwva-master-0 --patch '{"metadata": {"labels": {"node-role.kubernetes.io/master": ""}}}'
Error from server (NotFound): nodes "paws-dev-126-1-lurpkkeiwwva-master-0" not found
+ echo 'Trying to label master node with node-role.kubernetes.io/master=""'
+ sleep 5s
```
paws-dev-126-1-lurpkkeiwwva-master-0 does appear to exist in `openstack server list`

I'm deploying with tofu, my resources look like the following:

```
resource "openstack_containerinfra_cluster_v1" "k8s_126_1" {
  name = "paws${var.name[var.datacenter]}-126-1"
  cluster_template_id = resource.openstack_containerinfra_clustertemplate_v1.template_126_1.id
  master_count = 1
  node_count = var.workers[var.datacenter]
}

resource "openstack_containerinfra_clustertemplate_v1" "template_126_1" {
  name = "paws${var.name[var.datacenter]}-126-1"
  coe = "kubernetes"
  dns_nameserver = "8.8.8.8"
  docker_storage_driver = "overlay2"
  docker_volume_size = var.volume_size[var.datacenter]
  external_network_id = var.external_network_id[var.datacenter]
  fixed_subnet = var.fixed_subnet[var.datacenter]
  fixed_network = var.fixed_network[var.datacenter]
  flavor = var.worker_flavor[var.datacenter]
  floating_ip_enabled = "false"
  image = "Fedora-CoreOS-38-2"
  keypair_id = "pawsdev"
  master_flavor = var.control_flavor[var.datacenter]
  network_driver = "flannel"

  labels = {
    kube_tag = "v1.26.15-rancher1-linux-amd64"
    hyperkube_prefix = "docker.io/rancher/"
    cloud_provider_enabled = "true"
  }
}

```
My Fedora-CoreOS-38-2 image is a copy of fedora-coreos-38.20230806.3.0

[Vivian Rook (vrook)]

I found the labelThe following labels were needed
```
    container_runtime = "containerd"
    containerd_version = "1.6.20"
    containerd_tarball_sha256 = "1d86b534c7bba51b78a7eeb1b67dd2ac6c0edeb01c034cc5f590d5ccd824b416"
```
I found them in the documentation. This bug can be closed