refactor: rename k8s inventory groups for kubespray v2.30

kubespray v2.30 requires underscored group names:
- kube-master -> kube_control_plane
- kube-node -> kube_node
- k8s-cluster -> k8s_cluster

Updated inventory templates, group_vars filename, group_vars content,
and all playbook references. Directory paths (playbooks/k8s-cluster/)
are unchanged.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Signed-off-by: Douglas Holt <dholt@nvidia.com>

Author: dholt

config.example/group_vars/k8s-cluster.yml config.example/group_vars/k8s_cluster.ymlconfig.example/group_vars/k8s-cluster.yml renamed to config.example/group_vars/k8s_cluster.yml

@@ -44,7 +44,7 @@ dashboard_metrics_scrape_tagr: "v1.0.4"
 dashboard_metrics_scraper_repo: "kubernetesui/metrics-scraper"
 
 # Ensure hosts file generation only runs across k8s cluster
-hosts_add_ansible_managed_hosts_groups: ["k8s-cluster"]
+hosts_add_ansible_managed_hosts_groups: ["k8s_cluster"]
 
 # NFS Client Provisioner 
 # Playbook: nfs-client-provisioner.yml

config.example/inventory

@@ -19,7 +19,7 @@
 ######
 # KUBERNETES
 ######
-[kube-master]
+[kube_control_plane]
 #mgmt01
 #mgmt02
 #mgmt03
@@ -31,13 +31,13 @@
 #mgmt03
 
 # Also add mgmt/master nodes here if they will run non-control plane jobs
-[kube-node]
+[kube_node]
 #gpu01
 #gpu02
 
-[k8s-cluster:children]
-kube-master
-kube-node
+[k8s_cluster:children]
+kube_control_plane
+kube_node
 
 ######
 # SLURM

playbooks/bootstrap/bootstrap-openshift.yml

@@ -1,5 +1,5 @@
 ---
-- hosts: kube-master
+- hosts: kube_control_plane
   become: true
   tasks:
     - name: Install required Python OpenShift packages/libraries

playbooks/k8s-cluster.yml

@@ -68,7 +68,7 @@
   import_playbook: generic/hosts.yml
 
 - name: Set up a local cluster container registry
-  import_playbook: container/standalone-container-registry.yml hostlist=kube-master
+  import_playbook: container/standalone-container-registry.yml hostlist=kube_control_plane
   when: kube_enable_container_registry|default(false)
 
 # Install 'sshpass' program for: https://github.com/ansible/ansible/issues/56629
@@ -110,7 +110,7 @@
   environment: "{{proxy_env if proxy_env is defined else {}}}"
 
 # Install Kubernetes
-# for configuration, see: config/group_vars/k8s-cluster.yml
+# for configuration, see: config/group_vars/k8s_cluster.yml
 - name: Install Kubernetes
   import_playbook: ../submodules/kubespray/cluster.yml
   tags:
@@ -128,7 +128,7 @@
     - swap
 
 # Manage Kubernetes cluster access config file
-- hosts: k8s-cluster
+- hosts: k8s_cluster
   gather_facts: false
   vars:
     ansible_become: no
@@ -165,7 +165,7 @@
 - name: Install NVIDIA driver on GPU servers
   import_playbook: nvidia-software/nvidia-driver.yml
   vars:
-    hostlist: "k8s-cluster"
+    hostlist: "k8s_cluster"
   tags:
     - nvidia
   when: deepops_gpu_operator_enabled|default(true) | bool == false or
@@ -174,7 +174,7 @@
 - name: Install NVIDIA container runtime on GPU servers
   import_playbook: container/nvidia-docker.yml
   vars:
-    hostlist: "k8s-cluster"
+    hostlist: "k8s_cluster"
   tags:
     - nvidia
   when:
@@ -183,7 +183,7 @@
     - container_manager is defined and container_manager == "docker"
 
 # Manage kubectl binary
-- hosts: kube-master
+- hosts: kube_control_plane
   gather_facts: false
   vars:
     ansible_become: no
@@ -196,7 +196,7 @@
       run_once: true
   tags:
     - local
-- hosts: k8s-cluster
+- hosts: k8s_cluster
   gather_facts: false
   vars:
     config_dir: "../config"
@@ -252,9 +252,9 @@
   tags:
     - local
 
-# Remove taint from kube-master nodes.
+# Remove taint from kube_control_plane nodes.
 # This keeps backwards compatibility and allows a few services (monitoring/etc.) to run properly.
-- hosts: kube-master
+- hosts: kube_control_plane
   gather_facts: false
   vars:
     ansible_become: no
@@ -268,7 +268,7 @@
     - name: kubeadm | Remove taint for master with node role
       command: "{{ artifacts_dir }}/kubectl --kubeconfig {{ artifacts_dir }}/admin.conf taint node {{ inventory_hostname }} node-role.kubernetes.io/control-plane:NoSchedule-"
       delegate_to: localhost
-      failed_when: false # Taint will not be present if kube-master also under kube-node
+      failed_when: false # Taint will not be present if kube_control_plane also under kube_node
 
 - name: Install k8s GPU feature discovery
   import_playbook: k8s-cluster/nvidia-k8s-gpu-feature-discovery.yml
@@ -295,11 +295,11 @@
 - name: Setup rsyslog server
   import_playbook: generic/rsyslog-server.yml
   vars:
-    hostlist: "{{ rsyslog_server_hostname | default('kube-master[0]') }}"
+    hostlist: "{{ rsyslog_server_hostname | default('kube_control_plane[0]') }}"
   when: kube_enable_rsyslog_server|default(true)
 
 - name: Setup rsyslog client
   import_playbook: generic/rsyslog-client.yml
   vars:
-    hostlist: "{{ rsyslog_client_group | default('k8s-cluster') }}"
+    hostlist: "{{ rsyslog_client_group | default('k8s_cluster') }}"
   when: kube_enable_rsyslog_client|default(true)

playbooks/k8s-cluster/container-registry.yml

@@ -1,5 +1,5 @@
 ---
-- hosts: kube-master
+- hosts: kube_control_plane
   become: true
   tasks:
     - name: Install helm chart for container registry
@@ -9,14 +9,14 @@
   tags:
     - container-registry
 
-- hosts: kube-node
+- hosts: kube_node
   become: true
   vars:
     container_registry_hostname: registry.local
   tasks:
     - name: Set registry hostname in /etc/hosts
       lineinfile:
         path: /etc/hosts
-        line: "{{ hostvars[groups['kube-master'][0]]['ansible_host'] | default(hostvars[groups['kube-master'][0]]['ansible_default_ipv4']['address']) }} {{ container_registry_hostname }}"
+        line: "{{ hostvars[groups['kube_control_plane'][0]]['ansible_host'] | default(hostvars[groups['kube_control_plane'][0]]['ansible_default_ipv4']['address']) }} {{ container_registry_hostname }}"
   tags:
     - container-registry

playbooks/k8s-cluster/netapp-trident.yml

@@ -2,7 +2,7 @@
 # Playbook for deploying NetApp Trident
 
 - name: "Install NFS utils on worker nodes"
-  hosts: kube-node
+  hosts: kube_node
   become: true
   become_method: sudo
   tasks:
@@ -16,7 +16,7 @@
     when: ansible_os_family == "RedHat"
 
 - name: "Deploy NetApp Trident"
-  hosts: kube-master
+  hosts: kube_control_plane
   become: true
   vars_files:
   - ../../config/group_vars/netapp-trident.yml

playbooks/k8s-cluster/nfs-client-provisioner.yml

@@ -1,15 +1,15 @@
 ---
 # This playbook is used to setup the nfs-client-provisioner
 # If the cluster already has an nfs server setup run with --skip-tags="nfs_server,nfs_mkdir"
-#   Otherwhise an NFS server will be configured on kube-master[0] 
+#   Otherwhise an NFS server will be configured on kube_control_plane[0] 
 # After settting up an NFS server, nfs software is installed on all nodes
 # Lastly, the nfs-client-provisioner is helm installed using the role
 #
 # The default NFS configuration of this playbook is not meant for production systems
 #   For production deployments please configur NFS with storage-partner guidance
 #
 # Install the required NFS software on all nodes and then helm install the nfs-client 
-- hosts: kube-master[0]
+- hosts: kube_control_plane[0]
   become: yes
   tasks:
   - name: Make sure NFS directory exists and is globally writable
@@ -25,14 +25,14 @@
     - nfs_is_server: yes
     when: k8s_deploy_nfs_server
 
-- hosts: "k8s-cluster"
+- hosts: "k8s_cluster"
   become: yes
   roles:
     - nfs
   tags:
     - nfs_software
 
-- hosts: kube-master[0]
+- hosts: kube_control_plane[0]
   become: true
   tasks:
     - name: install nfs-client-provisioner

playbooks/k8s-cluster/nvidia-gpu-operator.yml

@@ -3,7 +3,7 @@
   import_playbook: ../bootstrap/bootstrap-openshift.yml
 
 # GPU operator
-- hosts: kube-master[0]
+- hosts: kube_control_plane[0]
   become: yes
   tasks:
     - name: Install helm chart for GPU operator

playbooks/k8s-cluster/nvidia-k8s-gpu-device-plugin.yml

@@ -1,5 +1,5 @@
 ---
-- hosts: kube-master[0]
+- hosts: kube_control_plane[0]
   become: true
   tasks:
     - name: install k8s GPU plugin

playbooks/k8s-cluster/nvidia-k8s-gpu-feature-discovery.yml

@@ -1,5 +1,5 @@
 ---
-- hosts: kube-master[0]
+- hosts: kube_control_plane[0]
   become: true
   tasks:
     - name: install k8s GPU feature discovery