Compute-init: Optimise dir copies + Numerical sort playbook + new nodes to existing cluster

ansible/roles/compute_init/README.md

@@ -151,24 +151,25 @@ a new image:
 3. Add metadata to a compute node e.g. via Horizon to turn on compute-init
    playbook functionality.
 
-4. Fake an image build to deploy the compute-init playbook:
+4. Stop ansible-init from running
+
+        ansible all -ba "systemctl stop ansible-init"
+
+5. Fake an image build to deploy the compute-init playbook:
 
         ansible-playbook ansible/fatimage.yml --tags compute_init
 
     NB: This will also re-export the compute hostvars, as the nodes are not
     in the builder group, which conveniently means any changes made to that
     play also get picked up.
 
-5. Fake a reimage of compute to run ansible-init and the compute-init playbook:
-
-    On compute node where metadata was added:
+6. Fake a reimage of compute to run ansible-init and the updated compute-init playbook:
 
-        [root@rl9-compute-0 rocky]# rm -f /var/lib/ansible-init.done && systemctl restart ansible-init
-        [root@rl9-compute-0 rocky]# systemctl status ansible-init
+        ansible all -ba "rm -f /var/lib/ansible-init.done && systemctl restart ansible-init"
 
     Use `systemctl status ansible-init` to view stdout/stderr from Ansible.
 
-Steps 4/5 can be repeated with changes to the compute script. If required,
+Steps 4/5/6 can be repeated with changes to the compute script. If required,
 reimage the compute node(s) first as in step 2 and/or add additional metadata
 as in step 3.
 

ansible/roles/compute_init/files/compute-init.yml

@@ -95,6 +95,20 @@
         - meta: end_play
       when: _mount_mnt_cluster.failed
 
+    - name: Check if hostvars exist
+      stat:
+        path: "/mnt/cluster/hostvars/{{ ansible_hostname }}/hostvars.yml"
+      register: hostvars_stat
+
+    - block:
+        - name: Report skipping initialization if host vars does not exist
+          # meta: end_play produces no output
+          debug:
+            msg: "Skipping compute initialization: hostvars does not exist"
+
+        - meta: end_play
+      when: not hostvars_stat.stat.exists
+
     - name: Load hostvars from NFS
       # this is higher priority than vars block = normal ansible's hostvars
       include_vars:

ansible/roles/compute_init/tasks/install.yml

@@ -16,12 +16,14 @@
     - roles
 
 - name: Inject files from roles
-  copy:
+  synchronize:
     src: '{{ item.src }}'
     dest: '/etc/ansible-init/playbooks/{{ item.dest }}'
-    owner: root
-    group: root
-    mode: 0644
+    archive: false
+    rsync_opts: ["-p", "--chmod=D770,F644", "--owner=root", "--group=root"]
+    recursive: true
+    use_ssh_args: true
+  become: true
   loop:
     - src: ../../resolv_conf/templates/resolv.conf.j2
       dest: templates/resolv.conf.j2
@@ -60,7 +62,7 @@
 - name: Add compute initialisation playbook
   copy:
     src: compute-init.yml
-    dest: /etc/ansible-init/playbooks/1-compute-init.yml
+    dest: /etc/ansible-init/playbooks/10-compute-init.yml
     owner: root
     group: root
     mode: 0644

environments/.stackhpc/tofu/cluster_image.auto.tfvars.json

@@ -1,6 +1,6 @@
 {
     "cluster_image": {
-        "RL8": "openhpc-RL8-250305-1110-534ed276",
-        "RL9": "openhpc-RL9-250305-1110-534ed276"
+        "RL8": "openhpc-RL8-250311-1020-d05208bc",
+        "RL9": "openhpc-RL9-250311-1020-d05208bc"
     }
 }