Merge branch 'main' into feat/isolated-env-2

Author: sjpb

.github/workflows/extra.yml

@@ -32,11 +32,11 @@ jobs:
           - image_name: openhpc-extra-RL8
             source_image_name_key: RL8 # key into environments/.stackhpc/tofu/cluster_image.auto.tfvars.json
             inventory_groups: doca,cuda,lustre
-            volume_size: 30 # needed for cuda
+            volume_size: 35 # needed for cuda
           - image_name: openhpc-extra-RL9
             source_image_name_key: RL9
             inventory_groups: doca,cuda,lustre
-            volume_size: 30 # needed for cuda
+            volume_size: 35 # needed for cuda
     env:
       ANSIBLE_FORCE_COLOR: True
       OS_CLOUD: openstack

.gitignore

@@ -5,3 +5,4 @@ venv
 *.pyc
 packer/openhpc2
 .vscode
+requirements.yml.last

README.md

@@ -25,8 +25,6 @@ The default configuration in this repository may be used to create a cluster to
 - Persistent state backed by an OpenStack volume.
 - NFS-based shared file system backed by another OpenStack volume.
 
-Note that the Open OnDemand portal and its remote apps are not usable with this default configuration.
-
 It requires an OpenStack cloud, and an Ansible "deploy host" with access to that cloud.
 
 Before starting ensure that:

ansible/.gitignore

@@ -90,3 +90,7 @@ roles/*
 !roles/gateway/**
 !roles/alertmanager/
 !roles/alertmanager/**
+!roles/slurm_recompile/**
+!roles/slurm_recompile/**
+!roles/nhc/
+!roles/nhc/**

ansible/bootstrap.yml

@@ -134,6 +134,7 @@
 
 - hosts: dnf_repos
   become: yes
+  tags: dnf_repos
   tasks:
   - name: Check that creds won't be leaked to users
     ansible.builtin.assert:
@@ -143,7 +144,7 @@
       - appliances_mode == 'configure'
       - not (dnf_repos_allow_insecure_creds | default(false)) # useful for development
 
-- hosts: cacerts:!builder
+- hosts: cacerts
   tags: cacerts
   gather_facts: false
   tasks:

ansible/disable-repos.yml

@@ -49,6 +49,20 @@
         name: cuda
         tasks_from: "{{ 'runtime.yml' if appliances_mode == 'configure' else 'install.yml' }}"
 
+- name: Setup vGPU
+  hosts: vgpu
+  become: yes
+  gather_facts: yes
+  tags: vgpu
+  tasks:
+    - include_role:
+        name: stackhpc.linux.vgpu
+        tasks_from: "{{ 'configure.yml' if appliances_mode == 'configure' else 'install.yml' }}"
+  handlers:
+    - name: reboot
+      fail:
+        msg: Reboot handler for stackhpc.linux.vgpu role fired unexpectedly. This was supposed to be unreachable.
+
 - name: Persist hostkeys across rebuilds
   # Must be after filesystems.yml (for storage)
   # and before portal.yml (where OOD login node hostkeys are scanned)
@@ -59,17 +73,6 @@
     - import_role:
         name: persist_hostkeys
 
-
-- name: Setup NFS export for compute node configuration
-  hosts: compute_init:!builder
-  # NB: has to be after eeesi and os-manila-mount
-  tags: compute_init
-  become: yes
-  tasks:
-    - include_role:
-        name: compute_init
-        tasks_from: export.yml
-
 - name: Install k9s
   become: yes
   hosts: k9s

ansible/extras.yml

@@ -252,14 +252,24 @@
     - include_role: # done in same play so it can use handlers from cloudalchemy.grafana
         name: grafana-dashboards
 
+- name: Add support for NVIDIA GPU auto detection to Slurm
+  hosts: cuda
+  become: yes
+  tasks:
+    - name: Recompile slurm
+      import_role:
+        name: slurm_recompile
+      vars:
+        slurm_recompile_with_nvml: "{{ groups.cuda | length > 0 }}"
+
 - name: Run post.yml hook
   vars:
     appliances_environment_root: "{{ lookup('env', 'APPLIANCES_ENVIRONMENT_ROOT') }}"
     hook_path: "{{ appliances_environment_root }}/hooks/post.yml"
   import_playbook: "{{ hook_path if hook_path | exists else 'noop.yml' }}"
   when: hook_path | exists
 
-- import_playbook: disable-repos.yml
+- import_playbook: final.yml
 
 - hosts: builder
   become: yes

ansible/fatimage.yml

@@ -0,0 +1,18 @@
+- hosts: dnf_repos
+  become: yes
+  tags: dnf_repos
+  tasks:
+    - name: Disable pulp repos
+      ansible.builtin.include_role:
+        name: dnf_repos
+        tasks_from: disable_repos.yml
+
+- name: Setup NFS export for compute_init
+  hosts: compute_init:!builder
+  # NB: done last so other roles can prepare configuration etc
+  tags: compute_init
+  become: yes
+  tasks:
+    - include_role:
+        name: compute_init
+        tasks_from: export.yml

ansible/final.yml

@@ -4,7 +4,7 @@ Configure CA certificates and trusts.
 
 ## Role variables
 
-- `ca-certificates`: Optional str. Path to directory containing certificates
+- `cacerts_cert_dir`: Optional str. Path to directory containing certificates
   in PEM or DER format. Any files here will be added to the list of CAs trusted
   by the system.