Merge branch 'main' into feat/auto-bump-timestamps

Author: wtripp180901

.github/workflows/doca.yml renamed to .github/workflows/extra.yml

@@ -1,4 +1,4 @@
-name: Test DOCA extra build
+name: Test extra build
 on:
   workflow_dispatch:
   push:
@@ -7,16 +7,18 @@ on:
     paths:
       - 'environments/.stackhpc/terraform/cluster_image.auto.tfvars.json'
       - 'ansible/roles/doca/**'
-      - '.github/workflows/doca'
+      - 'ansible/roles/cuda/**'
+      - '.github/workflows/extra.yml'
   pull_request:
     paths:
       - 'environments/.stackhpc/terraform/cluster_image.auto.tfvars.json'
       - 'ansible/roles/doca/**'
-      - '.github/workflows/doca'
+      - 'ansible/roles/cuda/**'
+      - '.github/workflows/extra.yml'
 
 jobs:
   doca:
-    name: doca-build
+    name: extra-build
     concurrency:
       group: ${{ github.workflow }}-${{ github.ref }}-${{ matrix.build.image_name }} # to branch/PR + OS
       cancel-in-progress: true
@@ -25,12 +27,14 @@ jobs:
       fail-fast: false # allow other matrix jobs to continue even if one fails
       matrix: # build RL8, RL9
         build:
-          - image_name: openhpc-doca-RL8
+          - image_name: openhpc-extra-RL8
             source_image_name_key: RL8 # key into environments/.stackhpc/terraform/cluster_image.auto.tfvars.json
-            inventory_groups: doca
-          - image_name: openhpc-doca-RL9
+            inventory_groups: doca,cuda
+            volume_size: 30 # needed for cuda
+          - image_name: openhpc-extra-RL9
             source_image_name_key: RL9
-            inventory_groups: doca
+            inventory_groups: doca,cuda
+            volume_size: 30 # needed for cuda
     env:
       ANSIBLE_FORCE_COLOR: True
       OS_CLOUD: openstack
@@ -95,6 +99,7 @@ jobs:
           -var "source_image_name=${{ fromJSON(env.FAT_IMAGES)['cluster_image'][matrix.build.source_image_name_key] }}" \
           -var "image_name=${{ matrix.build.image_name }}" \
           -var "inventory_groups=${{ matrix.build.inventory_groups }}" \
+          -var "volume_size=${{ matrix.build.volume_size }}" \
           openstack.pkr.hcl
         
       - name: Get created image names from manifest

.github/workflows/fatimage.yml

@@ -38,10 +38,10 @@ jobs:
       matrix: # build RL8, RL9
         build:
           - image_name: openhpc-RL8
-            source_image_name: Rocky-8-GenericCloud-Base-8.9-20231119.0.x86_64.qcow2
+            source_image_name: Rocky-8-GenericCloud-Base-8.10-20240528.0.x86_64.qcow2
             inventory_groups: control,compute,login,update
           - image_name: openhpc-RL9
-            source_image_name: Rocky-9-GenericCloud-Base-9.4-20240523.0.x86_64.qcow2
+            source_image_name: Rocky-9-GenericCloud-Base-9.5-20241118.0.x86_64.qcow2
             inventory_groups: control,compute,login,update
     env:
       ANSIBLE_FORCE_COLOR: True

.github/workflows/nightlybuild.yml

@@ -25,7 +25,7 @@ jobs:
       matrix: # build RL8, RL9
         build:
           - image_name: rocky-latest-RL8
-            source_image_name: Rocky-8-GenericCloud-Base-8.9-20231119.0.x86_64.qcow2
+            source_image_name: Rocky-8-GenericCloud-Base-8.10-20240528.0.x86_64.qcow2
             inventory_groups: update
           - image_name: rocky-latest-RL9
             source_image_name: Rocky-9-GenericCloud-Base-9.4-20240523.0.x86_64.qcow2

.github/workflows/trivyscan.yml

@@ -25,6 +25,20 @@ jobs:
     steps:
       - uses: actions/checkout@v2
 
+      - name: Override CI_CLOUD if PR label is present
+        if: ${{ github.event_name == 'pull_request' }}
+        run: |
+          # Iterate over the labels
+          labels=$(echo '${{ toJSON(github.event.pull_request.labels) }}' | jq -r '.[].name')
+          echo $labels
+          for label in $labels; do
+             if [[ $label == CI_CLOUD=* ]]; then
+              # Extract the value after 'CI_CLOUD='
+              CI_CLOUD_OVERRIDE=${label#CI_CLOUD=}
+              echo "CI_CLOUD=${CI_CLOUD_OVERRIDE}" >> $GITHUB_ENV
+            fi
+          done
+
       - name: Record settings for CI cloud
         run: |
           echo CI_CLOUD: ${{ env.CI_CLOUD }}

ansible/.gitignore

@@ -58,6 +58,8 @@ roles/*
 !roles/squid/**
 !roles/tuned/
 !roles/tuned/**
+!roles/compute_init/
+!roles/compute_init/**
 !roles/k3s/
 !roles/k3s/**
 !roles/k9s/

ansible/adhoc/sync-pulp.yml

@@ -6,5 +6,5 @@
       vars:
         pulp_site_target_arch: "x86_64"
         pulp_site_target_distribution: "rocky"
-        pulp_site_target_distribution_version: "9.4"
+        pulp_site_target_distribution_version: "9.5"
         pulp_site_target_distribution_version_major: "9"

ansible/bootstrap.yml

@@ -122,7 +122,6 @@
     ansible.builtin.include_role:
       name: dnf_repos
       tasks_from: set_repos.yml
-    when: ansible_distribution_major_version == "9" #TODO update role once RL8 config decided
 
 # --- tasks after here require access to package repos ---
 - hosts: squid

ansible/cleanup.yml

@@ -66,5 +66,4 @@
       slurm-ohpc: "{{ ansible_facts.packages['slurm-ohpc'].0.version | default('-') }}"
 
 - name: Show image summary
-  debug:
-    var: image_info
+  command: cat /var/lib/image/image.json

ansible/disable-repos.yml

@@ -5,4 +5,3 @@
       ansible.builtin.include_role:
         name: dnf_repos
         tasks_from: disable_repos.yml
-      when: ansible_distribution_major_version == "9" #TODO update role once RL8 config decided

ansible/extras.yml

@@ -24,8 +24,9 @@
   gather_facts: yes
   tags: cuda
   tasks:
-    - import_role:
+    - include_role:
         name: cuda
+        tasks_from: "{{ 'runtime.yml' if appliances_mode == 'configure' else 'install.yml' }}"
 
 - name: Persist hostkeys across rebuilds
   # Must be after filesystems.yml (for storage)
@@ -37,10 +38,32 @@
     - 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
+  name: Export hostvars
+  tasks:
+    - include_role:
+        name: compute_init
+        tasks_from: export.yml
+
 - name: Install k9s
   become: yes
   hosts: k9s
   tags: k9s
   tasks:
   - import_role:
       name: k9s
+
+- hosts: extra_packages
+  become: yes
+  tags:
+   - extra_packages
+  tasks:
+  - name: Install additional packages
+    dnf:
+      name: "{{ appliances_extra_packages }}"
+    when: appliances_mode != 'configure' or appliances_extra_packages_during_configure