Updating from Rocky8 plain images to Rocky9 (#5736)

Author: sudheer-quad

community/examples/hpc-build-slurm-image.yaml

@@ -24,7 +24,9 @@ vars:
   zone: us-central1-a
 
   image_build_machine_type: n2d-standard-16
-  build_from_image_family: hpc-rocky-linux-8
+  # Rocky 8: hpc-rocky-linux-8
+  # Rocky 9: hpc-rocky-linux-9
+  build_from_image_family: hpc-rocky-linux-9
   build_from_image_project: cloud-hpc-image-public
   build_from_git_ref: 6.12.1
   built_image_family: my-custom-slurm
@@ -49,15 +51,17 @@ deployment_groups:
         content: |
           #!/bin/bash
           set -e -o pipefail
-          # Slurm build on Rocky8 will upgrade to python3.12 as part of build
-          dnf install -y python3.12 python3.12-pip
-          alternatives --set python3 /usr/bin/python3.12
-          python3 -m pip install pip --upgrade
-          python3 -m pip install ansible==8.7.0
-          python3 -m pip install selinux
+          # Slurm build will install and set python3.12 as the default python3
+          dnf install -y python3.12 python3.12-pip perl-ExtUtils-MakeMaker perl-devel
+          python3.12 -m pip install pip --upgrade
+          python3.12 -m pip install ansible==8.7.0
           export PATH=/usr/local/bin:$PATH
           ansible --version
-          ansible-galaxy role install googlecloudplatform.google_cloud_ops_agents
+          n=0
+          until [ "$n" -ge 3 ] || ansible-galaxy role install googlecloudplatform.google_cloud_ops_agents; do
+            ((n=n+1))
+            sleep 5
+          done
       - type: data
         destination: /var/tmp/slurm_vars.json
         content: |
@@ -67,7 +71,8 @@ deployment_groups:
             "nvidia_version": "latest",
             "install_ompi": true,
             "install_lustre": false,
-            "install_gcsfuse": true
+            "install_gcsfuse": true,
+            "monitoring_agent": "cloud-ops"
           }
       - type: shell
         destination: install_slurm.sh

community/examples/hpc-slurm6-apptainer.yaml

@@ -63,7 +63,9 @@ deployment_groups:
     settings:
       source_image_project_id: [schedmd-slurm-public]
       # see latest in https://github.com/GoogleCloudPlatform/slurm-gcp/blob/master/docs/images.md#published-image-family
-      source_image_family: slurm-gcp-6-12-hpc-rocky-linux-8
+      # Rocky 8: slurm-gcp-6-12-hpc-rocky-linux-8
+      # Rocky 9: slurm-gcp-6-12-hpc-rocky-linux-9
+      source_image_family: slurm-gcp-6-12-hpc-rocky-linux-9
       # You can find size of source image by using following command
       # gcloud compute images describe-from-family <source_image_family> --project schedmd-slurm-public
       disk_size: $(vars.disk_size)

examples/README.md

@@ -878,7 +878,7 @@ For this example the following is needed in the selected region:
 ### [hpc-build-slurm-image.yaml] ![community-badge] ![experimental-badge]
 
 This blueprint demonstrates how to use Cluster Toolkit to build a Slurm image on top
-of an existing image, `hpc-rocky-linux-8` in the case of this example.
+of an existing image, `hpc-rocky-linux-9` in the case of this example.
 
 The blueprint contains 3 groups: