PoC of automating partition/nodegroup config

Author: sjpb

environments/.stackhpc/inventory/group_vars/all/z_partitions.yml

@@ -15,8 +15,21 @@ openhpc_slurmdbd_mysql_password: "{{ vault_mysql_slurm_password }}"
 openhpc_slurmdbd_mysql_username: slurm
 openhpc_slurm_control_host: "{{ groups['control'] | first }}" # avoid using hostvars for compute-init
 openhpc_slurmdbd_host: "{{ openhpc_slurm_control_host }}"
-openhpc_slurm_partitions:
-  - name: "compute"
+openhpc_rebuild_partition: # not a role var - could actually add more indirection here for things we're expecting to be modified, e.g. groups and maxtime
+  name: rebuild
+  groups: "{{ cluster_compute_groups }}"
+  default: NO
+  maxtime: 30
+  partition_params:
+    PriorityJobFactor: 65533
+    Hidden: YES
+    RootOnly: YES
+    DisableRootJobs: NO
+    PreemptMode: 'OFF'
+    OverSubscribe: EXCLUSIVE
+openhpc_nodegroups: "{{ cluster_compute_groups | map('community.general.dict_kv', 'name') }}" # create nodegroup for each compute group
+openhpc_user_partitions: "{{ openhpc_nodegroups }}" # create partition for each nodegroup (actually role default) - this is what we'd expect to be changed
+openhpc_partitions: "{{ openhpc_user_partitions + [openhpc_rebuild_partition] if groups['rebuild'] | length > 0 else [] }}" # auto-create rebuild partition if reqd.
 openhpc_packages_default:
   # system packages
   - podman

environments/common/inventory/group_vars/all/openhpc.yml

@@ -11,12 +11,3 @@ resource "local_file" "hosts" {
                           )
   filename = "../inventory/hosts.yml"
 }
-
-resource "local_file" "partitions" {
-    content  = templatefile("${path.module}/partitions.tpl",
-                            {
-                              "compute_groups": module.compute,
-                            },
-    )
-    filename = "../inventory/group_vars/all/partitions.yml" # as all/ is created by skeleton
-}

environments/skeleton/{{cookiecutter.environment}}/tofu/inventory.tf

@@ -2,6 +2,7 @@ all:
     vars:
         openhpc_cluster_name: ${cluster_name}
         cluster_domain_suffix: ${cluster_domain_suffix}
+        cluster_compute_groups: ${jsonencode(keys(compute_groups))}
 
 control:
     hosts:

environments/skeleton/{{cookiecutter.environment}}/tofu/inventory.tpl

@@ -4,7 +4,7 @@ roles:
     version: v25.3.2
     name: stackhpc.nfs
   - src: https://github.com/stackhpc/ansible-role-openhpc.git
-    version: v0.28.0
+    version: feat/nodegroups # TODO: bump to release
     name: stackhpc.openhpc
   - src: https://github.com/stackhpc/ansible-node-exporter.git
     version: stackhpc