configure openhpc for slurmdbd backup/update

Author: sjpb

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

@@ -58,7 +58,12 @@ ohpc_default_extra_repos:
   "9": []
   "8": []
 
-# systemd.service.unit.TimeoutStartSec to wait for slurmdbd startup
-# Set long enought to avoid problems with a major version upgrade
-# Currently implemented in environments/common/inventory/group_vars/all/systemd.yml
-openhpc_slurmdbd_timeout_start_sec: '45 minutes'
+# configure slurm database pre-upgrade backups:
+openhpc_slurm_accounting_storage_service: mysql
+openhpc_slurm_accounting_storage_backup_cmd: >-
+  openstack volume snapshot create
+  --volume {{ openhpc_cluster_name }}-state
+  --force
+  {{ openhpc_cluster_name }}-state-{{ ansible_date_time.iso8601_basic_short }}
+openhpc_slurm_accounting_storage_backup_host: localhost
+openhpc_slurm_accounting_storage_backup_become: false

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

@@ -14,13 +14,7 @@ systemd_dropins:
     content: "{{ _systemd_requiresmount_statedir }}"
   slurmdbd:
     group: openhpc
-    content: |
-      {{ _systemd_requiresmount_statedir }}
-      
-      [Service]
-      # Allow slurmdbd to complete major version upgrades
-      TimeoutStartSec={{ openhpc_slurmdbd_timeout_start_sec }}
-
+    content: "{{ _systemd_requiresmount_statedir }}"
   slurmctld:
     group: openhpc
     content: "{{ _systemd_requiresmount_statedir }}"

requirements.yml

@@ -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/upgrade-db' # local - TODO: bump on release
     name: stackhpc.openhpc
   - src: https://github.com/stackhpc/ansible-node-exporter.git
     version: stackhpc