Stop slurmd during slurmctld restart

Some Slurm configuration changes can cause compute nodes to go into
invalid state if slurmctld is restarted while slurmd services are still
running.

Stop slurmd services while slurmctld is being restarted. This has been
tested not to affect running jobs.

Closes #199

Author: priteau

tasks/runtime.yml

@@ -127,6 +127,23 @@
     - "_openhpc_slurmdbd_state.stdout == 'inactive'"
     - openhpc_enable.database | default(false)
 
+- name: Stop slurmd if configuration has changed
+  service:
+    name: "slurmd"
+    state: stopped
+  retries: 5
+  register: slurmd_stop
+  until: slurmd_stop is success
+  delay: 30
+  when:
+    - openhpc_slurm_service_started | bool
+    - openhpc_enable.batch | default(false) | bool
+    - openhpc_slurm_control_host in ansible_play_hosts
+    - hostvars[openhpc_slurm_control_host].ohpc_slurm_conf.changed or hostvars[openhpc_slurm_control_host].ohpc_gres_conf.changed # noqa no-handler
+
+- name: Flush handler
+  meta: flush_handlers # This will restart slurmctld while slurmd services are stopped, if needed
+
 - name: Notify handler for slurmd restart
   debug:
     msg: "notifying handlers" # meta: noop doesn't support 'when'