account for transitory down state in ci/check_slurm.yml

Author: bertiethorpe

ansible/ci/check_slurm.yml

@@ -6,7 +6,7 @@
       shell: 'sinfo --noheader --format="%N %P %a %l %D %t" | sort' # using --format ensures we control whitespace: Partition,partition_state,max_jobtime,num_nodes,node_state,node_name
       register: sinfo
       changed_when: false
-      until: not ("boot" in sinfo.stdout or "idle*" in sinfo.stdout)
+      until: not ("boot" in sinfo.stdout or "idle*" in sinfo.stdout or "down" in sinfo.stdout)
       retries: 10
       delay: 5
     - name: Check nodes have expected slurm state