Merge pull request #2269 from priti-parate/release_1.6

fix for slurmctld service which is failed to start after remove node
dell · May 22, 2024 · 3b030ad · 3b030ad
2 parents 5435bab + b04881c
commit 3b030ad
Show file tree

Hide file tree

Showing 2 changed files with 3 additions and 3 deletions.
diff --git a/utils/roles/remove_node/remove_kube_node/tasks/main.yml b/utils/roles/remove_node/remove_kube_node/tasks/main.yml
@@ -148,7 +148,7 @@
   ansible.builtin.set_fact:
     node_dict: "{{ node_dict | combine({item.item: {'admin_ip': item.query_result[0].admin_ip,
       'kube_override_hostname': (item.query_result[0].service_tag | upper) if (item.query_result[0].service_tag | upper) in kubectl_output.stdout
-         else ((item.query_result[0].service_tag | lower ) if (item.query_result[0].service_tag | lower) in kubectl_output.stdout
+         else ((item.query_result[0].service_tag | lower) if (item.query_result[0].service_tag | lower) in kubectl_output.stdout
          else (item.query_result[0].admin_ip if item.query_result[0].admin_ip in kubectl_output.stdout
          else (item.query_result[0].hostname if item.query_result[0].hostname in kubectl_output.stdout
          else (item.query_result[0].node if item.query_result[0].node in kubectl_output.stdout else '')))),

diff --git a/utils/roles/remove_node/remove_slurm_node/tasks/main.yml b/utils/roles/remove_node/remove_slurm_node/tasks/main.yml
@@ -89,8 +89,8 @@
 - name: Remove hostname from partition details of slurm.conf
   ansible.builtin.replace:
     path: '{{ conf_path }}'
-    regexp: '^(PartitionName=.*\s){{ node_hostname }}(\s.*)$'
-    replace: '\1\2'
+    regexp: '^(PartitionName=normal Nodes=.*?)({{ node_hostname }},|,{{ node_hostname }})(.*?)'
+    replace: '\1\3'
   failed_when: false
   delegate_to: "{{ groups['slurm_control_node'] | first }}"