Auto-restart slurmctld on failure after 1 second.

gmarciani · gmarciani · commit 951ee9e24c5c · 2024-11-26T11:52:59.000+01:00
Signed-off-by: Giacomo Marciani &lt;mgiacomo@amazon.com&gt;
diff --git a/cookbooks/aws-parallelcluster-slurm/templates/default/slurm/head_node/slurmctld.service.erb b/cookbooks/aws-parallelcluster-slurm/templates/default/slurm/head_node/slurmctld.service.erb
@@ -12,6 +12,8 @@ ExecReload=/bin/kill -HUP $MAINPID
 LimitNOFILE=562930
 LimitMEMLOCK=infinity
 LimitSTACK=infinity
+Restart=on-failure
+RestartSec=1s
 
 [Install]
 WantedBy=multi-user.target
