<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body text="#000000" bgcolor="#FFFFFF">
Hello Arielle,<br>
<br>
I don't have a full answer, but here is a start:<br>
<br>
Yes, you first need at least <tt>"AccountingStorageEnforce=associations,limits</tt><tt>"
(and qos is you want to use it) so that the limits you set are
enforced (see <a class="moz-txt-link-freetext" href="https://slurm.schedmd.com/resource_limits.html">https://slurm.schedmd.com/resource_limits.html</a>)<br>
</tt><br>
Then you can set limits for each user eg with "sacctmgr modify
account where account=<account> set MaxJobs=50" to set the max
number of jobs for each of the users of this account (replace
<account> as needed). Then maybe set it for a single user:
"sacctmgr modify account where account=<account>
user=<user> set MaxJobs=500".<br>
<br>
You can check what is set with "sacctmgr show associations", and
remove what has been set by setting it to -1.<br>
<br>
I hope this helps,<br>
Colas<br>
<br>
<div class="moz-cite-prefix">On 2018-01-18 12:22, Arielle Willm
wrote:<br>
</div>
<blockquote type="cite"
cite="mid:b95b1918-bdde-722a-bd44-f74920630776@oca.eu">
<meta http-equiv="content-type" content="text/html; charset=utf-8">
<br>
<tt>Hi,</tt><tt><br>
</tt><tt> </tt><tt><br>
</tt><tt>slurm is installed in a minimal configuration for a
cluster of 3000cores/170 nodes.We have 4 partitions, one for
each type of nodes; each partition is available for all users.</tt><tt><br>
</tt><tt> </tt><tt>We want to prevent each user from taking more
than 1000 cores running on up to 50 jobs on all the cluster, and
I'm looking for guidance on how we should do so.</tt><tt><br>
</tt><tt> </tt><tt><br>
</tt><tt> </tt><tt>I have already read a lot of documentation
about it , i tried but i have no result</tt><tt><br>
</tt><tt> <br>
- Below a list of some config parameters : </tt>
<blockquote> <tt>ClusterName=cluster</tt><tt><br>
</tt> <tt># Accounting.</tt><tt><br>
</tt><tt> AccountingStorageType=accounting_storage/slurmdbd</tt><tt><br>
</tt><tt> AccountingStorageUser=slurm</tt><tt><br>
</tt><tt> #AccountingStorageEnforce=qos,limits</tt><tt><br>
</tt><tt> #AccountingStoragePass=</tt><tt><br>
</tt><tt> #AccountingStoragePort=</tt><tt><br>
</tt> <tt>EnforcePartLimits=YES</tt><tt><br>
</tt> <tt># SCHEDULING</tt><tt><br>
</tt><tt> SchedulerType=sched/backfill</tt><tt><br>
</tt><tt> # Consumable Resource</tt><tt><br>
</tt><tt> SelectType=select/cons_res</tt><tt><br>
</tt><tt> SelectTypeParameters=CR_Core</tt><br>
</blockquote>
<tt>- about the config in the slurmDB , there are not many things,
only account root and only user root</tt><tt><br>
</tt><tt> </tt><tt><br>
</tt><tt>sacctmgr show account</tt><tt><br>
</tt><tt> Account Descr Org </tt><tt><br>
</tt><tt> ---------- -------------------- -------------------- </tt><tt><br>
</tt><tt> root default root account root </tt><tt><br>
</tt><tt> </tt><tt><br>
</tt><tt>I have tried : sacctmgr modify cluster set
GrpTRES=cpu=1040</tt><tt><br>
</tt><tt> </tt><tt><br>
</tt><tt>but it does not work !</tt><tt><br>
</tt><tt> </tt><tt><br>
</tt><tt>What is wrong ?</tt><tt><br>
</tt><tt> </tt><tt><br>
</tt><tt> </tt><tt><br>
</tt><tt>Did we need to uncomment the parameter
AccountingStorageEnforce ?</tt><tt><br>
</tt><tt> </tt><tt><br>
</tt><tt>Did we need to define all the users ? <br>
<br>
Can someone show the way to proceed?</tt><tt><br>
</tt><tt> </tt><tt><br>
</tt><tt>Thanks for your help<br>
<br>
A Willm<br>
<br>
</tt> -- </blockquote>
<br>
</body>
</html>