
<html><body><div style="font-family: arial, helvetica, sans-serif; font-size: 12pt; color: #000000"><div>Hello,<br><br>I tried using, in slurm.conf<br>TaskPlugin=task/affinity, task/cgroup<br>SelectTypeParameters=CR_CPU_Memory<br>MemLimitEnforce=yes<br><br>and in cgroup.conf:<br>CgroupAutomount=yes<br>ConstrainCores=yes<br>ConstrainRAMSpace=yes<br>ConstrainSwapSpace=yes<br>MaxSwapPercent=10<br>TaskAffinity=no<br><br>But when the job reaches its limit, it passes alternately from R to D state without being killed, even when it exceeds the 10% of swap partition allowed.<br><br>Do you have an idea to do this?<br></div><div><br></div><div>Regards,<br data-mce-bogus="1"></div><div><br data-mce-bogus="1"></div><div>Jean-Mathieu<span style="font-size: 8pt; font-family: arial, helvetica, sans-serif;"></span></div></div></body></html>