[slurm-users] MaxMemPerCPU not enforced?

[Angel de Vicente]

Hello,

Angel de Vicente <angel.de.vicente at iac.es> writes:

> From my limited tests today, somehow in the interactive queue all seems
> OK now, but not so in the 'batch' queue. For example, I just submitted
> three jobs with different amount of CPUs per job (4, 8 and 16 processes
> respectively). MaxMemPerCPU is set to 2GB, and these jobs run the
> 'stress' command, consuming 3GB per process.

OK, this was again me reading the documentation too quickly, as this is
the expected behaviour as per the NOTE in ConstrainRAMSpace
(cgroup.conf).

[I was led off track by the fact that it is behaving differently in
interactive and batch modes: in the interactive tests all the processes
were killed, while in batch mode only some of the processes get the
ax]. 

Sorry for the noise,
-- 
Ángel de Vicente
 Research Software Engineer (Supercomputing and BigData)
 Tel.: +34 922-605-747
 Web.: http://research.iac.es/proyecto/polmag/

 GPG: 0x8BDC390B69033F52
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 5877 bytes
Desc: not available
URL: <http://lists.schedmd.com/pipermail/slurm-users/attachments/20230725/60436550/attachment-0001.bin>