[slurm-users] possible to set memory slack space before killing jobs?

[Raymond Wan]

Hi,


On 7/12/2018 6:23 PM, Bjørn-Helge Mevik wrote:
> Raymond Wan <rwan.work at gmail.com> writes:
> 
>> However, a more general question... I thought there is no fool-proof
>> way to watch the amount of memory a job is using.  What if within the
>> script they ran another program using "nohup", for example.  Wouldn't
>> slurm be unable to include the memory usage of that program?
> 
> If you use jobacct_gather/cgroup, slurm (via cgroups) will count the
> memory of all processes in the cgroup, and I don't think a regular user
> is able to move any processes outside the current cgroup (or start a
> process outside from a process inside).


Oh...  I see.  I never looked into cgroups before.  I will 
read more about it if it helps me solve one of the problems 
that's been bugging me.

Thank you!

Ray