Hi Bjørn,
Thank you very much for your help. That's right, we use IntelMPI.
At the moment, we are telling users to ignore the warning. But we
think it's a Slurm bug, so we've reported it, in case the developers
want to take a look at it.
Regards,
Ginés
El mié, 17 nov 2021 a las 5:13, Bjørn-Hel
Hi,
I encountered an issue a while back when Slurm would expect a cgroups v1
hierarchy, but Debian 11 had switched to cgroups v2.
I don't know whether this has been tackled in Slurm since then, but I
worked around it by adding "systemd.unified_cgroup_hierarchy=0" to the
kernel command line t
On Wed, 17 Nov 2021, Bjørn-Helge Mevik wrote:
...
We are using basically the same setup, and have not found any other way
than running "scontrol show job ..." in the prolog (even though it is
not recommended). I have yet to see any problems arising from it, but
YMMW.
If you find a different way
Hi Crhis,
That makes sense, thanks a lot !
Arthur
De: "Christopher Samuel"
À: "slurm-users"
Envoyé: Mercredi 17 Novembre 2021 02:36:36
Objet: Re: [slurm-users] Can't use cgroups on debian 11 : unable to get
parameter 'tasks' for '/sys/fs/cgroup/cpuset/'
On 11/16/21 8:04 am, Arthur Tous
Mark Dixon writes:
> Unfortunately, I've not found anything in the Prolog environment (or
> stored on disk under /var/spool/slurmd) containing the gres
> allocations for the job.
[...]
> Is there a better way to get the job's gres information from within
> the prolog, please?
We are using basi
Ginés Guerrero writes:
> Hi,
>
> If I submit this script:
>
> #!/bin/bash
> #SBATCH --get-user-env
> #SBATCH -p slims
> #SBATCH -N 2
> #SBATCH -n 40
> #SBATCH --ntasks-per-node=20
> #SBATCH -o log
> #SBATCH -e log
>
> /bin/env
>
> srun hostname
>
> I get th
