[slurm-users] SLURM_JOB_IB for heterogeneous jobs

Hello, the docu for heterogeneous jobs [1] says that the envVar SLURM_JOB_ID should be different for each component. However, I cannot reproduce this on a fresh slurm-19.05.1 installation. $ salloc -pcompute -N1 : -pcompute2 -N1 [...] salloc: Granted job allocation 108453 [...] bash-4.1$ sque

Re: [slurm-users] allocate last MPI-rank to an exclusive node?

Hi, we had the same issue and solved it by using the 'plane' distribution in combination with MPMD style srun, e.g. in your example #SBATCH -N 3 # 3 nodes with 10 cores each #SBATCH -n 21 # 21 MPI-tasks in sum #SBATCH --cpus-per-task=1 # if you do not want hyperthreading cat > mpmd.conf <<

Re: [slurm-users] Checking memory requirements in job_submit.lua

Hi, based on information given in job_submit_lua.c we decided not to use pn_min_memory any more. The comment in src says: /* * FIXME: Remove this in the future, lua can't handle 64bit * numbers!!!. Use min_mem_per_node|cpu instead. */ Instead we check in job_submit.lua for s,th, like if

[slurm-users] sbatch option --propagate ignored

Hello, we recently updated from slurm 16.05.x to 17.11.5 and found that the sbatch option --propagate is no longer followed. Although written in the man pages the following does not modify the core file and stack size limits on the compute nodes #SBATCH --propagate=STACK,CORE but it can sti

Re: [slurm-users] Extreme long db upgrade 16.05.6 -> 17.11.3

st 15 minutes! Best, Hendryk -- Dr. Hendryk Bockelmann Wissenschaftliches Rechnen Abteilung Anwendungen Deutsches Klimarechenzentrum GmbH (DKRZ) Bundesstraße 45 a, D-20146 Hamburg, Germany smime.p7s Description: S/MIME Cryptographic Signature