Le 26/05/2026 à 11:13, Anders Frederiksen via AMBER a écrit :
> srun $AMBERHOME/bin/sander.MPI -O -i colvar.mdin \
Dear all,
Sorry to hijack the thread (I'm not expert on QM/MM) but I would
recommend AGAINST srun,
since at first it seems interesting, but SLURM with override your
environment with many variables,
and the problem you describe seems exactly to be this. Try using a
classical mpirun -np $NB_PROC
(or a similar variable) and see if the problem persist.
In my script I have a specific "env" command call, just to ensure the
complete environement where
the job is run will not contain garbage transferred to sander, pmed, and
others ...
HTH,
Stéphane
--
Maître de conférences Hors Classe, PEDR, HDR
US2B, Nantes Université, CNRS, UMR 6286, Team Structural Bioinformatics, F-44000 Nantes
http://www.us2b.univ-nantes.fr/ - http://www.steletch.org
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue May 26 2026 - 09:00:02 PDT
