Re: [AMBER] Error "forrtl: severe (174) : SIGSEGV, segmentation fault" while running MD simulation production run

From: Horn, Anselm <anselm.horn.fau.de>
Date: Wed, 28 Aug 2019 08:56:06 +0000

Dear Leena,

in the past, I observed a crash of MD-executables on our HPC systems unless I unlimited the stacksize in the shell.
I put the following in my bash-jobscript just before the invocation of pmemd:
  ulimit -s unlimited
Maybe this is helpful for you as well.

Regards,

Anselm

________________________________________
Von: Leena Aggarwal <leena.hrc.gmail.com>
Gesendet: Mittwoch, 28. August 2019 09:31:31
An: AMBER Mailing List
Betreff: Re: [AMBER] Error "forrtl: severe (174) : SIGSEGV, segmentation fault" while running MD simulation production run

I am running calculation on HPC resource. I am able to run simulation for
other systems in AMBER 18. I have tried both with sander.MPI as well as
with pmemd.MPI for this system but it is giving the error as I have
mentioned in the previous mail. Till NVT equilibration, it is running
absolutely fine, however it is giving the error for the NPT production run.

On Tue, Aug 27, 2019 at 8:43 PM Andrew Sugarman <axs1724.case.edu> wrote:

> Leena,
>
> Are you running calculations locally or on an HPC resource? I have
> encountered segmentation faults in the past when the instruction logs for
> CPUs have been updated or changed. From the output file, it looks like the
> executable 'sander.MPI' is not being recognized. Are you able to run any
> other AMBER18 executables in your current environment?
>
> Best,
> Andrew Sugarman
>
> On Tue, Aug 27, 2019 at 10:03 AM Leena Aggarwal <leena.hrc.gmail.com>
> wrote:
>
> > Dear Amber Users
> >
> > I am getting an error *forrtl: severe (174) : SIGSEGV, segmentation
> fault*
> > error while running the MD simulation production run in AMBER18.
> > The simulation is smoothly running without any error in AMBER 12 but
> > giving the above mentioned error in AMBER 18.
> >
> > For the reference, I have attached the logfile or error file.
> >
> > From
> > Leena Aggarwal
> > University of Delhi
> > India
> > _______________________________________________
> > AMBER mailing list
> > AMBER.ambermd.org
> > http://lists.ambermd.org/mailman/listinfo/amber
> >
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber

_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Aug 28 2019 - 02:00:02 PDT
Custom Search