On Tue, Oct 2, 2012 at 1:51 AM, Anogna <anognal.gmail.com> wrote:
> Dear David,
>
> Sorry for not conveying my problem clearly.
>
> I submitted my complex for production phase. After period of time its
> getting stopped abnormally.
> As per I know if any error occurs Amber will print a file named "mdout"
> which shows all information about the error and "mdinfo" file which shows
> the step were the process stopped.
>
> But in my job....its not even generating any error output file. And more
> before going to production i faced the same problem while running
> equilibration. At first it got stopped abnormally after reaching
> 30,000/50,000 steps. Later when i ran the equil again using same input
> files as used in my first run....it didn't stop in middle of the process.
>
The output file (mdout) is always printed. It is not just printed if there
is an error. If there is an error, it will most likely be printed to the
output file (although it's a good idea to check stderr as well). If no
mdout file was printed, I would probably lean toward blaming the computers
you're running on, since opening the mdout file is one of the first things
that happens.
If you are using PBS, make sure you know what directory your "sander" or
"pmemd" command is being run from. (i.e., use a command like "cd
$PBS_O_WORKDIR" at the beginning of your script so you know the output file
will be placed in the directory you submitted the job from).
>
> Can you please tell me what would be the problem.
>
Not without the error message. There are too many things that could be
going wrong. If there really is no error message, we would need a way of
reproducing your issue in order to have a chance of determining what is
wrong.
HTH,
Jason
--
Jason M. Swails
Quantum Theory Project,
University of Florida
Ph.D. Candidate
352-392-4032
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Oct 02 2012 - 05:00:04 PDT
