[gmx-users] Re: mpispawn.c:303 Unexpected exit status

Justin A. Lemkul jalemkul at vt.edu
Tue Dec 8 02:43:44 CET 2009

Please keep all Gromacs-related correspondence on the gmx-users list.  I am not 
a private help service.  Furthermore, you stand a better chance of getting a reply.

shivkumar bale wrote:
> Hi Justin,
> The error:
> mpispawn.c:303 Unexpected exit status
> Child exited abnormally!
> Killing remote processes...DONE
> Following is the link which I was talking about:
> http://www.ece.unm.edu/~etanner/MPI.html

This article is talking about memory usage, which I doubt is relevant.  The 
article is talking about hundreds of grid points, which may or may not be 
pertinent to your system.

> I think the above error is because of the PME calculations and changing 
> the grid spacing might work. But I dont know how to change the grid 
> spacing. Do you have any idea regarding that? Or if you think the error 
> is due to some other reason let me know.

I think several of your .mdp settings are inappropriate, and should be fixed 
before you make any conclusions like this.

1. define = -DFLEXIBLE is inappropriate, per the link I cited previously.
2. pbc = full is for infinite systems (carbon nanotubes, graphene, polymers, 
etc) so if your system is a normal bilayer, as your title suggests, this setting 
is wrong.

To fix grid spacing, adjust the fourierspacing parameter.  But address the two 
points above before drawing any conclusions that are more remote.

Also, see the following in general, in case your system is simply crashing 
because it is unstable.  Gromacs rarely exits without an informative error message.



> I am attaching the .mdp file with this email.
> Thanks for the help.
> Regards
> Shivkumar


Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080


More information about the gromacs.org_gmx-users mailing list