[gmx-users] Re: segmentation fault with grompp

[Mark Abraham]

On 09/02/10 09:04, Justin A. Lemkul wrote:
> Quoting Gard Nelson<gardnersnake23 at gmail.com>:
>
>> Ok, here's what i've got:
>>
>> Gromacs version 4.0.5.
>> Both platforms are Linux, icc 10, fftw 3
>> compile option:  ./configure --enable-mpi
>>
>
> The vast majority of Gromacs programs (read: anything not called "mdrun") are
> not MPI-enabled.  Users have reported weird activity when using, for instance,
> mpicc to compile non-MPI code, although I don't know enough about that to
> specifically say that's the problem.  I would advise following the installation
> instructions on the Gromacs site exactly, using MPI-enable compilers for mdrun
> only, to see if that fixes your issue.

A quicker diagnostic may be whether grompp works on other simulation 
systems.

Also, inspect this ff_dum.itp for well-formedness. Try commenting out 
its #include and see what breaks then.

Mark

>> thanks for your help!
>> Gard
>>
>>
>> Gard Nelson wrote:
>>> Hi all,
>>> I'm trying to simulate a solvated membrane using Berger's lipids.  When
>>> I run grompp, i get the following error:
>>>
>>> checking input for internal consistency...
>>> processing topology...
>>> Opening library file ...lipid.itp
>>> Opening library file ...nb_lipid.itp
>>> Opening library file ...bon_lipid.itp
>>> Opening library file ...ff_dum.itp
>>> Segmentation fault (core dumped)
>>>
>>> I've tried this on two different machines with two different ff_dum.itp
>>> files with the same results. Moreover, I've never touched the ff_dum.itp
>>> file.  I have modified the other forcefield files, both to include
>>> Berger's lipids as well as other non-standard residues.  It runs fine
>>> with solvated proteins, but this is the first time I've run a system
>>> with the membrane, so i'm guessing the error is coming from there.  I've
>>> looked through my files and cant find any errors.  Does anyone have any
>>> ideas of what might be causing this?
>>>
>>
>> Lots of things.  But to get better help, you'll have to provide some more
>> information:
>>
>> 1. Gromacs version
>> 2. Platform
>> 3. How Gromacs was compiled (commands, compilers used, etc)
>>
>> -Justin
>>
>>> Thanks
>>> Gard Nelson
>>>
>>
>
>
>
> ========================================
>
> Justin A. Lemkul
> Graduate Research Assistant
> Department of Biochemistry
> Virginia Tech
> Blacksburg, VA
> jalemkul at vt.edu | (540) 231-9080
> http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin/
>
> ========================================