[gmx-users] Segmentation Fault - Water on Substrate

[Mark Abraham]

Hi,

You're presumably making one of your formerly periodic dimensions
differently periodic when you combine your boxes. If your input structure
has molecules with coordinates that are broken across the former boundary,
you need to make them whole before your merge can work sanely.

Mark

On Sun, Jun 26, 2016 at 11:45 AM Diez Fernandez, Amanda <
amanda.diez10 at imperial.ac.uk> wrote:

> Dear Gromacs users,
>
> I am want to simulate water on a  substrate.
> I have a water.pdb file with the coordinates of a cuboid of water and a
> substrate.pdb file with the coordinates of the substrate. The box
> dimensions along the x and y directions are identical for both  files.
> I run two simulations to minimise both structures separately. I get no
> error, meaning the topology file must be correct (is this a correct
> assumption?) This also ensures that all atoms in the minimised.pdb files
> are within the box I have specified  (as opposed to before, when some atoms
> were slightly outside).
> I then have a script to merge both minimised .pdb files.
> I minimise the system and get a segmentation fault:
>
> 40783 Segmentation fault
>
> Further, in the file.log I get:
>
>
> Steepest Descents:
>
>    Tolerance (Fmax)   =  6.00000e+01
>
>    Number of steps    =       100000
>
>            Step           Time         Lambda
>
>               0        0.00000        0.00000
>
>
>    Energies (kJ/mol)
>
>            Bond          Angle          LJ-14     Coulomb-14        LJ (SR)
>
>     1.36501e+04    6.58753e+04    4.22206e+04   -1.70332e+06           -nan
>
>    Coulomb (SR)   Coul. recip.      Potential Pressure (bar)
>
>     1.18196e+06    2.65781e+03           -nan           -nan
>
>
>
> I have visualised the input merged.pdb file and have also checked the
> minimum and maximum coordinates of water and substrate and I am sure there
> is no atom overlap. In fact, water and substrate atoms are separate by more
> than 3A.  ( I have also tried previously with smaller separations).
>
> What could else could I check?
>
> Thanks,
> Amanda
> --
> Gromacs Users mailing list
>
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
> posting!
>
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
> * For (un)subscribe requests visit
> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> send a mail to gmx-users-request at gromacs.org.
>