[gmx-users] No domain decomposition

[soumadwip ghosh]

Hi Justin and Tsjerk,
                                  thanks for your valuable suggestions.
Justin, I am attaching the structure of my system after the energy
minimization step i.e, as follows

https://drive.google.com/file/d/0B7SBnQ5YXQSLbEdhZEZKVUtJTHM/view?usp=sharing

I dont see any distortion in the structure. But when I proceed for position
restrained dynamics, I am getting the following error on issuing mdrun
-deffnm nvt

Step 910:
The charge group starting at atom 1539 moved more than the distance allowed
by the domain decomposition (3.333333) in direction Z
distance out of cell -8.503349
Old coordinates:    4.604    5.037    4.690
New coordinates:   -3.702    5.989   -5.170
Old cell boundaries in direction Z:    3.333    6.667
New cell boundaries in direction Z:    3.333    6.667

-------------------------------------------------------
Program mdrun, VERSION 4.5.6
Source code file: domdec.c, line: 4124

Fatal error:
A charge group moved too far between two domain decomposition steps


Then following Tsjerk's advice, I issued mdrun -deffnm nvt -nt 8 after
which I get the same error. I can't understand what's going wrong from the
.log file.

There are: 98940 Atoms
Charge group distribution at step 0: 4039 4205 4385 4423 4386 4270 4317 4241
Grid: 5 x 15 x 15 cells

Constraining the starting coordinates (step 0)

Constraining the coordinates at t0-dt (step 0)
RMS relative constraint deviation after constraining: 0.00e+00
Initial temperature: 397.477 K

Started mdrun on node 0 Thu Sep 24 08:52:44 2015

           Step           Time         Lambda
              0        0.00000        0.00000

   Energies (kJ/mol)
          Angle            U-B    Proper Dih.  Improper Dih.          LJ-14
    2.97004e+03    3.84714e+02    1.17214e+04    3.55064e+02    3.34059e+04
     Coulomb-14        LJ (SR)  Disper. corr.   Coulomb (SR)   Coul. recip.
   -1.60787e+03    2.88658e+05   -1.89372e+03   -2.01160e+06   -3.71978e+04
 Position Rest.      Potential    Kinetic En.   Total Energy  Conserved En.
    9.46864e+00   -1.71480e+06    2.87409e+05   -1.42739e+06   -1.42739e+06
    Temperature Pres. DC (bar) Pressure (bar)   Constr. rmsd
    3.50934e+02   -6.28921e+01   -4.78393e+05    2.81540e-02

DD  step 4 load imb.: force  5.6%

DD  step 499 load imb.: force  6.0%

           Step           Time         Lambda
            500        0.50000        0.00000

   Energies (kJ/mol)
          Angle            U-B    Proper Dih.  Improper Dih.          LJ-14
    2.15265e+04    4.81825e+03    2.24841e+04    5.26390e+02    7.79415e+04
     Coulomb-14        LJ (SR)  Disper. corr.   Coulomb (SR)   Coul. recip.
   -1.90445e+03    3.20893e+05   -1.89372e+03   -1.94686e+06   -3.66598e+04
 Position Rest.      Potential    Kinetic En.   Total Energy  Conserved En.
    2.39384e+04   -1.51519e+06    2.47287e+05   -1.26790e+06   -1.13929e+06
    Temperature Pres. DC (bar) Pressure (bar)   Constr. rmsd
    3.01944e+02   -6.28921e+01   -1.66734e+02    4.42351e-04


-------------------------------------------------------
Program mdrun, VERSION 4.5.6
Source code file: domdec.c, line: 4124

Fatal error:
A charge group moved too far between two domain decomposition steps
This usually means that your system is not well equilibrated
For more information and tips for troubleshooting, please check the GROMACS
website at http://www.gromacs.org/Documentation/Errors
-------------------------------------------------------

"Chemical gases filling lungs of little ones" (Black Eyed Peas)

If anybody knows how to settle this issue I will really appreciate it. As
of now I can't see a way out apart from setting up the system again. But I
am 100% sure that there was nothing wrong with the topologies.

Thanks and regards,
Soumadwip Ghosh
Research Fellow
IITB
Mumbai
India