[gmx-users] mdrun segmentation fault

[Jim Fonseca]

Hi,
	I'm having a problem with mdrun seg-faulting after ~240 ps of  
unrestrained simulation.  I'm using 3.3.1 on a beowulf cluster.   
Below are the tails of the output files.  I've also listed the .mdp  
at the end.

I've tried varying some of the parameters(dt and tau_p), but the  
simulations always seem to run to about 200 ps before crashing.  How  
can I find out more information about what's going on here?  The EM  
(run to machine convergence) and PR (100 ps) always run fine.  It  
seems to me that if it was an equilibration problem, it'd happen  
before running for so long.


Thanks for any help!
Jim Fonseca



*log from parallel node 0*
            Step           Time         Lambda
          120000      240.00000        0.00000

    Rel. Constraint Deviation:  Max    between atoms     RMS
        Before LINCS         0.067432   7188   7189   0.005393
         After LINCS         0.000131   8250   8253   0.000021

    Energies (kJ/mol)
           Angle    Proper Dih. Ryckaert-Bell.  Improper  
Dih.          LJ-14
     7.02762e+04    2.34867e+04    2.53603e+04    7.54128e+03     
2.46027e+04
      Coulomb-14        LJ (SR)   Coulomb (SR)   Coul. recip.       
Potential
     1.54268e+05    2.16502e+05   -2.96632e+06   -6.02749e+05    
-3.04703e+06
     Kinetic En.   Total Energy    Temperature Pressure (bar)
     5.09231e+05   -2.53780e+06    2.99996e+02    3.08941e+01

*output redirect fom mdrun*
^Mstep 120180, will finish at Wed Jun 28 13:04:09 2006
^Mstep 120190, will finish at Wed Jun 28 13:04:13 2006
^Mstep 120200, will finish at Wed Jun 28 13:04:17 2006

Back Off! I just backed up step120205.pdb to ./#step120205.pdb.1#

Back Off! I just backed up step120206.pdb to ./#step120206.pdb.1#
Wrote pdb files with previous and current coordinates
Wrote pdb files with previous and current coordinates

Back Off! I just backed up step120206.pdb to ./#step120206.pdb.2#

Back Off! I just backed up step120206.pdb to ./#step120206.pdb.3#

Back Off! I just backed up step120207.pdb to ./#step120207.pdb.1#
Wrote pdb files with previous and current coordinates
Wrote pdb files with previous and current coordinates

Back Off! I just backed up step120207.pdb to ./#step120207.pdb.2#

Back Off! I just backed up step120207.pdb to ./#step120207.pdb.3#

Back Off! I just backed up step120207.pdb to ./#step120207.pdb.4#

Back Off! I just backed up step120208.pdb to ./#step120208.pdb.1#

Back Off! I just backed up step120208.pdb to ./#step120208.pdb.2#

Back Off! I just backed up step120208.pdb to ./#step120208.pdb.3#
Wrote pdb files with previous and current coordinates
Wrote pdb files with previous and current coordinates
Wrote pdb files with previous and current coordinates
Wrote pdb files with previous and current coordinates
mpiexec: Warning: tasks 7-13 died with signal 11 (Segmentation fault).



*full.mdp*
title                    = 2agv_full
cpp                      = /usr/bin/cpp
;include                  = -I../top
;define                   = -DPROPOSRES -DLIPPOSRES
constraints              = all-bonds
integrator               = md
tinit                    = 0.0
dt                       = 0.002
nsteps                   = 500000   ;1000 ps
nstxout                  = 10000
nstvout                  = 10000
nstlog                   = 10000
nstenergy                = 1000
nstxtcout                = 1000
xtc_grps                 = Protein  SOL POP Na
energygrps               = Protein  SOL POP Na
nstlist                  = 10
ns_type                  = grid
rlist                    = 1.4
coulombtype              = PME
rcoulomb                 = 1.4
rvdw                     = 1.4
tcoupl                   = Berendsen
tc-grps                  = Protein      SOL    POP Na
tau_t                    = 0.1  0.1  0.1 0.1
ref_t                    = 300  300  300 300
Pcoupl                   = Berendsen
pcoupltype               = anisotropic
tau_p                    = 1.00
compressibility          = 4.5e-5 4.5e-5 4.5e-5 0.0 0.0 0.0
ref_p                    = 1.0 1.0 1.0 0.0 0.0 0.0
gen_vel                  = yes
gen_temp                 = 300
gen_seed                 = -1
optimize_fft             = yes