FATAL ERROR: Bad global improper count!

From: andrea carotti (andcar_at_chimfarm.unipg.it)
Date: Mon Jul 23 2007 - 03:47:16 CDT

Hi all,
I'm getting this error and my simulation (protein) stops.
If then i restart the MD it goes on for a while and after some steps it
stops qith the same error. From a visual inspection everything is ok.
I'm running it using the last version in binary format (tcp).
My computing system is a cluster ibm based x3665, with mpich and sge
queue system (ROCKS cluster). I'm using 18 processors.

This is the LOG with the error:

ENERGY: 495500 391.3126 1007.2228 376.7549
107.2033 -353880.6957 32054.3436 0.0000
0.0000 62914.4988 -257029.3596 311.9644 -256981.5243
-257019.8884 311.7083 -116.1457 -56.7914
1004811.0739 0.0489 0.6264

WRITING EXTENDED SYSTEM TO RESTART FILE AT STEP 495500
FATAL ERROR: Bad global improper count!

FATAL ERROR: See http://www.ks.uiuc.edu/Research/namd/bugreport.html
Stack Traceback:
  [0] _ZN10Controller9integrateEv+0x5a0 [0x821bddc]
  [1] _ZN10Controller9algorithmEv+0x518 [0x82146d0]
  [2] _ZN10Controller9threadRunEPS_+0xc [0x8221770]
  [3] /home/install/namd/namd2 [0x831aa19]
  [4] Charm++ Runtime: Converse thread (qt_args+0x72 [0x83939ca])

Here is my configuration file:
;# molecular system
structure 1YCR_charmm.psf

# force field
paratypecharmm on
parameters PARM.PRM
exclude scaled1-4
1-4scaling 1.0

# approximations
switching on
switchdist 8
cutoff 12
pairlistdist 13.5
margin 0
stepspercycle 20
splitPatch hydrogen
dielectric 1
rigidBonds all
COMmotion no
useSettle on
fullElectFrequency 2
nonbondedFreq 1

# Periodic Boundary conditions
cellBasisVector1 102.493 0. 0.
cellBasisVector2 0. 102.493 0.
cellBasisVector3 0. 0. 102.493
cellOrigin 51.246 51.246 51.246
wrapWater on # wrap water to central cell
wrapAll on # wrap other molecules too
wrapNearest off # use for non-rectangular cells

#PME (for full-system periodic electrostatics)
PME yes
PMEGridSizeX 122 # Valore > vicino a quello della cella
PMEGridSizeY 122 #
PMEGridSizeZ 122 #

#integrator
timestep 2.0 #unita in fs

#output
outputenergies 500 #timestep per output ene
outputtiming 1000 #timestep per output per CPU memoria
e altro
outputpressure 1000 #timestep per output pressure (bar)

# continuing a run
binaryoutput yes
binCoordinates 1YCR_min_equ.coor # coordinates from last run (binary)
binVelocities 1YCR_min_equ.vel # velocities from last run (binary)
extendedSystem 1YCR_min_equ.xsc # cell dimensions from last run
numsteps 500000 # run stops when this step is
reached:attiva per dyn. 1ns

# molecular system
coordinates 1YCR_charmm.pdb

#Output
outputname 1YCR_min_dyn
DCDfile 1YCR_min_dyn.dcd
DCDfreq 2000
restartname 1YCR_min_dyn
restartfreq 500
binaryrestart yes

seed 12345

# Constant Temperature Control Langevin dyn
langevin on # langevin dynamics
langevinDamping 10 # damping coefficient of 10/ps
langevinTemp 310 # temp (K)
langevinHydrogen on # Couple bath to hydrogens (on,off)

# Constant Pressure Control
useGroupPressure yes # needed for rigid bonds
useFlexibleCell no # no for water box, yes for membrane
useConstantArea no # no for water box, maybe for membrane

#Uso Langevin (volume variabile)
langevinPiston on #
langevinPistonTarget 1.01325 ;# pressure in bar -> 1 atm
langevinPistonPeriod 200. ;# oscillation period around 100 fs
langevinPistonDecay 100. ;# oscillation decay time of 50 fs
langevinPistonTemp 310 ;# coupled to heat bath

Thanks to all
Andrea

-- 
¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
Andrea Carotti
Dipartimento di Chimica e Tecnologia del Farmaco
Via del Liceo, 1
06123 Perugia, Italy
phone: +39 075 585 5169
fax: +39 075 585 5161
www http://rpg.unipg.it
personal www http://iris.chimfarm.unipg.it/users/andcar

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:44:59 CST