[gmx-users] Range checking error: ci barely out of bounds.

[chris.neale at utoronto.ca]

I am attempting to solve my range checking errors where ci is just  
barely out of bounds. This particular run is on 4cpus. I have reposted  
because now I have noticed that a0.log differs from the other log  
files with respect to the reported cells in the grid and that this may  
be important. Any help is greatly appreciated.

My error message is:
Variable ci has value 1625. It should have been within [ 0 .. 1584 ]
And I notice that 12 x 12 x 11 = 1584, but 13 x 12 x 11 = 1716 (which  
would put ci=1625 in bounds).

grep 'Grid:' a0.log | wc -l = 24 lines
grep 'Grid:' (a1.log or a2.log or a3.log) | wc -l = 25 lines

The straight grep 'Grid:' output is:
...
Grid: 12 x 12 x 11 cells
Grid: 13 x 12 x 11 cells  (a0.log does not contain this line)

The same pattern tends to occur for different jobs. For another  
example, Grid: 13 x 13 x 10 cells --> Grid: 13 x 13 x 11 cells (not on  
a0.log) and ci = 1721. It should have been within [ 0 .. 1690 ] (But  
13 x 13 x 11 = 1859 putting 1721 in bounds).

Can the grid sizes be different for the different processors or is  
this an output artifact?

Here is my MDP file:
title               =  seriousMD
cpp                 =  /home/cneale/exe/cpp
integrator          =  md
nsteps              =  350000
tinit               =  0
dt                  =  0.002
comm_mode           =  linear
nstcomm             =  1
comm_grps           =  System
nstxout             =  50000
nstvout             =  50000
nstfout             =  50000
nstlog              =  1000
nstlist             =  10
nstenergy           =  5000
nstxtcout           =  5000
ns_type             =  grid
pbc                 =  xyz
coulombtype         =  PME
rcoulomb            =  0.9
fourierspacing      =  0.12
pme_order           =  4
vdwtype             =  cut-off
rvdw_switch         =  0
rvdw                =  1.4
rlist               =  0.9
DispCorr            =  no
Pcoupl              =  Berendsen
pcoupltype          =  semiisotropic
compressibility     =  4.5e-5 4.5e-5
ref_p               =  1.     1.
tau_p               =  4.0    4.0
tcoupl              =  Berendsen
tc_grps             =  Protein     LDA_POPE_DMPE     SOL_NA+
tau_t               =  0.1         0.1               0.1
ref_t               =  300.        300.              300.
annealing           =  no
gen_vel             =  yes
unconstrained-start =  no
gen_temp            =  300.
gen_seed            =  9896
constraints         =  all-bonds
constraint_algorithm=  lincs
lincs-iter          =  1
lincs-order         =  4

There is more background information and run data here:
http://www.gromacs.org/pipermail/gmx-users/2006-October/024154.html
http://www.gromacs.org/pipermail/gmx-users/2006-September/024003.html