hangup (signal1) CO2 simulation

Hello everyone,

i'm trying to simulate CO2-molecules, but I receive the following
error message:
APPLICATION TERMINATED WITH THE EXIT STRING: Hangup (signal 1)

At first I thought that the pppm-style charge distribution represented
the problem, but no changes there helped. Even if I omit charges
completely, the problem persists.
I checked the mailing list and found only one entry about this
problem, considering a too large cutoff radius, which shouldn't be the
problem in my case.

I'm currently at a loss to determine the origin of this failure and I
would appreciate any help or hint, where to look!

Thank you for your time and help in advance,
Jan

The input for read_data is attached and the setup is:

in.co2:

dimension 3
boundary p p p

# 512 CO2 in a 3d-box

units real
atom_style hybrid angle charge
pair_style lj/cut/coul/long 21.0 21.0

bond_style harmonic
angle_style harmonic
kspace_style pppm 1.0e-4

atom_modify map array sort 1000 0
read_data data.co2zhang
group co2 molecule <= 512
group Cs type 1

pair_coeff 1 1 0.0573209 2.7918 21.0
pair_coeff 2 2 0.164254 3.0 21.0

pair_modify table 12 mix arithmetic

neighbor 3.0 bin
neigh_modify delay 0 every 1 check yes
neigh_modify exclude molecule co2
velocity co2 create 290.0 12345 dist gaussian mom yes

fix 1 co2 shake 0.00001 20 1000 b 1 t 1 2
fix 1 all nvt temp 290.0 290.0 100.0
timestep 1

compute clusterdef Cs cluster/atom 6.3728
thermo_style multi
thermo 1000

dump 1000 all custom 1000 dump.co2 id type x y z c_clusterdef

run 10000

data.co2zhang (68.6 KB)

I can;t run this until I'm back in the office next week,
but there is no other error message? If you
are trying to use SHAKE with a constraint
angle of 180 for CO2, that could be causing
a problem - see the mail list archives for
discussion of that.

Steve