FATAL ERROR: PBC

From: Raha YS (rahays62_at_gmail.com)
Date: Thu Oct 03 2019 - 14:20:40 CDT

Dear all,
I want to run a simulation (unit cell protein inside water) on NAMD with
non-orthogonal PBC (hexagonal box). As I start minimization, using shake
algorithm (rigidBonds on), after several (ten or so) runs it crashes with
the following error:

Segmentation fault (core dumped)

and inside my log file it includes the following error for some atoms:

ERROR: Constraint failure in RATTLE algorithm for atom 3450!

when I look at these atoms, they are all located at the boundary with (x y
z)s related as: (-40.460 -29.028 -6.030) & (40.460 29.028 -6.030) for
each pair, which means that these atoms overlap (taking into account PBC).
As for minimization, I turn the rigidBonds off, it can finish minimization,
however it contains warning as fellows:

Warning: Low global exclusion count! (97031 vs 97869)
Warning: This warning is not unusual during minimization.
Warning: Increasing pairlistdist or cutoff may avoid this.

and crashes as it starts running MD with the following error:

OPENING COORDINATE DCD FILE
WRITING COORDINATES TO DCD FILE ./output/ionizedparal_output.dcd AT STEP
1000
REINITIALIZING VELOCITIES AT STEP 1000 TO 277.15 KELVIN.
TCL: Running for 200000 steps
FATAL ERROR: Low global exclusion count! (97031 vs 97869) System unstable
or pairlistdist or cutoff too small.
FATAL ERROR: See http://www.ks.uiuc.edu/Research/namd/bugreport.html

playing with the cutOff, pairListDist, and timeStep values does not resolve
the problem. I also tried making the box a bit bigger and it does not help
either.

Can anybody help me with this? thanks!

This archive was generated by hypermail 2.1.6 : Thu Dec 31 2020 - 23:17:12 CST