[gmx-users] Fatal error when using Hbond restraints

Robert Johnson bobjohnson1981 at gmail.com
Sat Mar 17 19:29:15 CET 2007

Hello everyone,
I'm performing a simulation that includes a 2D infinite SiO2
substrate. I'm not interested in the high frequency motion of the
substrate bonds, so I'm applying position restraints to all the SiO2
atoms. I have other molecules in the system such as water and DNA that
require hydrogen bond restraints. However, I'm getting the following
error when trying to run the system across several nodes:
Fatal error:
Shake block crossing node boundaries
constraint between atoms (1680,1716)

The atoms listed are in the substrate. Since I'm already restraining
the substrate atoms, I don't really need to constrain the Hbonds in
the SiO2. Is there any way to apply Hbond restraints only to a group
of atoms?

Alternatively, how does Gromacs decide which bonds contain hydrogen? I
could just alter the topology file for the substrate to trick Gromacs
into thinking that the hydrogens in the SiO2 are really a different
type of atom.
Bob Johnson

More information about the gromacs.org_gmx-users mailing list