Thanks for your quick reply, David. You wrote: > Have you looked at the pdb files? Anything strange with them? The residue+its surroundings look normal before I get the errors. In the pdb files that were produced when ths run crashed, two hydrogens in the lysine are very close to each other. I can't see any other irregularities. You wrote: > You are not maybe running with redhat linux 6.2 > the standard kernel, are you? I don't think so. How would this affect the stability of my the MD run ? Thanks very much for your help ! Best wishes, Chris