[gmx-users] Continue run after a crush

[itai bloch]

Hi all,

I had a 30ns simulation on 4 nodes but it crushed after ~24ns.
I am now trying to continue the remaining time on different machine
(both are Linux).
I have followed the orders I found in the FAQ (in the gromacs site),
BUT I  am having doubts regarding the constraints!
The only constraints I used were the "all-bonds" constraints in the
original run and now after the crush the procedure I followed was:

1. tpbconv_mpi -s before.tpr -f before.trr -n deshuf.ndx -o crush1 -time 24411
2. mdrun - s crush1 .......

Is it OK to do it like that or should I do it differently and regards
the constraints in a different way?

I will be very happy to hear your ideas/suggestions.

Thanks, Itai.