[gmx-users] Magic number error in XTC file

GAYATHRI S gayathri.s at iitb.ac.in
Thu Jul 28 07:36:35 CEST 2016


Dear all,

I had a run an MD simulation (GROMACS 4.6.6) which was abruptly stopped
due to power failure. Then, I extended the simulation using the commands:

$tpbconv -s production.tpr -o production_2.tpr -untill 50000

$mdrun -v -s production_2.tpr -deffnm production_2 -cpi production.cpt

Then, when the simulation was complete, I renamed the output files of the
first simulation as production_1. Next, I wanted to merge the trajectories
of both the runs; hence, I used the following command:

$trjcat -f production_1.xtc production_2.xtc -o production.xtc -overwrite

Fatal Error: Magic number error in XTC file

When I went through the mailing list archive, I found that this error is
shown when the trajectory is corrupted.
So, I used gmxcheck to find out which of the trajectories were corrupted.
I discovered that production_1.trr did not show any error, but
production_1.xtc showed the same error.
Hence, I used trjconv to make a new production_1.xtc

$trjconv -f production_1.trr -o production_1.xtc

Again, I checked the new file for error using gmxcheck. When everything
looked good, I tried the trjcat command again, only to get the following
error

Fatal error: Different numbers of atoms (584885/9944) in the files.

Please suggest how to resolve the issue.

Thank you.

Regards,
Gayathri.



More information about the gromacs.org_gmx-users mailing list