[gmx-users] Magic Number Error in XTC file - Cannot determine precision of trn file

[Walter Ash]

Hi everybody,

I've got about 80 trajectories (4o each on two different systems) with 1 
month's worth of data in each trajectory, and every single one appears 
to be corrupted in some mysterious way.   Is this a known bug in 
gromacs?  Can I recover the files? 

If I run "gmxcheck" on the XTC file I get:
 > Checking file umb.1.xtc
 > Reading frame       0 time 6500.000
 > # Atoms  29428
 > Precision 0.001 (nm)
 > Reading frame      30 time 6800.000   
-------------------------------------------------------
 > Program gmxcheck, VERSION 3.3.1
 > Source code file: xtcio.c, line: 83
 >
 > Fatal error:
 > Magic Number Error in XTC file (read 0, should be 1995)

The error message with the .trr file is "Can not determine precision of 
trn file".  I cannot view these files in VMD ("Error reading timestep, 
file does not match format"), and I can't continue them with trjconv 
("Can not determine precision of trn file").

This is with gromacs 3.3.1, running umbrella sampling simulations.