[gmx-users] Magic Number Error

Mark Abraham Mark.Abraham at anu.edu.au
Mon Jan 28 11:51:55 CET 2008

alessandro casoni wrote:
> I have a little problem...when I use g_covar I get this message:
> Program g_covar, VERSION 3.3.2
> Source code file: xtcio.c, line: 83
> Fatal error:
> Magic Number Error in XTC file (read 0, should be 1995)
> ..could you help me?

Use ls -l to check that your .xtc file has non-zero size. Otherwise, 
you've somehow managed to mangle the file so that the "magic" numbers at 
the start of the file (which confirm to GROMACS that this binary file is 
indeed suitably-constructed .xtc trajectory data) are not correct. This 
should not be caused merely by moving from big- to little-endian 
architecture (or vice-versa), but if you generated it on a different 
system to the one on which you're running g_covar, this would be a point 
of interest.


More information about the gromacs.org_gmx-users mailing list