[gmx-users] gmxcheck : Magic Number Error in XTC file (read 0, should be 1995)

[Mark Abraham]

Hi,

The most likely explanation is that you were using a network file system
that dropped out for some time, was later restored, and it lost some data
in the meantime. This might happen if it was caching data before writing it
to disk, telling mdrun everything was OK, and something traumatic happened
in the meantime. mdrun can't reasonably find out anything until its next
attempt to write a file to disk, and whether a file system can, does or
should report failure at that time is up for debate. I would make inquiries
of your system administrators, and perhaps encourage them to make the file
system more robust. You might like to take steps to write your files to
real disk that is local to the compute node of MPI rank 0, and transfer the
file back at the end of the run. Since you've had problems, you should
definitely add a gmxcheck stage to your workflow between restarts.

Mark


On Thu, Dec 25, 2014 at 10:00 AM, Amit Kumawat <amit.kumawat20 at gmail.com>
wrote:

> Thank you very much David. Out of the complete trajectory I identified all
> those frames which are corrupted and manually concatenated the rest of the
> trajectory. Can you provide me suggestion for the possible source of error.
> Because in the log file I could not see any out of disk space error and I
> have been using the same version of gromacs  for initiating or extending
> the simulation
>
> Regards,
> Amit R. Kumawat
> Junior Research Fellow,
> C/O Dr. Suman Chakrabarty
> Physical & Materials Chemistry Division
> CSIR-National Chemical Laboratory
> Pune, India - 411008
> Mobile 09579195769
>
>
> On Wed, Dec 24, 2014 at 8:56 PM, David van der Spoel <spoel at xray.bmc.uu.se
> >
> wrote:
>
> > On 2014-12-24 15:33, Amit Kumawat wrote:
> >
> >> Dear Gromacs Users,
> >>
> >> I have a trajectory of 2microsecond.
> >> I am trying to use gmxcheck to analyse trajectory but I am getting
> >> following error;
> >>
> >> Program gmxcheck, VERSION 4.6.5
> >> Source code file: /home/suman/sources/gromacs-4.6.5/src/gmxlib/xtcio.c,
> >> line: 92
> >>
> >> Fatal error:
> >> Magic Number Error in XTC file (read 0, should be 1995)
> >>
> >> I have seen similar questions in the mailing list describing the
> >> trajectory
> >> may be corrupted. If its true in all the cases of this error, Is there
> any
> >> way to recover the trajectory?
> >>
> >> Please provide suggestion to solve the error.
> >>
> > start with
> > trjconv -f old.xtc -o new.xtc
> > to see how much is corrupt.
> >
> >
> >
> >> Regards,
> >> ​Amit
> >>
> >> =============================​
> >>
> >> Amit R. Kumawat
> >> Junior Research Fellow,
> >> Physical & Materials Chemistry Division
> >> CSIR-National Chemical Laboratory
> >> ​Pune, INDIA​
> >>
> >> ​=============================​
> >>
> >>
> >
> > --
> > David van der Spoel, Ph.D., Professor of Biology
> > Dept. of Cell & Molec. Biol., Uppsala University.
> > Box 596, 75124 Uppsala, Sweden. Phone:  +46184714205.
> > spoel at xray.bmc.uu.se    http://folding.bmc.uu.se
> > --
> > Gromacs Users mailing list
> >
> > * Please search the archive at http://www.gromacs.org/
> > Support/Mailing_Lists/GMX-Users_List before posting!
> >
> > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> >
> > * For (un)subscribe requests visit
> > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> > send a mail to gmx-users-request at gromacs.org.
> >
> --
> Gromacs Users mailing list
>
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
> posting!
>
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
> * For (un)subscribe requests visit
> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> send a mail to gmx-users-request at gromacs.org.
>