Anna -
So what does the restrt actually look like? They are not that hard to read
in a text editor, you know. Any fields filled with asterisks? That sort of
thing is typically what messes up a mdcrd, and wrapping typically solves
that problem. Any chance the disk went full on you, due to the mdcrd and
mdvel (or other people writing to the same disk), and the restrt write
attempt failed? Any chance the disk has hardware problems? If you don't
have an integral amount of data, then possibly one of these things happened,
or the master process died without flushing the restrt completely to disk.
All sorts of things can happen on linux clusters that have nothing to do
with the amber software; feel free to mail me gzipped copies of the mdout
and problem restrt if you wish (just looking in the mdout to get the atom
count/box data). I take it that doing everything you are doing here (ie.,
an nmropt run of this sort) routinely works for you?
Regards - Bob Duke
----- Original Message -----
From: <anna.schrey.gmx.de>
To: <amber.scripps.edu>
Sent: Saturday, June 17, 2006 8:30 AM
Subject: AMBER: Problems with restart file from pmemd sim.
> Dear specialists,
>
> I ran a simulation with a setup as follows:
>
>
> cat << eof > dyn.in
>
> # 400ps dynamics to drop the restraints
> &cntrl
> imin=0, nmropt=1,
> ntx=5, irest=1,
> ntrx=1, ntpr=1000,
> ntwr=1000,
> ntwx=1000,ntwv=1000,
> ntwe=1000,
> ntf=2,ntb=1,
> dielc=1,cut=8,
> scnb=2,scee=1.2,
> ntr=0,nstlim=200000,
> dt=0.002,temp0=280,
> ntt=1,
> tautp=1.0,ntc=2,
> &end
> &wt type='END' &end
> LISTIN=SH3_13COM_HB_NTV_eq2.restraints
> LISTOUT=SH3_13COM_HB_NTV_eq2.violations
> DISANG=SH3_cryst_COM_HB_end.rst
> eof
>
> /usr/local/encap/mpich-126-intel-32/bin/mpirun -machinefile
> mpich_hosts -np 8 \
> /swc/amber8/exe/pmemd -O -i dyn.in -o SH3_13COM_HB_NTV_eq2.out \
> -p SH3_13COM.top -c SH3_13COM_HB_NTV_eq.crd -ref SH3_13COM_HB_NTV_eq.crd \
> -r SH3_13COM_HB_NTV_eq2.crd -x SH3_13COM_HB_NTV_eq2.mdcrd -v
> SH3_13COM_HB_NTV_eq2.mdvel \
> -inf SH3_13COM_HB_NTV_eq2.inf -e SH3_13COM_HB_NTV_eq2.mden
>
> Trying to use the restart file for a follow-up simulation, it failed.
> Trying to create a PDB file from the restart file gave the following error
> message:
> | New format PARM file being parsed.
> | Version = 1.000 Date = 03/24/06 Time = 13:17:12
> fmt: read unexpected character
> apparent state: unit 5 (unnamed)
> last format: (6F12.7)
> lately reading sequential formatted external IO
> Abort
>
> The .mdcrd file can be read properly by software such as vmd, whereas a
> CARNAL dump of the last frame resulted in a distorted structure.
>
> Sometimes, this problem could be avoided by wrappig the trajectory, but
> not this time.
>
>
> Thanx for answers,
>
> Anna
>
> --
>
>
> Der GMX SmartSurfer hilft bis zu 70% Ihrer Onlinekosten zu sparen!
> Ideal für Modem und ISDN: http://www.gmx.net/de/go/smartsurfer
>
> --
>
>
> Der GMX SmartSurfer hilft bis zu 70% Ihrer Onlinekosten zu sparen!
> Ideal für Modem und ISDN: http://www.gmx.net/de/go/smartsurfer
> -----------------------------------------------------------------------
> The AMBER Mail Reflector
> To post, send mail to amber.scripps.edu
> To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
>
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber.scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
Received on Sun Jun 18 2006 - 06:07:25 PDT
