[gmx-users] Some confusion related to extending a simulation

[Justin Lemkul]

On 6/8/14, 8:05 PM, bharat gupta wrote:
> Hi,
>
> I extended a crashed simulation using the command :
>
> mpirun -np 24 mdrun -v -s mdrun.tpr -cpi mdrun.cpt
>
> I just stopped the restarted simulation after some time and found that
> mdrun.log was not updated, because log file and results obtained from
> gmxcheck are different.
>
> Here's the last information about the simulation that I stopped:
>
> vol 0.64  imb F  1% pme/F 0.88 step 1431100, will finish Sat Jun 21
> 03:51:12 2014
>
> (It means 1431100*0.002 = 2.8 ns has finished)
>
> Here's is the mdrun.log file details :
>
> DD  step 219999  vol min/aver 0.575  load imb.: force 37.9%  pme mesh/force
> 0.960
>
>             Step           Time         Lambda
>           220000      440.00000        0.00000
>
> When I checked the trajectory file using gmxcheck it shows only 400ps,
> here's the details:
> Reading frame     200 time  400.000
>
>
> Item        #frames Timestep (ps)
> Step           221    2
> Time           221    2
> Lambda         221    2
> Coords         221    2
> Velocities     221    2
> Forces           0
> Box            221    2
>
> I am confused now what is going on ?? How can I get those 2.8 completed
> trajectory ??
>

Looks like your run re-started.  Likely this is due to a problem with a .cpt 
file, which should have been noted in the .log file.  mdrun won't over-write 
your previous trajectory, so it should have been backed up (e.g. #traj.xtc.1#).

-Justin

-- 
==================================================

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 601
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

==================================================