[gmx-users] reg plumed gromacs

[Mark Abraham]

On 15/10/2011 1:26 AM, vidhya sankar wrote:
> Dear justin,  thank you for your previous reply
>
>    I am new user of plumed gromacs  i am using the following file to 
> do meta dynamics by taking ELSTPOT as CV . i have used the following 
> files as plumed.dat file  withe RESTART key word . But when i restart  
> by actual procedure of Gromacs MD i have got  the follwing error
> Reading checkpoint file state.cpt generated: Thu Oct 13 23:07:42 2011
>   #PME-nodes mismatch,
>     current program: -1
>     checkpoint file: 0
> Gromacs binary or parallel settings not identical to previous run.
> Continuation is exact, but is not guaranteed to be binary identical.
>
> IS it error or not ?. What should i do to avoid this information ?
> But still Gromacs MD is running continously  i am getting output with 
> updated output(COLVAR, HILL)  . i am expectiong your valuable reply
>
>

See http://www.gromacs.org/Documentation/Terminology/Reproducibility

Your current run is using mdrun -npme -1 (the default). Probably the 
former run was also, but zero was chosen by mdrun. The latter was 
recorded in the checkpoint file, and now mdrun is telling you that there 
may be some degree of discontinuity between the runs if you were to have 
changed something else that would lead to zero not being chosen this 
time (e.g. running on more nodes). If you specify the same suitable 
value each time for -npme, this warning will never occur.

Mark
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20111015/b964c4fc/attachment.html>