[gmx-users] Replica Exchange MD using Gromacs

[Mark Abraham]

On 1/04/2011 6:52 PM, bsmith wrote:
> On 01/04/2011, at 6:29 PM, Mark Abraham wrote:
>
>> On 1/04/2011 6:03 PM, Ruchi Gupta wrote:
>>> Dear gmx-users,
>>>
>>> I am facing some problems while running replica exchange MD using 
>>> Gromacs.
>>>
>>> Few seconds after the job submission it ends with the following 
>>> error message:
>>>
>>> "
>>> Initializing Replica Exchange
>>>
>>> Repl  There are 6 replicas:
>>> Multi-checking the number of atoms ... OK
>>> Multi-checking the integrator ... OK
>>> Multi-checking init_step+nsteps ...
>>> init_step+nsteps is not equal for all subsystems
>>>   subsystem 0: 0
>>>
>>>   subsystem 1: 50000
>>>   subsystem 2: 0
>>>   subsystem 3: 50000
>>>   subsystem 4: 0
>>>   subsystem 5: 50000
>>>
>>> -------------------------------------------------------
>>> Program mdrun, VERSION 4.5.4
>>> Source code file: main.c, line: 249
>>>
>>>
>>> Fatal error:
>>> The 6 subsystems are not compatible
>>>
>>> For more information and tips for troubleshooting, please check the 
>>> GROMACS
>>> website athttp://www.gromacs.org/Documentation/Errors
>>>
>>> -------------------------------------------------------
>>>
>>> "
>>>
>>> I am using a "single .gro file" and "6 different md.mdp" files (at 6 
>>> different temperatures) for generating "6 .tpr files", respectively. 
>>> But the simulation doesn't work.
>>>
>>>
>>> I have compared the ".tpr files" using "gmxdump" and "gmxcheck" 
>>> Gromacs commands and they seem fine with respect to nsteps and 
>>> init_step information.
>>
>> So, to be clear, gmxcheck says that for no pair of .tprs does 
>> init_step or nsteps differ...
>>
>> What is your mdrun command line? Are you using checkpoint files?
>>
>> Mark
>> -- 
>> gmx-users mailing list gmx-users at gromacs.org 
>> <mailto:gmx-users at gromacs.org>
>> http://lists.gromacs.org/mailman/listinfo/gmx-users
>> Please search the archive at 
>> http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
>> Please don't post (un)subscribe requests to the list. Use the www 
>> interface or send it to gmx-users-request at gromacs.org.
>> Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
> 1) Comparison of the .tpr files using gmxdump reveals no differences 
> but for the temperatures and the velocities in each i.e. init_step and 
> nsteps have the same values (0 and 50000, respectively) in all files.
>
> 2) We have tried running mdrun without checkpointing (-cpt -1) and 
> with checkpointing with no change in the outcome.

That sounds like a bug. Please create a new Redmine issue here 
http://redmine.gromacs.org and attach your .tpr and .mdp files - 
preferably in a compressed archive file. You should get the option to 
assign it to me - please do.

Mark
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20110401/471a7c08/attachment.html>