It is possible, if you are not wrapping coordinates back into the primary
unit cell, that you have run long enough for some of your atoms to go "out
of range" for the restart file also. Look for strings of asterisks "*" in
the file "restrt" to make sure this has not happened. There really is not
nearly enough information here to know what is happening for sure. As Tom
says, it could also be an intermittent problem with your mpi-related
hardware (the interconnect); it could also be that your simulation blew up
or a copy of pmemd died on one of the processors for some other reason (even
a bug, though I would like to believe that most of those have been fixed by
now...).
Regards - Bob Duke
----- Original Message -----
From: "Tom Joseph" <ttjoseph.gmail.com>
To: "AMBER Mailing List" <amber.ambermd.org>
Sent: Monday, September 14, 2009 9:34 PM
Subject: Re: [AMBER] Problem on PMEMD
This is an MPI error message, not a pmemd error message. You need to
look at the mdout file (the file you specified with the -o option to
pmemd) for error messages that will help you determine what happened
here.
--Tom
2009/9/14 zyxism <zyxism.gmail.com>:
> Dear Amber,
> I'm a new subscriber of this mail list.
> When I did Nucleic Acid Simulation with PMEMD, using ff99
> forcefield(modified with bsc0 and ion parameters ionsjc_tip3p),I made 300
> scripts , 0.1ns each. But when the simulation was prolonged to about 15ns,
> the simulation was stopped, and the system message was like this :
> rank 0 in job 271 local.localdomain 56517 caused collective abort of all
> ranks
> exit status of rank 0:killed by signal 9
> Is there something wrong with my scripts or with the use of forcefield?
> Zheng
>
>
>
>
> zyxism
> 2009-09-15
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Sep 14 2009 - 19:00:04 PDT
