Re: [AMBER] Stopping sander in production phase. Why?

From: Jason Swails <jason.swails.gmail.com>
Date: Fri, 23 Oct 2015 11:24:06 -0400

On Fri, Oct 23, 2015 at 11:12 AM, Atila Petrosian <atila.petrosian.gmail.com
> wrote:

> Dear Dr Roitberg,
>
> Thanks for your answers.
>
> My system is one of nodes (compute-0-2) in rocks system. But I just have
> access to it and use for MD simulation, only.This node has 8 processors.
>
> Unfortunately, I have remote connection to system and I have internet
> problem and I can not access to files to Monday. Now is weekend in my
> country.
>
> I will check and visualize it.
>
> In this case, in the info file,
>
> NSTEP = 82500 TIME(PS) = 6665.000
>
> I have another question. I have many protein-ligand complex to do MD
> simulation.
>
> In an another complex on another node of rocks (that I just have access to
> it ), in the first production phase, I encountered with an interesting
> point. sander was stopped. End of the out file is as follows:
>
> NSTEP = 862500 TIME(PS) = 3225.000 TEMP(K) = 300.70 PRESS =
> 66.4
> Etot = -91819.1456 EKtot = 23880.5373 EPtot =
> -115699.6829
> BOND = 991.8114 ANGLE = 2600.1910 DIHED =
> 3410.4648
> 1-4 NB = 1178.1472 1-4 EEL = 14041.4112 VDWAALS =
> 12746.1452
> EELEC = -150667.8536 EHBOND = 0.0000 RESTRAINT =
> 0.0000
> EKCMT = 10042.1420 VIRIAL = 9489.3132 VOLUME =
> 385716.3526
> Density =
> 1.0271
> Ewald error estimate: 0.3641E-06
>
> ------------------------------------------------------------------------------
>
> NSTEP = 865000 TIME(PS) = 3230.000 TEMP(K) = 301.75 PRESS =
> 57.9
> Etot = -92106.4028 EKtot = 23964.1744 EPtot =
> -116070.5772
> BOND = 1005.2157 ANGLE = 2636.4786 DIHED =
> 3385.8516
> 1-4 NB = 1142.1352 1-4 EEL = 13988.7767 VDWAALS =
> 13041.7090
> EELEC = -151270.7440 EHBOND = 0.0000 RESTRAINT =
> 0.0000
> EKCMT = 10141.9697 VIRIAL = 9659.8830 VOLUME =
> 385573.9303
> Density =
> 1.0275
> Ewald error estimate: 0.1193E-03
>
> ------------------------------------------------------------------------------
>
> NSTEP = 867500 TIME(PS) = 3235.000 TEMP(K) = 301.48 PRESS
>
>
> ***************************************************************************************************
> While end of the info file is as follows:
>
> NSTEP = 2500000 TIME(PS) = 6500.000 TEMP(K) = 299.05 PRESS =
> 197.7
>

​Did you ask for more steps than this?



> Etot = -92057.6812 EKtot = 23749.9389 EPtot =
> -115807.6201
> BOND = 970.4243 ANGLE = 2638.3735 DIHED =
> 3421.2260
> 1-4 NB = 1168.8488 1-4 EEL = 13997.9525 VDWAALS =
> 12941.0212
> EELEC = -150945.4663 EHBOND = 0.0000 RESTRAINT =
> 0.0000
> EKCMT = 10058.1946 VIRIAL = 8412.3404 VOLUME =
> 385481.4470
> Density =
> 1.0277
> Ewald error estimate: 0.3833E-04
>
> What is the reason of this issue? Have this md simulation ended?
>

​There is no error message and nothing looks suspicious, so we have
absolutely no way of knowing. It could be that the job was simply killed.​
 It could be a problem during cleanup of sander (some kind of double-free
or other bug that killed sander during the cleanup process). The error
message, then, is almost certainly in the standard error output, not the
standard output printout (which is captured in the mdout).

It could also be in the standard output of one of the non-master threads if
it was an error in a slave node that sander detected and handled (in which
case it would be captured in the stdout stream NOT redirected to the mdout).

Long story short, without those outputs, there are many possible causes
(all very different) that are equally likely. So we can't even really
speculate.

HTH,
Jason

-- 
Jason M. Swails
BioMaPS,
Rutgers University
Postdoctoral Researcher
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Oct 23 2015 - 08:30:06 PDT
Custom Search