NAMD2.6 repeatedly freezing on new desktop

From: Jonathan Bourne (jwb268_at_gmail.com)
Date: Sat Nov 01 2008 - 16:03:53 CDT

Hello All,

Our lab recently added a custom built desktop for dedicated NAMD
simulations. The PC has an Intel Core 2 Quad Processor Q6600 @ 2.40
GHz, Kingston 2 GB RAM, Asus P5N-T Deluxe motherboard, aftermarket
cooling, Windows XP Pro (SP3), and is running NAMD 2.6 i686 version
from http://www.ks.uiuc.edu/Research/namd/

Using this new desktop described above, I tried running restarts of a
simulation that can be successfully restarted and ran on an AMD 5600+
dual core system (WinXP Pro SP2), a Pentium4 w/HT (WinXP Pro SP2). The
simulation proceeded fine for a variable length of time (minutes
hours) and then NAMD seems to spontaneously hang (system clock stops,
mouse input fails to register) with no error message, or system error
log generated by NAMD or Windows on the new system, and requires the
computer to be hard restarted. This occurs when the identically when
NAMD program is run alone, or with Charmrun.
  # namd2 +p4 file.conf > file.log
  # charmrun namd2 ++local +p4 file.conf > file.log

One unique aspect of this is that when the system hangs, my monitor
continues to show whatever was showing when everything crashed. Up to
the point of no return, my CPU's were all running full load (98%),
memory was around 500mb usage, and CPU and motherboard temps were
38-42 deg C (according to the ASUS motherboard temperature probes).
In order to make sure this wasn't a simple hardware malfunction, I
tried using Prime95 on the blended stress test setting, and the system
ran error free for over 24 hours.

To check whether this "system hang" issue was unique to the Windows
port, I used a Kubuntu 8.04.1 LiveCD, to re-run the program under
Linux (NAMD_Linux-amd64 port also from ks.uiuc.edu), and got an
identical spontaneous hang results.

Because I got the same spontaneous "hang" in Kubuntu-linux, I moved
back to my native Win32 install, and tried some other variations. In
order to check that the problem was not due to my restart files, I
tried using an earlier set of restart files and then tried running my
simulation from t=0. Both cases failed with the same spontaneous
freeze. I'm pretty confident that my psf and pdb files are fine, and
the parameter file is good. Again, these all worked on two other
WinXP desktops, and my equilibration run on the AMD dual core system
was up to 3 ns. However, just in case, I tried running the tutorial
simulation (1-2 ubq_ws_eq) with all of the provided files. Again,
these simulations also spontaneously hang, identical to my other
attempts.

I'm now out of ideas, and still not certain where my problem is.
Because Prime95 works, I presume the problem is not hardware based.
Does anyone have NAMD2.6 working on a Core2 Quad Q6600 system, or know
a reason why these processors would be unable to run NAMD? Are there
any other diagnostic procedures, programs, or suggestions to further
troubleshoot this, or does anyone have any suggestions on how to solve
this problem?

Thank you for your time and consideration.

Sincerely,
Jonathan

jwb268(at)gmail.com

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:50:02 CST