AMBER Archive (2006)

Subject: AMBER: Amber: Can't get sander to run

From: sethl_at_gatech.edu
Date: Tue Jun 06 2006 - 13:15:14 CDT

Quoting Ross Walker <ross_at_rosswalker.co.uk>:

>
> > When running sander in as outlined in the DNA tutorial, I experience a
> > segmentation fault. I could not find a solution in the
>
> Unfortunately just saying the code crashed does not help at all. Can you
> provide many more details. What version of Amber, what compiler, what
> operating system, what hardware. What configure options did you use? Is this
> in serial or parallel? Do the amber test cases run properly? Does amber work
> for other calculations or does it always segfault? Did it write anything to
> the output file before segfaulting? Was anything else printed to the screen?
>
> Without this sort of information we cannot possibly work out what might be
> going wrong.
>
> All the best
> Ross
>

Ok, so I've narrowed it down.

Operating System: RedHat Enterprise 4
Hardware: AMD 64
Compilier: Intel iFort 9.0.031
Software Version: Amber8

I did a "./configure -athlon -static" but replaced the config.h with one with
the "CC=gcc -m32" to get gcc to force gcc to work in 32 bit mode. Then I did a
"make serial" to install amber8.

Sander fails when running a test case on it... here's the output on the screen:

cd dmp; ./Run.dmp
Segmentation fault
  ./Run.dmp: Program error
make: *** [test.sander.no_lmod] Error 1

Thank you always,
Seth Lilavivat
Georgia Institute of Technology

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu