AMBER Archive (2006)

Subject: RE: AMBER: Amber9 serial installation -tests- severe (174): SIGSEGV, segmentation fault occurred

From: Ross Walker (ross_at_rosswalker.co.uk)
Date: Fri Oct 20 2006 - 11:06:49 CDT


Dear Mina,

> I tried the divcon_intel.bugfix and I get no segmentation
> faults any more ;)

Excellent, I will post it as a bugfix then.

> I was a bit worried to see though, that the results obtained
> by running test.antechamber have quite a few discrepancies
> from the expected ones, and as I can not assess the
> seriousness of the differences, I would appreciate it if you
> could have a quick look and tell me if these differences are
> serious or not. I attach the TEST_FAILURES.diff created by
> running the test.antechamber.

This is weird though. Normally I just see roundoff differences in the last
column, the charges, which is to be expected. Here though you see
fundamentally different atom types:

< 3 HH32 1.486 2.454 0.890 hc 1 ACE 0.066
> 3 HH32 1.486 2.454 0.890 ha 1 ACE 0.066

< 18 C2 4.713 6.129 0. c 1 ASH 0.639
> 18 C2 4.713 6.129 0. c2 1 ASH 0.637

There is something going very wrong here. Junmei should probably look into
it as he wrote the antechamber code. Can you just try one thing though.
Create a completely new directory and untar your original amber9.tgz file
into that directory so that you have a completely clean copy of amber9. Then
get apply the bugfix.all file that is attached to this email (this is the
one from the amber website with the divcon bugfix also added to it). Then do
a full serial compile and run the full test suite again. If the problem is
still showing up then it needs looking into.

All the best
Ross

/\
\/
|\oss Walker

| HPC Consultant and Staff Scientist |
| San Diego Supercomputer Center |
| Tel: +1 858 822 0854 | EMail:- ross_at_rosswalker.co.uk |
| http://www.rosswalker.co.uk | PGP Key available on request |

Note: Electronic Mail is not secure, has no guarantee of delivery, may not
be read every day, and should not be used for urgent or sensitive issues.



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