[1455] in Release_7.7_team
more on BioSym problems
daemon@ATHENA.MIT.EDU (Alex T Prengel)
Thu Aug 13 15:37:24 1998
To: wdc@MIT.EDU, mbarker@MIT.EDU, ghudson@MIT.EDU, rbasch@MIT.EDU,
release-team@MIT.EDU
Cc: alexp@MIT.EDU
Date: Thu, 13 Aug 1998 15:37:17 EDT
From: Alex T Prengel <alexp@MIT.EDU>
I just got this additional information.
A.
------- Forwarded Message
Date: Thu, 13 Aug 1998 12:14:55 -0700
To: Alex T Prengel <alexp@MIT.EDU>
From: Stanley Kurdziel <sek@msi.com>
Subject: Re: SR#104851 - core_trap errors
Cc: iris131!104851!srf@msi.com
In-Reply-To: <199808121616.MAA29130@w20-575-106.mit.edu>
References: <Your message of "Tue, 11 Aug 1998 16:41:34 PDT." <3.0.3.32.19980811164134.00707e40@146.202.6.218>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Hi Alex,
Yeah, I would make sure that the eoe.sw32.gfx subsystem is installed on the
systems that are having problems. You can do a "versions | grep
eoe.sw32.gfx" to check.
You are correct that there is a distinction between Insight 400 and 970 in
the graphics they use. Insight 400 uses dgl and Insight 970 can use either
dgl, or OpenGL (both of them can also use X-graphics with an axxess
license). Depending on your system, 970 will run in either OpenGl or dgl
by default. You can override this default however, by running either
"insightII -gl" for dgl, or "insightII -opengl" for OpenGL.
> also- the errors I reported when running 400 on the same machines where
> the problem shows up only happen when you first try to run 970, have it
> crash, then use "set_biosym_context 400" and try to launch 400; it doesn't
> happen when you just log on and launch 400 directly. Any idea why?
It could be because the LD_LIBRARY_PATH is getting set incorrectly when you
use the "set_biosym_context 400" command. Try opening a new shell, and see
if it will then start Insight 400. If that is not the case, then I think
it must be some sort of graphics initialization problem.
In every case that I have dealt with where people were getting the
core_trap error, and I heard back from, there have been one of two
resolutions. Either the problem mysteriously disappears, or it turned out
to be a patch or subsystem issue. I would make sure that all of the N32
subsystems listed in the last email were installed, then install path 1746,
and remove any of the 3135 or predecessor patches. If at this point it
does not work, then I am not sure what the problem is. However, at least
two people just started removing patches and were able to solve the problem
that way. Of course, those patches were installed to solve other problems,
and removing them all is perhaps not the best idea.
I do not think that there are any graphics cards which have incompatibility
problems. I am running on a Z-buffer less right now. I have seen this
error before on this machine, when removing and adding certain patches and
components, but it goes away when the correct subsystems and patches are
installed. If you send me your "hinv" and "/usr/gfx/gfxinfo" output from a
machine with the problem, then I can try to verify that insightII runs on
machines with the same graphics card.
Best Regards,
Stanley
- --
Stanley Kurdziel
Technical Support Specialist
/ \
/ Email:support@msi.com \
\ Toll Free: 800-756-4674 /
\-- Fax: 619-458-0431 --/
/- Phone: 619-546-5509 -\
/-------------------------\
/ Molecular Simulations Inc.\
/ 9685 Scranton Rd. \
\ San Diego, CA 92121-3752 /
\ http://www.msi.com /
\_________________________/
------- End of Forwarded Message
