[5826] in Release_7.7_team

home help back first fref pref prev next nref lref last post

Re: Trouble with onboard video for Dell Optiplex 745 and Athena installs

daemon@ATHENA.MIT.EDU (andrew m. boardman)
Fri Aug 31 12:45:16 2007

Message-Id: <200708311644.l7VGiXwK011729@pothole.mit.edu>
To: Myron Freeman <fletch1@eecs.MIT.EDU>
cc: release-team@MIT.EDU
In-Reply-To: Your message of "Fri, 31 Aug 2007 12:05:38 EDT."
             <D9D44BDE-4CE9-42D9-B190-3B6FB78BC3E9@eecs.mit.edu> 
Date: Fri, 31 Aug 2007 12:44:33 -0400
From: "andrew m. boardman" <amb@MIT.EDU>
X-Spam-Flag: NO
X-Spam-Score: 0.00


Hi!  Thanks for the bug report.  Many questions and possible suggestions:

Could you better describe lala land?  Blank, frozen, something else?  
Have you tried hitting control-P a bunch of times to see if the text mode
login will come up?  Is it just the display wedged or is it the whole
machine?  (Can the affected machines be pinged from elsewhere?)

> Everything appears to revolve around having to use the generic VESA
> drivers as opposed to something specific for the Intel GMA3000 on the
> motherboard.

Which drivers are they using now?  Our cluster 745s come with ATI Radeon
hardware; per lspci:

01:00.0 VGA compatible controller: ATI Technologies Inc RV516 [Radeon X1300/X1550 Series]

On these machines, the default installer uses the vesa driver, but
there's a script that runs at boot time which, on machines with
PUBLIC=true set, swaps the vesa driver for the proprietary ATI drivers
which are *much* faster.  There's a bug with this, however, in that the
script only works on machines which have cluster info in hesiod.
(There's a new version of the script in
/afs/dev/user/amb/scripts/installproprietary-x-drivers-new.sh which
doesn't need cluster info.)  On the other hand, Intel-specific drivers
are included in the base release and the installer should likely have
used the "intel" driver (vice vesa).

Could you send me (or leave in AFS somewhere readable) copies of
/etc/X11/xorg.conf and the output of "lspci" from one of the affected
machines?

Thanks for the bug report; unless something obvious turns up, could I
come visit one of the affected machines?  (Probably next week;
ironically, I was walking by your lab yesterday afternoon and almost
popped in to see how your Athena machines were doing.)

Andrew Boardman/ZINC/ISDA/IS&T/MIT  +1.617.253.4040  amb@mit.edu

home help back first fref pref prev next nref lref last post