[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#285081: marked as done (xbase-clients: startx could try to fail back to vesa|vga if X does not start up because of a driver issue)

Your message dated Fri, 13 Apr 2007 01:08:29 +0200
with message-id <461EBBED.4000000@ens-lyon.org>
and subject line Bug#285081: closed by Brice Goglin <Brice.Goglin@ens-lyon.org> (Bug#285081: xbase-clients: startx could try to fail back to vesa|vga if X does not start up because of a driver issue)
has caused the attached Bug report to be marked as done.

This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.

(NB: If you are a system administrator and have no idea what I am
talking about this indicates a serious mail system misconfiguration
somewhere.  Please contact me immediately.)

Debian bug tracking system administrator
(administrator, Debian Bugs database)
--- Begin Message --- 
Package: xbase-clients
Version: 4.3.0.dfsg.1-8
Priority: wishlist

I have thought on this issue while fixing a sarge installation in a Dell 
desktop  (I believe it was an OptiPlex 170L) which should use the i810 
device driver. The problem with it, however, is that after a standard 
Debian installation (sarge rc2 net install) even though the X configuration 
discovered the i810 card and suggested the driver the 2.6.8 kernel would 
not load the driver and X wouldn't work.

I offered the user an alternate possibility, that is, configuring X with 
the vesa driver, it wouldn't have all the bells&whistles, but it would 
probably work. And it did work, as it turned out.

Could it be possible (I'm not sure if this should be done by startx or if 
this should be implemented in other tools, for example, gdm) to have X 
failback to a "safe" configuration for VGA displays if it's unable to load 
a given driver.

The startx could try to load the XF86Config-4 and, if a driver error was
detected (cannot setup the configured display) try to startup with an
alternate configuration in which the 'vesa' (or 'vga'?) driver is used for
the video card instead? (maintaining the other settings)

Doing that and providing some proper input (through log files, maybe 
syslog) could avoid having users wonder why don't they get the fancy 
graphics they are expecting and instead get thrown to an obscure shell.

Googling I've found this has also been asked for Red Hat [1] already :-)

Regards

Javier

[1] https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=102289

--- End Message ---
--- Begin Message --- 
Version: 2:1.2.99.905-1


Javier Fernández-Sanguino Peña wrote:
> reopen 285081 
> tags 285081 fixed-in-experimental
> thanks
>
> On Thu, Apr 12, 2007 at 09:00:09PM +0000, Debian Bug Tracking System wrote:
>   
>> With the latest X server (currently in experimental), you don't even
>> need a configuration file. The server takes care of choosing the right
>> driver for you. So this bug does not matter anymore, closing.
>>     
>
> ? Since when do we close bugs because a feature is fixed in experimental? We
> have 'fixed-in-experimental' for that! Reopening and tagging.
>   

No need to use fixed-in-experimental, the recommended way to handle this
is to close with a Version: pseudo-header. The BTS will mark as fixed
with this version first, and take care of archiving the bug as soon as
this package version or later is uploaded to unstable.

Sorry I forgot the pseudo-header earlier, closing correctly now.

Brice

--- End Message ---
Reply to: