X does not work with ATI based machines

Antonio Olivares olivares14031 at yahoo.com
Thu May 14 00:38:53 UTC 2009




--- On Wed, 5/13/09, Adam Williamson <awilliam at redhat.com> wrote:

> From: Adam Williamson <awilliam at redhat.com>
> Subject: Re: X does not work with ATI based machines
> To: "For testers of Fedora Core development releases" <fedora-test-list at redhat.com>
> Date: Wednesday, May 13, 2009, 4:30 PM
> On Wed, 2009-05-13 at 14:24 -0700,
> Antonio Olivares wrote:
> 
> > I logged into level 3 and startx and its
> working.  The other one, however is not.  I see
> the following :
> > 
> > ERROR:  failed in exec of /bin/echo: input/output
> error
> > ERROR:  failed in exec of /bin/sleep:
> input/output error
> > ERROR:  failed in exec of /bin/sleep: 
> input/output error
> > Kernel panic - not syncing:  Attempted to kill
> init!
> > [drm:radeonfb_panic] : ERROR:  panic occurred,
> switching back to text console.  
> > 
> > I started in level 3 and this is the message that I
> see.  The caps lock and scroll lock lights in the
> keyboard are flashing at me also.  I will try to login
> to level 3 under a previous kernel and see if I can boot and
> report back.
> 
> That's got nothing to do with your graphics card. The
> system's stuffed.
> Likely the RAM or hard disk. But something's dead (or just
> loose).
> 
> BTW, FWIW, I always find the output of plain lspci about as
> useful as a
> chocolate teapot. What I want to know is the PCI ID, and it
> doesn't give
> you that. "lspci -nn" does, though. I don't know if the X
> maintainers
> prefer that output or if they're able to work with just
> plain lspci...
> -- 

One machine now works correctly, but this one does not.  Any kernel I choose bombs out with the same error message :(

Here's smolt profile of that machine in case it is relevant.

http://www.smolts.org/client/show/pub_808fc0dc-61b4-4f57-a41a-62cdc710058d

According to smolt it is an:

ATI Technologies Inc Radeon 2100 (wiki)
http://www.smolts.org/reports/view_device/Radeon%202100

I just tried to update that machine.  I shut it down and I tried to restart and that is what I get.  It runs windows Ok, so I conclude that some update broke it.  Ram is fine, why?  Windows does not complain.  Disk is fine, why?  Windows does not complain and disk does not screech/scratch/ or output weird sounds.  This is one of the machines that the little annoying program said that one or more hard disks are failing by the way :(, It was working, it worked since late March and before.  Can the disk really be failing? only on the Fedora side and not on the Windows Side?  This would be really sad and I would get feelings to just nuke Fedora and leave the evil empire installed to the whole disk and hope that it truly fails and then I can conclude that the utility was right and apologize to all for doubting "One or more hard disks are failing" :(, but the laws of probability tell me otherwise.  I have a hard drive that scratches/screams/(I can hear sounds
 coming from it) for 3 years and it is still running ?  This is a weird case, till that happens in this disks case, I will conclude that the utility was right, but for now it is wrong and this system is indeed nuked, how sad :(

I tried booting in rescue mode, and chmount /mnt/sysimage and check out things, but I can't make it work :(, I get that /bin/sh is not there or other messages.  Should I just nuke it and try to reinstall/install Fedora 11 Preview on it and update it back to where it was?  Or just drop it in peace, Fedora 11 is very close to becoming a reality.  What are your recommendations, can this system be rescued or is it badly damaged/hosed?  

Thanks,

Antonio  


      




More information about the fedora-test-list mailing list