[SOLVED] Re: X gone

[Jason Filippou <jason.filippou@gmail.com>]

Dear list,

I am really really sorry for wasting your time with such a lengthy
message. Once I understood that I could actually load a shell with
ctrl + alt + f1 from my newest kernel I was able to run the 256.35
NVIDIA installer (given that I had all linux-headers and linux-source
packages for that kernerl installed) that I had and everything runs
smootlhy now.

Really sorry once again,

Jason

On Wed, Dec 29, 2010 at 2:52 PM, Jason Filippou
<jason.filippou@gmail.com> wrote:
> Hello list,
>
> I was having some trouble running openGL in my Debian testing,
> 2.6.32-5-686 kernel version system. I have an NVIDIA card and ran the
> driver provided by nvidia-kernel-dkms. The problem was that when I ran
> glxinfo, I got:
>
> glxinfo: symbol lookup error: /usr/lib/libGL.so.1: undefined symbol: _nv000006gl
>
> so I decided to google a bit, ending up on an arch linux distribution
> link which suggested that a certain filesystem folder be deleted:
> https://bugs.archlinux.org/task/12771. This folder was /usr/lib/tls.
>
> After I -apparently wrongfully- deleted it, my  2.6.32-5-686 kernel
> gets stuck during loading of the X server, and once I see or skip
> through the error messages, I don't even get a shell to do some typing
> to. Luckily, I had a 2.6.32-3-686 kernel in my system, which,
> unfortunately, doesn't load X either. This particular kernel version,
> does, however, give me a shell, yet I have tried lots of things, such
> as reinstalling the 5-686 kernel ( I don't want to mess around with
> the 3-686 image and end up with an unusable installation),
> re-installing the xserver-xorg package, even removing
> nvidia-kernel-dkms and trying to install one of three NVIDIA
> installers I already had in my home folder: 190.53, 195.36.08,
> 256.35). However, trying to run every one of those installer scripts
> tells me that I do not have the appropriate source files for the
> kernel that I'm running (which, I would like to remind you, is
> currently 2.6.32-3-686). I tried installing the appropriate
> linux-headers package, however, there seems to not exist a
> linux-headers-2.6.32-3-686 package. I also reinstalled
> linux-source-2.6.32 and linux-source-2.6, yet the NVIDIA installers
> will still not run. Installing nvidia-kernel-source doesn't help
> either.
>
> So I'm stuck with a kernel I can not even run a shell from, and
> another older one which doesn't appear to be of much help to me.
> Anything I could do?
>
> Currently posting from my laptop. When I run startx, I get a bunch of
> errors, such as:
>
> Error: API mismatch: the NVIDIA kernel module has version 190.53, but
> this NVIDIA driver component has version 195.36.31. Please make sure
> that the kernel module and all NVIDIA driver components have the same
> version.
> (EE) Dec 29 14:38:15 NVIDIA(0): Failed to initialize the NVIDIa kernel module.
> .
> .
> .
> (EE) NVIDIA(0): ***Aborting***
> (EE) Screen(s) found, but none have a usable configuration.
>
> Fatar server error:
> no screens found.
> .
> .
> .
> xinit: No such file or directory (errno 2): unable to connect to X server
> xinit: No such process (errno 3): Server error.
>
> Thanks for any help, as I'm not sure what to do here. Please do CC me
> as I am not on the list.
>
> --
> Jason Filippou
>
> http://cgi.di.uoa.gr/~std06142
>



-- 
Jason Filippou

http://cgi.di.uoa.gr/~std06142