I got the NVIDIA drivers working properly after installing them with Yast, while using kernel 2.6.25.5-1.1-default.
However, after I updated my kernel to 2.6.25.18-0.2 , I can only get to Kde, by selecting failsafe. When I try to boot normally, x11 does not start. It fails with loading the NVIDIA drivers (same as when trying to load the NVIDIA drivers with the modprobe command) FATAL: Error inserting nvidia (/lib/modules/2.6.25.20-0.1-default/weak-updates/nvidia.ko): No such device. I have looked, and the file is there, neither did I find anything wrong in the xorg.conf. I can boot with failsafe normally, but no 3d acceleration (as it does not load the NVIDIA drivers). I have tried to fix it for 3 days now, searching on google, and trying different kernel updates and NVIDIA drivers. I think, that downgrading kernel to 2.6.25.5-1.1, and then reinstalling drivers could make it work, but I haven’t tried it yet. I have tried installing them with Yast and the NVIDIA script.
I’m using Suse 11, and have a GeForce 8500gt (no other videocards show up in any device listing).
If anyone can, then please help, as I’m running out of options here.
You need to re-install the drivers when you install or update the kernel.
I reinstalled the drivers every time I updated the kernel. I tried Yast (worked with 2.6.25.5-1.1), and NVIDIA’s script, but I couldn’t get them to work.
Are you reintalled or recompiled the nvidia driver after the kernel update. Please specify what command you’ve used for doing this.
you need to get the kernel-source for the kernel you are using. Yast should have done that, I think.
I install by downloading from nvidia and running their install script. If you grab the kernel sources for your current kernel and then run the installer, it should work. You have to go into init 3 to install the drivers this way.
I installed the kernel-source with Yast, and updated it every time, I updated the kernel. And I installed the drivers with the script from NVIDIA’s site (NVIDIA-Linux-x86-180.22.pkg1.run is the file’s name).
And for reinstalling the driver with the script, I used just a “sh NVIDIA-Linux-x86-180.22.pkg1.run”.
Okay, now check what stays in the nvidia HOW-TO.
I couldn’t find a solution there.
Of course there is no solution - in how-to you don’t have any problem. 
Are you saw this line:
IMPORTANT: You need to recompile and install the nvidia kernel module
after each kernel update.
sh NVIDIA-Linux-x86-173.14.09-pkg1.run -K
?
When you ran the nvidia script, did you exit to console (ctrl+alt+f1)? and then type init 3? before running the installer?
When I installed the drivers with the NVIDIA script, I switched to runlevel 3 (with the “init 3” command). I have also tried running the script with " -K", but that gave me a error in the end.
So, that’s the problem. What is that error message? Are you have kernel-sources / header, gcc installed?
I have the kernel-source and kernel-headers for my current kernel version along with the newest gcc installed. The error that the NVIDIA script displays is:
ERROR: The file '/lib/modules/2.6.25.20-0.1-default/kernel/drivers/video/nvidia.ko` already exists as part of this driver installation.
On the next screen it says:
ERROR: Installation has failed. Please see the file '/var/log/nvidia-installer.log' for details. You may find suggestions on fixing installation problems in the README available on the Linux driver download page at www.nvidia.com.
Try to rename the nvidia.ko and move the entire nvidia file/directory with:
mv nvidia.ko nvidia.ko_old
mv nvidia/ nvidia_old/
in directory /lib/modules/2.6.25.20-0.1-default/kernel/drivers/video/
and try to install it again. Maybe som junk files are left after the last unsuccessful installation.
Good luck.
I renamed the nvidia.ko and nvidia folder, and ran the script again, and got it working fine after a reboot.
Thanks!
Glad to hear!
Thank you for posting back.