Updated Kernel Causing Kernel Panic - Image Provided

[arpsmack]

I recently installed FC5 on my PC. However I couldn't get nvidia drivers to install. I tried nvidia's method of installing via their .run installation package, but I kept getting an error after compiling the kernel interface that "nvidia.ko" could not be loaded.

So I found a 3rd party installation from the livna repository and decided to install that. However, this requires the latest kernel. Upon installation, and reboot, I ran into a kernel panic.

Well, I decided to start over with a fresh new installation of FC5 and update the kernel and other packages first thing. So I installed clean, with all the default options, and ran a "yum update" first thing. After downloading and installing all updates, I rebooted. It attempted to boot the new kernel, and once again I received a kernel panic.

From what I can see on the screen, the new kernel appears to have a problem with mounting the filesystem. I am using two Western Digital rapter drives in a RAID 0 stripe. If I boot into the old kernel, there is not problem and everything works great (except that I still can't install my nvidia drivers).

Here is the image, any ideas?

[ianbigboy]

Check that the SELinux contexts are set correctly on the nvidia drivers. They should match all the other Xorg drivers. Do a "ll -alZ /usr/lib/xorg/modules/drivers/" If they don't match, correct them with "chcon system_u:object_r:lib_t /usr/lib/xorg/modules/drivers/nvidia_drv.*" Then try to starting X up with the nvidia driver.

[arpsmack]

Upon installation, I set SELinux into "Permissive" mode so it wouldn't interfere, but nevertheless, I have time to spare and will try what you suggested. However, this still doesn't resolve the fact that the latest kernel causes a system lockup on boot, while the older default installed kernel works fine. I would think that a newer version of the kernel would increase my compatibility, not decrease.

[JaRaEeZ]

arpsmack : did you update the kernel AND the nvidia drivers (if your using the knod-nvidia) or just the kernel?

[arpsmack]

I did exactly this and only this:

1) Install Fedora Core 5 fresh off the DVD.
2) Ran "yum update" from a terminal
3) Yum proceeded to install 988 MB of updates.
4) I rebooted
5) Kernel panic

I did no messing around with the nvidia drivers with this fresh install. I didn't even download them.

However, after I read iambigboy's post, I attempted to install the nvidia drivers again using the .run package available on their site. I downloaded kernel headers that match the current kernel I am using ( 2.6.15-1.2054_FC5 ) and ran the install package as root upon rebooting into init level 3. It successfully (I presume) compiled a kernel interface and then proceeded once again to tell me that "nvidia.ko" could not be loaded. It says this may be because I provided kernel headers that don't match my kernel (not true, they do match) or because my kernel was compiled using a different version of gcc (also not true, I checked this as well). Actually, here, I uploaded the install log. The results have been exactly the same every time I've tried to run nvidia's installation package. NVIDIA-INSTALL-LOG

Near the bottom there is a little excerpt that appears on screen when the error occurs

ERROR: Unable to load the kernel module 'nvidia.ko'.  This happens most
       frequently when this kernel module was built against the wrong or
       improperly configured kernel sources, with a version of gcc that differs
       from the one used to build the target kernel, or if a driver such as
       rivafb/nvidiafb is present and prevents the NVIDIA kernel module from
       obtaining ownership of the NVIDIA graphics device(s).

[ianbigboy]

Can you post the output of "rpm -qa |grep kernel |sort" and a " uname -a"

[arpsmack]

Sure thing.

[mma@localhost ~]$ rpm -qa |grep kernel |sort
kernel-2.6.15-1.2054_FC5
kernel-2.6.17-1.2174_FC5
kernel-devel-2.6.15-1.2054_FC5
[mma@localhost ~]$ uname -a
Linux localhost.localdomain 2.6.15-1.2054_FC5 #1 Tue Mar 14 15:48:33 EST 2006 i686 athlon i386 GNU/Linux

[ianbigboy]

I would try to delete the updated kernel. Then try reinstalling it with it's devel package.