It seems that the latest Nvidia drivers (340.32_k3.11.6_4-30.1) from the official repo (16.08.2014 04:46:00) for 13.1 x64 are broken. After zypper update libglx modules failes to load:
11.766] (II) LoadModule: "glx"
11.861] (II) Loading /usr/lib64/xorg/modules/updates/extensions/libglx.so
11.861] (EE) Failed to load /usr/lib64/xorg/modules/updates/extensions/libglx.so: /usr/lib64/xorg/modules/updates/extensions/libglx.so: cannot open shared object file: No such file or directory
11.861] (II) UnloadModule: "glx"
11.861] (II) Unloading glx
11.861] (EE) Failed to load module "glx" (loader failed, 7)
ordos:/usr/lib64/xorg/modules/updates/extensions # ls -la /usr/lib64/xorg/modules/updates/extensions
total 12
drwxr-xr-x 3 root root 4096 Aug 16 22:07 .
drwxr-xr-x 8 root root 4096 Aug 15 10:46 ..
lrwxrwxrwx 1 root root 16 Aug 16 22:07 libglx.so -> libglx.so.340.32
drwxr-xr-x 2 root root 4096 Aug 16 22:07 nvidia
libglx doesnt exist…
Anyway, manual installation of the driver (.run) works fine.
I didn’t run this update yet. Mostly because the I’m cautious due to a previous fiasco and it threw up a complaint about the signing key having been changed.
how can I fix the poblem with nvidia driver update?
I’ve tried to create the link, but I miss the /usr/lib64/xorg/modules/updates/extensions/nvidia folder. I’ve tried to reinstall the nvidia driver and xorg-x11-server package and it didn’t help.
The problem should already be fixed with the latest packages from yesterday.
I’ve tried to create the link, but I miss the /usr/lib64/xorg/modules/updates/extensions/nvidia folder.
What openSUSE version are you using?
There is no /usr/lib64/xorg/modules/updates/extensions/nvidia folder on 13.1 or 12.3. The nvidia rpms install libglx.so.340.32 and the libglx.so symlink directly to the /usr/lib64/xorg/modules/updates/extensions folder.
They got installed to the nvidia folder by mistake, and this caused the problem in the first place I think.
So if you don’t have the nvidia folder, you should not have a problem either.
Can you post a directory listing maybe?
ls -la /usr/lib64/xorg/modules/updates/extensions
I’ve tried to reinstall the nvidia driver and xorg-x11-server package and it didn’t help.
Which nvidia and kernel packages do you actually have installed now?
rpm -qa | egrep "kernel|nvidia"
Reinstalling xorg-x11-server cannot possibly fix the nvidia driver. It would even break it if you would have installed it “the hard way”, i.e. via the .run installer from nvidia’s homepage.
And maybe you should describe what problem you actually have. Why do you think the nvidia driver is not working?
It would also be interesting if it ever worked on your system.
A side-note: have you tried to reboot already? On some systems at least the driver does not seem to work at the first boot after installation/update of the driver packages. This should be “fixed” on subsequent boots though.
Thank you for the answer. System (openSuse 13.1) worked properly till yesterday update (I´m not sure what was updated, but the nvidia driver was there). Now I have problem, that X server don´t start.
Should not have happened. The problem with the packages only broke GLX, i.e. hardware 3D support.
This prevents GNOME from working, but the X server should still start and other DE’s should work.
Can you boot to recovery mode? (“Advanced Options” in the boot menu)
The folder
/usr/lib64/xorg/modules/updates/extensions
is empty
Yes. That’s because you don’t have nvidia-glG03 installed, this contains libglx.
Install it and the folder should contain the necessary files.
Looks fine except for the missing nvidia-glG03 that is needed for hardware accelerated 3D support as mentioned.
But even without that, at least the Xserver should start.
I do see a kernel-debug in that list.
Why do you have that?
Which kernel are you actually using?
uname -a
If it’s kernel-debug this would of course explain why the nvidia driver does not work. You don’t have the driver’s kernel modules installed for kernel-debug (they don’t exist).
You should better uninstall that kernel.
And try to select a different one in “Advanced Options” in the boot menu. Does X start then?
If that doesn’t help either, try to reinstall the kernel module:
Note 2 of the same package. Can’t imagen that would cause a problme but…[/QUOTE]
Please stop talking nonsense!
I already told you a few times in the last days that it’s perfectly valid and normal to have multiple kmp packages installed. That’s the multiversion feature that allows multiple kernel versions installed at the same time.
Those packages do not have a problem!
And he doesn’t have any kernel-default that you refer to now I suppose.
And just before you mention it: the nvidia packages cannot possibly pull in kernel-debug, as there are no nvidia packages for kernel-debug.
It is not nonsense that 2 package of the same version are installed. Would it cause a problem probably not but note there are only 2 kernels. So something went wrong somewhere it arrive at this state…
Linux Kosmik 3.11.10-21-desktop #1 SMP PREEMPT Mon Jul 21 15:28:46 UTC 2014 (9a9565d) x86_64 x86_64 x86_64 GNU/Linux
and the installation of nvidia-gfxG03-kmp-desktop finished with some error:
Získávám: nvidia-gfxG03-kmp-desktop-340.32_k3.11.6_4-31[dokončeno (202,4 KiB/s)]
(1/1) Instaluje se: nvidia-gfxG03-kmp-desktop-340.32_k3.11.6_4-31.1 .[dokončeno]
Další výstup programu rpm:
make: *** /usr/src/linux-obj/x86_64/desktop: No such file or directory. Stop.
/usr/src/kernel-modules/nvidia-340.32-desktop /
NVIDIA: calling KBUILD...
make: *** /usr/src/linux: No such file or directory. Stop.
NVIDIA: left KBUILD.
nvidia.ko failed to build!
make: *** [nvidia.ko] Error 1
/
install: cannot stat '/usr/src/kernel-modules/nvidia-340.32-desktop/nvidia.ko': No such file or directory
Modprobe blacklist files have been created at /etc/modprobe.d to prevent Nouveau from loading. This can be reverted by deleting /etc/modprobe.d/nvidia-*.conf.
*** Reboot your computer and verify that the NVIDIA graphics driver can be loaded. ***
WARNING: Can't read module /lib/modules/3.11.10-17-desktop/weak-updates/updates/nvidia-uvm.ko: No such file or directory
WARNING: Can't read module /lib/modules/3.11.10-21-desktop/weak-updates/updates/nvidia-uvm.ko: No such file or directory
I’ve removed the debug kernel, but it didn’t help.
I can boot in recovery mode and the X server runs.
Right. That would explain of course why the driver is not working.
make: *** /usr/src/linux-obj/x86_64/desktop: No such file or directory. Stop.
/usr/src/kernel-modules/nvidia-340.32-desktop /
NVIDIA: calling KBUILD...
make: *** /usr/src/linux: No such file or directory. Stop.
...
Your missing vital kernel development files so the module fails to build, resulting in:
install: cannot stat '/usr/src/kernel-modules/nvidia-340.32-desktop/nvidia.ko': No such file or directory
Modprobe blacklist files have been created at /etc/modprobe.d to prevent Nouveau from loading. This can be reverted by deleting /etc/modprobe.d/nvidia-*.conf.
*** Reboot your computer and verify that the NVIDIA graphics driver can be loaded. ***
WARNING: Can't read module /lib/modules/3.11.10-17-desktop/weak-updates/updates/nvidia-uvm.ko: No such file or directory
WARNING: Can't read module /lib/modules/3.11.10-21-desktop/weak-updates/updates/nvidia-uvm.ko: No such file or directory
Something seems to have deleted those files/dirs for whatever reason.
Re-install the corresponding packages, and try to install the nvidia kmp packages again:
sudo zypper in -f kernel-desktop-devel kernel-devel
sudo zypper in -f nvidia-gfxG03-kmp-desktop nvidia-uvm-gfxG03-kmp-desktop
(it’s possible it has something to do with my router, as I have to use a temp ATM. But everything else is OK, just nvidia. I already changed the DNS servers in the router but it didn’t help)
And which file is not found?
What URL are you using?
I had no problem updating the driver on one system yesterday, and at least Skeletonix seemed to have been able to access the repo without problem as well. Otherwise he couldn’t have re-installed the kmp packages.
zypper up
Retrieving repository 'packman' metadata ......................................................[done]
Building repository 'packman' cache ...........................................................[done]
Loading repository data...
Reading installed packages...
The following 18 package updates will NOT be installed:
amarok gstreamer-0_10-plugin-gnomevfs kde4-kgreeter-plugins kdebase4-openSUSE
kdebase4-runtime-branding-openSUSE kdebase4-workspace kdebase4-workspace-branding-openSUSE
kdebase4-workspace-ksysguardd kdebase4-workspace-liboxygenstyle kdebase4-workspace-plasma-calendar
kdm kscreen kwin libepub0 libfreetype6 libfreetype6-32bit python-kdebase4 python-setuptools
The following 2 NEW packages are going to be installed:
nvidia-gfxG03-kmp-desktop-340.32_k3.11.6_4-30.1
nvidia-uvm-gfxG03-kmp-desktop-340.32_k3.11.6_4-30.1
The following 8 packages are going to be upgraded:
h264enc kmediafactory kwrite libktexteditor libyui-qt-pkg5 nvidia-computeG03 nvidia-glG03
x11-video-nvidiaG03
8 packages to upgrade, 2 new.
Overall download size: 73.3 MiB. After the operation, additional 100.2 MiB will be used.
Continue? [y/n/? shows all options] (y): y
Retrieving package nvidia-gfxG03-kmp-desktop-340.32_k3.11.6_4-30.1.x86_64
(1/10), 3.6 MiB ( 32.3 MiB unpacked)
Retrieving: nvidia-gfxG03-kmp-desktop-340.32_k3.11.6_4-30.1.x86_64.rpm .......................[error]
File './x86_64/nvidia-gfxG03-kmp-desktop-340.32_k3.11.6_4-30.1.x86_64.rpm' not found on medium 'http://download.nvidia.com/opensuse/13.1/'