Bug 479329

NVIDIA was too fast putting the new modules online. This is not really controllable from us here.

I hope that we can release the updated kernel today, but we needed to some other fixes to make it install good enough.

There should be better coordination between openSUSE and Nvidia.  At least report the bug to them and work on a mutual resolution.

*** Bug 479588 has been marked as a duplicate of this bug. ***

*** Bug 478436 has been marked as a duplicate of this bug. ***

the kernel update that matches this bug was just released and should be available in 1 - 2 hours for everyone, and then it should work :)

Here's how it works:

# zypper in x11-video-nvidiaG02 nvidia-gfxG02-kmp-default
Loading repository data...
Reading installed packages...
Resolving package dependencies...

The following NEW packages are going to be installed:
  nvidia-gfxG02-kmp-default x11-video-nvidiaG02 


Overall download size: 13.4 M. After the operation, additional 59.5 M will be used.
Continue? [YES/no]: y
Retrieving package x11-video-nvidiaG02-180.29-0.1.x86_64 (1/2), 11.3 M (48.7 M unpacked)
Retrieving: x11-video-nvidiaG02-180.29-0.1.x86_64.rpm [done (685.8 K/s)]
Installing: x11-video-nvidiaG02-180.29-0.1 [done]
Retrieving package nvidia-gfxG02-kmp-default-180.29_2.6.27.18_0.3-0.1.x86_64 (2/2), 2.1 M (10.9 M unpacked)
Retrieving: nvidia-gfxG02-kmp-default-180.29_2.6.27.18_0.3-0.1.x86_64.rpm [startRetrieving: nvidia-gfxG02-kmp-default-180.29_2.6.27.18_0.3-0.1.x86_64.rpm [20% (Retrieving: nvidia-gfxG02-kmp-default-180.29_2.6.27.18_0.3-0.1.x86_64.rpm [55% (Retrieving: nvidia-gfxG02-kmp-default-180.29_2.6.27.18_0.3-0.1.x86_64.rpm [88% (Retrieving: nvidia-gfxG02-kmp-default-180.29_2.6.27.18_0.3-0.1.x86_64.rpm [done (639.2 K/s)]
Installing: nvidia-gfxG02-kmp-default-180.29_2.6.27.18_0.3-0.1 [error]
Installation of nvidia-gfxG02-kmp-default-180.29_2.6.27.18_0.3-0.1 failed:
(with --nodeps --force) Error: Subprocess failed. Error: RPM failed: cat: /var/tmp/weak-modules2.AeUIcF/krel-nvidia-gfxG02-kmp-default-180.29_2.6.27.18_0.3-0.1: No such file or directory
/usr/lib/module-init-tools/weak-modules2: line 53: /var/tmp/weak-modules2.AeUIcF/modules-nvidia-gfxG02-kmp-default-180.29_2.6.27.18_0.3-0.1: No such file or directory
/usr/lib/module-init-tools/weak-modules2: line 476: /var/tmp/weak-modules2.AeUIcF/basenames-nvidia-gfxG02-kmp-default-180.29_2.6.27.18_0.3-0.1: No such file or directory
error: %post(nvidia-gfxG02-kmp-default-180.29_2.6.27.18_0.3-0.1.x86_64) scriptlet failed, exit status 1


Abort, retry, ignore? [A/r/i]:

this looks bad.

please make sure you have the module-init-tools installed that we released yesterday before proceeding.

what is: rpm -q module-init-tools ?

# rpm -q module-init-tools
module-init-tools-3.4-56.9.1.x86_64

My system is fully updated.

(In reply to comment #8)
> # rpm -q module-init-tools
> module-init-tools-3.4-56.9.1.x86_64
                               ^^^^^^

Did you tweak your rpm popt config to show the architecture in rpm -q output? The weak-modules2 script does use plain rpm -qa in one place and such a change could indeed cause the failure you are seeing. The nvidia rpm installs fine here btw.

Created attachment 276360 [details]
fixed weak-modules2

Please save this file as /usr/lib/module-init-tools/weak-modules2 and try again.

I don't suffer the same problem as Igor. After the updated kernel hit the update repository, the problem I originally reported is fixed for me (using the 2.6.27.19-3.2-pae kernel). Hence I cannot function as the info provider and will therefore change "Info Provider" to Igor.

Michal, you were right, I tweaked the rpm configuration a little bit by adding the following line into my /etc/rpm/macros:

%_query_all_fmt %%{name}-%%{version}-%%{release}.%%{arch}

Erasing that line did the trick, thank you very much. I guess that tweaking rpm was not a good idea.

OK, thanks for confirming. In the meantime, I submitted the fix from comment #10 to Factory.

Update released for: module-init-tools
Products:
openSUSE 11.1 (debug, i586, ppc, x86_64)