when trying to upgrade to the newest version of silverblue (which includes a Linux kernel upgrade: kernel 5.0.17-300.fc30 → 5.1.5-300.fc30), i get following error:
error: Running %post for akmod-nvidia: Executing bwrap(/bin/sh): Child process killed by signal 1; run journalctl -t 'rpm-ostree(akmod-nvidia.post)' for more information
does anyone have a suggestion how to proceed?
Can you try running the journalctl command it gave and post the output?
Jun 01 13:01:17 cu rpm-ostree(akmod-nvidia.post)[2583]: Building /usr/src/akmods/nvidia-kmod-418.56-2.fc30.src.rpm for kernel 5.1.5-300.fc30.x86_64
Jun 01 13:01:44 cu rpm-ostree(akmod-nvidia.post)[2583]: 525 | .fault = uvm_vm_fault_wrapper,
Jun 01 13:01:44 cu rpm-ostree(akmod-nvidia.post)[2583]: | ^~~~~~~~~~~~~~~~~~~~
Jun 01 13:01:44 cu rpm-ostree(akmod-nvidia.post)[2583]: /tmp/akmodsbuild.DkhJGFhu/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-300.fc30.x86_64/nvidia-uvm/uvm8.c:525:21: note: (near initialization for>
Jun 01 13:01:44 cu rpm-ostree(akmod-nvidia.post)[2583]: /tmp/akmodsbuild.DkhJGFhu/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-300.fc30.x86_64/nvidia-uvm/uvm8.c:526:21: error: initialization of 'vm_f>
Jun 01 13:01:44 cu rpm-ostree(akmod-nvidia.post)[2583]: 526 | .page_mkwrite = uvm_vm_fault_wrapper,
Jun 01 13:01:44 cu rpm-ostree(akmod-nvidia.post)[2583]: | ^~~~~~~~~~~~~~~~~~~~
Jun 01 13:01:44 cu rpm-ostree(akmod-nvidia.post)[2583]: /tmp/akmodsbuild.DkhJGFhu/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-300.fc30.x86_64/nvidia-uvm/uvm8.c:526:21: note: (near initialization for>
Jun 01 13:01:44 cu rpm-ostree(akmod-nvidia.post)[2583]: /tmp/akmodsbuild.DkhJGFhu/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-300.fc30.x86_64/nvidia-uvm/uvm8.c:609:21: error: initialization of 'vm_f>
Jun 01 13:01:44 cu rpm-ostree(akmod-nvidia.post)[2583]: 609 | .fault = uvm_vm_fault_sigbus_wrapper,
Jun 01 13:01:44 cu rpm-ostree(akmod-nvidia.post)[2583]: | ^~~~~~~~~~~~~~~~~~~~~~~~~~~
Jun 01 13:01:44 cu rpm-ostree(akmod-nvidia.post)[2583]: /tmp/akmodsbuild.DkhJGFhu/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-300.fc30.x86_64/nvidia-uvm/uvm8.c:609:21: note: (near initialization for>
Jun 01 13:01:44 cu rpm-ostree(akmod-nvidia.post)[2583]: cc1: some warnings being treated as errors
Jun 01 13:01:44 cu rpm-ostree(akmod-nvidia.post)[2583]: make[2]: *** [scripts/Makefile.build:276: /tmp/akmodsbuild.DkhJGFhu/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-300.fc30.x86_64/nvidia-uvm/uvm>
Jun 01 13:01:44 cu rpm-ostree(akmod-nvidia.post)[2583]: make[2]: *** Waiting for unfinished jobs....
Jun 01 13:01:44 cu rpm-ostree(akmod-nvidia.post)[2583]: ./tools/objtool/objtool orc generate --module --no-fp --retpoline /tmp/akmodsbuild.DkhJGFhu/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-30>
Jun 01 13:01:44 cu rpm-ostree(akmod-nvidia.post)[2583]: ./tools/objtool/objtool orc generate --module --no-fp --retpoline /tmp/akmodsbuild.DkhJGFhu/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-30>
Jun 01 13:01:44 cu rpm-ostree(akmod-nvidia.post)[2583]: ./tools/objtool/objtool orc generate --module --no-fp --retpoline /tmp/akmodsbuild.DkhJGFhu/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-30>
Jun 01 13:01:44 cu rpm-ostree(akmod-nvidia.post)[2583]: ./tools/objtool/objtool orc generate --module --no-fp --retpoline /tmp/akmodsbuild.DkhJGFhu/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-30>
Jun 01 13:01:44 cu rpm-ostree(akmod-nvidia.post)[2583]: ./tools/objtool/objtool orc generate --module --no-fp --retpoline /tmp/akmodsbuild.DkhJGFhu/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-30>
Jun 01 13:01:44 cu rpm-ostree(akmod-nvidia.post)[2583]: ./tools/objtool/objtool orc generate --module --no-fp --retpoline /tmp/akmodsbuild.DkhJGFhu/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-30>
Jun 01 13:01:44 cu rpm-ostree(akmod-nvidia.post)[2583]: ./tools/objtool/objtool orc generate --module --no-fp --retpoline /tmp/akmodsbuild.DkhJGFhu/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-30>
Jun 01 13:01:44 cu rpm-ostree(akmod-nvidia.post)[2583]: make[1]: *** [Makefile:1575: _module_/tmp/akmodsbuild.DkhJGFhu/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-300.fc30.x86_64] Error 2
Jun 01 13:01:44 cu rpm-ostree(akmod-nvidia.post)[2583]: make[1]: Leaving directory '/usr/src/kernels/5.1.5-300.fc30.x86_64'
Jun 01 13:01:44 cu rpm-ostree(akmod-nvidia.post)[2583]: make: *** [Makefile:79: modules] Error 2
Jun 01 13:01:44 cu rpm-ostree(akmod-nvidia.post)[2583]: error: Bad exit status from /var/tmp/rpm-tmp.OQ3Zt1 (%build)
Jun 01 13:01:44 cu rpm-ostree(akmod-nvidia.post)[2583]: RPM build errors:
Jun 01 13:01:44 cu rpm-ostree(akmod-nvidia.post)[2583]: cannot open Packages database in /var/lib/rpm
Jun 01 13:01:44 cu rpm-ostree(akmod-nvidia.post)[2583]: user mockbuild does not exist - using root
Jun 01 13:01:44 cu rpm-ostree(akmod-nvidia.post)[2583]: group mock does not exist - using root
Jun 01 13:01:44 cu rpm-ostree(akmod-nvidia.post)[2583]: user mockbuild does not exist - using root
Jun 01 13:01:44 cu rpm-ostree(akmod-nvidia.post)[2583]: group mock does not exist - using root
Jun 01 13:01:44 cu rpm-ostree(akmod-nvidia.post)[2583]: cannot open Packages database in /var/lib/rpm
Jun 01 13:01:44 cu rpm-ostree(akmod-nvidia.post)[2583]: Bad exit status from /var/tmp/rpm-tmp.OQ3Zt1 (%build)
Jun 01 13:04:17 cu rpm-ostree(akmod-nvidia.post)[2583]: Building /usr/src/akmods/nvidia-kmod-418.56-2.fc30.src.rpm for kernel 5.1.5-300.fc30.x86_64
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: 525 | .fault = uvm_vm_fault_wrapper,
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: | ^~~~~~~~~~~~~~~~~~~~
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: /tmp/akmodsbuild.Tx8vulIH/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-300.fc30.x86_64/nvidia-uvm/uvm8.c:525:21: note: (near initialization for>
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: /tmp/akmodsbuild.Tx8vulIH/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-300.fc30.x86_64/nvidia-uvm/uvm8.c:526:21: error: initialization of 'vm_f>
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: 526 | .page_mkwrite = uvm_vm_fault_wrapper,
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: | ^~~~~~~~~~~~~~~~~~~~
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: /tmp/akmodsbuild.Tx8vulIH/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-300.fc30.x86_64/nvidia-uvm/uvm8.c:526:21: note: (near initialization for>
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: /tmp/akmodsbuild.Tx8vulIH/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-300.fc30.x86_64/nvidia-uvm/uvm8.c:609:21: error: initialization of 'vm_f>
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: 609 | .fault = uvm_vm_fault_sigbus_wrapper,
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: | ^~~~~~~~~~~~~~~~~~~~~~~~~~~
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: /tmp/akmodsbuild.Tx8vulIH/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-300.fc30.x86_64/nvidia-uvm/uvm8.c:609:21: note: (near initialization for>
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: cc1: some warnings being treated as errors
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: make[2]: *** [scripts/Makefile.build:276: /tmp/akmodsbuild.Tx8vulIH/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-300.fc30.x86_64/nvidia-uvm/uvm>
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: | ^~~~~~~~~~~~~~~~~~~~
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: /tmp/akmodsbuild.Tx8vulIH/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-300.fc30.x86_64/nvidia-uvm/uvm8.c:525:21: note: (near initialization for>
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: /tmp/akmodsbuild.Tx8vulIH/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-300.fc30.x86_64/nvidia-uvm/uvm8.c:526:21: error: initialization of 'vm_f>
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: 526 | .page_mkwrite = uvm_vm_fault_wrapper,
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: | ^~~~~~~~~~~~~~~~~~~~
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: /tmp/akmodsbuild.Tx8vulIH/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-300.fc30.x86_64/nvidia-uvm/uvm8.c:526:21: note: (near initialization for>
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: /tmp/akmodsbuild.Tx8vulIH/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-300.fc30.x86_64/nvidia-uvm/uvm8.c:609:21: error: initialization of 'vm_f>
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: 609 | .fault = uvm_vm_fault_sigbus_wrapper,
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: | ^~~~~~~~~~~~~~~~~~~~~~~~~~~
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: /tmp/akmodsbuild.Tx8vulIH/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-300.fc30.x86_64/nvidia-uvm/uvm8.c:609:21: note: (near initialization for>
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: cc1: some warnings being treated as errors
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: make[2]: *** [scripts/Makefile.build:276: /tmp/akmodsbuild.Tx8vulIH/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-300.fc30.x86_64/nvidia-uvm/uvm>
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: make[2]: *** Waiting for unfinished jobs....
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: ./tools/objtool/objtool orc generate --module --no-fp --retpoline /tmp/akmodsbuild.Tx8vulIH/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-30>
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: ./tools/objtool/objtool orc generate --module --no-fp --retpoline /tmp/akmodsbuild.Tx8vulIH/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-30>
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: ./tools/objtool/objtool orc generate --module --no-fp --retpoline /tmp/akmodsbuild.Tx8vulIH/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-30>
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: ./tools/objtool/objtool orc generate --module --no-fp --retpoline /tmp/akmodsbuild.Tx8vulIH/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-30>
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: ./tools/objtool/objtool orc generate --module --no-fp --retpoline /tmp/akmodsbuild.Tx8vulIH/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-30>
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: ./tools/objtool/objtool orc generate --module --no-fp --retpoline /tmp/akmodsbuild.Tx8vulIH/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-30>
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: ./tools/objtool/objtool orc generate --module --no-fp --retpoline /tmp/akmodsbuild.Tx8vulIH/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-30>
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: make[1]: *** [Makefile:1575: _module_/tmp/akmodsbuild.Tx8vulIH/BUILD/nvidia-kmod-418.56/_kmod_build_5.1.5-300.fc30.x86_64] Error 2
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: make[1]: Leaving directory '/usr/src/kernels/5.1.5-300.fc30.x86_64'
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: make: *** [Makefile:79: modules] Error 2
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: error: Bad exit status from /var/tmp/rpm-tmp.kCtvQa (%build)
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: RPM build errors:
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: cannot open Packages database in /var/lib/rpm
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: user mockbuild does not exist - using root
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: group mock does not exist - using root
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: user mockbuild does not exist - using root
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: group mock does not exist - using root
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: cannot open Packages database in /var/lib/rpm
Jun 01 13:04:44 cu rpm-ostree(akmod-nvidia.post)[2583]: Bad exit status from /var/tmp/rpm-tmp.kCtvQa (%build)
-- Reboot --
Looks like kernel 5.1 just made stable right before the weekend, but nvidia 430.14, which is needed for 5.1, hasn’t been pushed yet. Most likely builds will start up again tomorrow, so the best advise I can give is to wait it out a bit for 430.14 to be pushed.
1 Like
Thanks for the advice, i was hoping it was something like that and i just had to wait a little
This seems to be happening to other akmods too (e.g. wl).
To be honest, I’m glad this errors when updating rather than potentially being left with a broken system, so kudos to Silverblue.
There is already an updated akmod-nvidia 418.74 that is compatible with 5.1 in repos.
So if you have kept the rpmfusion-nonfree yum repository, you should have received the update few weeks “before” the kernel 5.1 update was moved to stable (in either fedora or silverblue).
The same apply with wl-kmod (despite there is another issue there because of wpa_supplicant incompatible feature that is been worked on).
About been left with a broken system with nvidia, this shouldn’t be the case anyone thx for the nvidia-fallback service that I’ve designed (as RPM Fusion nvidia package maintainer) on the original idea from Hans de Goede , from Red Hat.
Basically, if anything broken with nvidia, the fallback service will enable nouveau so you won’t have a broken system.
1 Like
thanks for the information, i ended up having to reinstall rpm-fusion
can you be more specific? i just came across same issue with following kernel:
kernel 5.1.18-300.fc30 → 5.2.5-200.fc30
thanks!
I think i might have been on a version of a rpm-fusion repo where the updated package was not showing up (maybe for fedora 29 or something) so removing the rpm-fusion repo and then reinstalling it, made the package available but it’s been a while now so i am not quite sure. Now I don’t have the nvidia drivers installed anymore because i was having some stability issues and did not notice any improvement over the nouveau drivers in my daily use.
I had exactly same issue and was pulling my hair down yesterday to fix it without any luck.
kernel 5.1.20 → 5.2.5
I tried removing RPM Fusion from sources, dnf clean all, dnf update, then added RPM Fusion repositories back again and tried dnf update and to my surprise there was a nVidia driver update (430.40) which fixed the problem.
I hope it helps.
(Thanks to mhack for his updates and details)
1 Like
I attempted an update again and it was successful without any issues. I didn’t actually change any of my repos. ¯\_(ツ)_/¯
Even better! 
Generally, as a ROT, the proprietary driver will not update as fast as the open sourced versions due to their lack of direct involvement in Fedora Linux. This bites both ways of course since the proprietary driver will logically exploit new hardware features before the community that supports the open source effort is aware the feature exists. It means in practice that a bit of patience pays in these instances.