After upgrading from F43 to F44 (through Discover) I got the “stuck on Plymouth” problem. Resolved by booting with nomodeset instead of all the nvidia kernel module settings and rebuilding the driver with sudo akmods --rebuild --force. Driver version is 595.71.05 from RPMFusion and the GPU is a laptop RTX A1000.
Happened again after a kernel upgrade
In theory, akmods should have automatically built the new kmod on boot. It would be good to check that specific boot’s log to see why it failed.
Find the list of boot logs with journalctl --list-boots and then review the relevant log by providing its number, e.g. journalctl --boot -1
If you share that here, someone may be able to identify the issue.
When akmods builds the modules there should be a log in /var/cache/akmods/nvidia with the driver version and kernel version in the name.
When it fails there should be a similar log with ‘failed’ in the name.
That log of the failure should tell us what actually failed.
If it succeeds the rpm used to install the driver should also be in the same location.
Not seeing anything suspicious during the update, but in the reboot after updates are installed there’s this:
(base) nnedelko@fedora:~$ journalctl --boot -8 | grep nvidia
May 10 01:24:24 fedora kernel: Command line: BOOT_IMAGE=(hd1,gpt2)/vmlinuz-7.0.4-200.fc44.x86_64 root=UUID=7b941001-f105-4f76-90ff-96c53fa32025 ro rootflags=subvol=root rhgb quiet rd.driver.blacklist=nouveau,nova_core modprobe.blacklist=nouveau,nova_core nvidia-drm.modeset=1 nvidia-drm.fbdev=1
May 10 01:24:24 fedora kernel: Kernel command line: BOOT_IMAGE=(hd1,gpt2)/vmlinuz-7.0.4-200.fc44.x86_64 root=UUID=7b941001-f105-4f76-90ff-96c53fa32025 ro rootflags=subvol=root rhgb quiet rd.driver.blacklist=nouveau,nova_core modprobe.blacklist=nouveau,nova_core nvidia-drm.modeset=1 nvidia-drm.fbdev=1
May 10 01:24:24 fedora dracut-cmdline[412]: Using kernel command line parameters: rd.driver.pre=btrfs BOOT_IMAGE=(hd1,gpt2)/vmlinuz-7.0.4-200.fc44.x86_64 root=UUID=7b941001-f105-4f76-90ff-96c53fa32025 ro rootflags=subvol=root rhgb quiet rd.driver.blacklist=nouveau,nova_core modprobe.blacklist=nouveau,nova_core nvidia-drm.modeset=1 nvidia-drm.fbdev=1
May 10 01:24:30 fedora systemd[1]: Starting nvidia-persistenced.service - NVIDIA Persistence Daemon...
May 10 01:24:30 fedora systemd[1]: Started nvidia-powerd.service - nvidia-powerd service.
May 10 01:24:30 fedora audit[1]: SERVICE_START pid=1 uid=0 auid=4294967295 ses=4294967295 subj=kernel msg='unit=nvidia-powerd comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
May 10 01:24:30 fedora nvidia-persistenced[1447]: Started (1447)
May 10 01:24:30 fedora nvidia-powerd[1438]: nvidia-powerd version:2.0 (build 1)
May 10 01:24:31 fedora nvidia-persistenced[1447]: Failed to query NVIDIA devices. Please ensure that the NVIDIA device files (/dev/nvidia*) exist, and that user 0 has read and write permissions for those files.
May 10 01:24:31 fedora nvidia-powerd[1438]: ERROR! Allocate Root client failed 0x59
May 10 01:24:31 fedora nvidia-persistenced[1436]: nvidia-persistenced failed to initialize. Check syslog for more details.
May 10 01:24:31 fedora nvidia-persistenced[1447]: Shutdown (1447)
May 10 01:24:31 fedora systemd[1]: nvidia-persistenced.service: Control process exited, code=exited, status=1/FAILURE
May 10 01:24:31 fedora systemd[1]: nvidia-powerd.service: Deactivated successfully.
May 10 01:24:31 fedora audit[1]: SERVICE_STOP pid=1 uid=0 auid=4294967295 ses=4294967295 subj=kernel msg='unit=nvidia-powerd comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
May 10 01:24:31 fedora systemd[1]: nvidia-persistenced.service: Failed with result 'exit-code'.
May 10 01:24:31 fedora systemd[1]: Failed to start nvidia-persistenced.service - NVIDIA Persistence Daemon.
May 10 01:24:31 fedora audit[1]: SERVICE_START pid=1 uid=0 auid=4294967295 ses=4294967295 subj=kernel msg='unit=nvidia-persistenced comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=failed'
Hmm, upon a second look at the upgrade process there is something:
I can see the updater uninstalling the old kmod-nvidia
May 10 01:22:50 fedora pk-offline-update[1326]: package removing kmod-nvidia-6.19.13-200.fc43.x86_64-3:580.142-2.fc43.x86_64 (installed)
but I don’t see it installing the new one. No errors or anything, there just isn’t a line with package updating kmod-nvidia.
Found a failure log for the build during the May 2nd upgrade from 43 to 44 with
2026/05/01 20:44:44 akmodsbuild: Failed to process /tmp/akmodsbuild.fql2CHYw/BUILD/nvidia-kmod-595.71.05-build/BUILDROOT/usr: IO error for operation on /tmp/akmodsbuild.fql2CHYw/BUILD/nvidia-kmod-595.71.05-build/BUILDROOT/usr: No such file or directory (os error 2) but no failure log for the recent kernel update. In fact, for May 10th there is only one build log (which is from me manually rebuilding), so it looks like that time the driver really was just silently not rebuilt for some reason.
Hmm, the same Failed to process line is also present in the successful manual rebuild. Other than that the failed build just says
Package Arch Version Repository Size
Removing:
kernel x86_64 0:6.19.12-200.fc43 updates 0.0 B
kernel-core x86_64 0:6.19.12-200.fc43 updates 98.0 MiB
kernel-devel x86_64 0:6.19.12-200.fc43 updates 87.5 MiB
kernel-modules x86_64 0:6.19.12-200.fc43 updates 96.5 MiB
kernel-modules-core x86_64 0:6.19.12-200.fc43 updates 70.7 MiB
kernel-modules-extra x86_64 0:6.19.12-200.fc43 updates 4.2 MiB
Removing dependent packages:
kmod-nvidia-6.19.12-200.fc43.x86_64 x86_64 3:580.142-2.fc43 @commandline 10.4 MiB
Installing:
kmod-nvidia-6.19.14-300.fc44.x86_64 x86_64 3:595.71.05-1.fc44 @commandline 10.5 MiB
Transaction Summary:
Installing: 1 package
Removing: 7 packages
Total size of inbound packages is 9 MiB. Need to download 0 B.
After this operation, 357 MiB will be freed (install 10 MiB, remove 367 MiB).
Running transaction
Transaction failed: Rpm transaction failed.
Can you stop the worthless spam and post something useful, see bugreport.
- I posted what I was asked to post
- The driver works just fine after it was rebuilt and the bug report script didn’t find any issues
- I’m not even sure that this is an issue with the driver component itself and not with the update configuration or something