HELP! Nvidia kernel module missing falling back to nouveau (Fedora 42 KDE Plasma)

Ask Fedora
, , , ,
1

Hello there, I’m a newbie at linux in general and trying to migrate from windows to linux.
I have been banging my head against the wall over the last couple of days trying to install the Nvidia drivers but when I did they just won’t show up on the Nvidia X Settings and I keep getting this error on the start-up screen (Nvidia kernel module missing falling back to nouveau).
Can anyone please help me and explain what I have to do because all of the resources I have found is just command lines with minimal explanations which confuses me even more?

2

Did you install the drivers according to the instructions at RPMFusion?
This is the recommended way for Fedora.

I don’t know if (I doubt that) NvidiaX works properly on Fedora, or if you need it.

3

Hello. I used an Youtube tutorial (This one to be exact: https://www.youtube.com/watch?v=k5uxX2U3tYE) instead of the RPMFusion instructions , I’ll try to follow the instructions and report back if I have any issues. Thanks!

1 Like
4

It does work, although the range of settings offered is narrower on Wayland than on X11.

5

The issue may also be secure boot.
If secure boot is enabled then the instructions at the rpmfusion how-to for secure boot or in the file /usr/share/doc/akmods/README.secureboot must be followed to ensure the nvidia driver module is signed and the signing key has been enrolled in bios to enable loading the module at boot time.

6

Hello again, I have manage to successfully (I hope) follow the instructions up to the point of “Installing the drivers” part of it (I have an NVIDIA Corporation GF106 [GeForce GTS 450]) . It’s giving this message on the console:
sudo dnf install xorg-x11-drv-nvidia-390xx akmod-nvidia-390xx
Updating and loading repositories:
Repositories loaded.
Failed to resolve the transaction:
Problem 1: installed package kmod-nvidia-6.17.4-200.fc42.x86_64-3:580.95.05-1.fc42.x86_64 requires nvidia-kmod-common >= 3:580.95.05, but none of the providers can be installed

  • installed package xorg-x11-drv-nvidia-3:580.95.05-1.fc42.x86_64 conflicts with xorg-x11-drv-nvidia-390xx provided by xorg-x11-drv-nvidia-390xx-3:390.157-4.fc42.x86_64 from rpmfusion-nonfree
  • package xorg-x11-drv-nvidia-3:580.95.05-1.fc42.x86_64 from rpmfusion-nonfree-updates conflicts with xorg-x11-drv-nvidia-390xx provided by xorg-x11-drv-nvidia-390xx-3:390.157-4.fc42.x86_64 from rpmfusion-nonfree
  • conflicting requests
  • installed package xorg-x11-drv-nvidia-3:580.95.05-1.fc42.x86_64 conflicts with xorg-x11-drv-nvidia-390xx provided by xorg-x11-drv-nvidia-390xx-3:390.157-5.fc42.x86_64 from rpmfusion-nonfree-updates
  • package xorg-x11-drv-nvidia-390xx-3:390.157-5.fc42.x86_64 from rpmfusion-nonfree-updates conflicts with xorg-x11-drv-nvidia provided by xorg-x11-drv-nvidia-3:580.95.05-1.fc42.x86_64 from rpmfusion-nonfree-updates
  • problem with installed package
    Problem 2: conflicting requests
  • package akmod-nvidia-390xx-3:390.157-18.fc42.x86_64 from rpmfusion-nonfree requires nvidia-390xx-kmod-common >= 3:390.157, but none of the providers can be installed
  • package akmod-nvidia-390xx-3:390.157-22.fc42.x86_64 from rpmfusion-nonfree-updates requires nvidia-390xx-kmod-common >= 3:390.157, but none of the providers can be installed
  • problem with installed package
  • installed package xorg-x11-drv-nvidia-3:580.95.05-1.fc42.x86_64 conflicts with xorg-x11-drv-nvidia-390xx provided by xorg-x11-drv-nvidia-390xx-3:390.157-4.fc42.x86_64 from rpmfusion-nonfree
  • package xorg-x11-drv-nvidia-3:580.95.05-1.fc42.x86_64 from rpmfusion-nonfree-updates conflicts with xorg-x11-drv-nvidia-390xx provided by xorg-x11-drv-nvidia-390xx-3:390.157-4.fc42.x86_64 from rpmfusion-nonfree
  • package xorg-x11-drv-nvidia-390xx-3:390.157-4.fc42.x86_64 from rpmfusion-nonfree conflicts with xorg-x11-drv-nvidia provided by xorg-x11-drv-nvidia-3:570.133.07-1.fc42.x86_64 from rpmfusion-nonfree
  • installed package xorg-x11-drv-nvidia-3:580.95.05-1.fc42.x86_64 conflicts with xorg-x11-drv-nvidia-390xx provided by xorg-x11-drv-nvidia-390xx-3:390.157-5.fc42.x86_64 from rpmfusion-nonfree-updates
  • package xorg-x11-drv-nvidia-390xx-3:390.157-5.fc42.x86_64 from rpmfusion-nonfree-updates conflicts with xorg-x11-drv-nvidia provided by xorg-x11-drv-nvidia-3:580.95.05-1.fc42.x86_64 from rpmfusion-nonfree-updates
  • package xorg-x11-drv-nvidia-390xx-3:390.157-5.fc42.x86_64 from rpmfusion-nonfree-updates conflicts with xorg-x11-drv-nvidia provided by xorg-x11-drv-nvidia-3:570.133.07-1.fc42.x86_64 from rpmfusion-nonfree
    You can try to add to command line:
    –allowerasing to allow removing of installed packages to resolve problems
    –skip-broken to skip uninstallable packages

Should I use the allow erasing command line, skip broken or is there any other solutions that I don’t know about/ did something wrong?

7

Edit: see Jeff’s post below.

8

(But now we know that OP has a GTS450, so is restricted to the 390xx drivers which support only X11, not Wayland).

9

We also know from the last post from the OP that he has already installed the latest nvidia drivers and that caused the conflict with installing the 390xx driver. This can easily be fixed with

  1. sudo dnf remove \*nvidia\* --exclude nvidia-gpu-firmware to remove everything currently installed
  2. sudo dnf install akmod-nvidia-390xx which should pull in everything necessary except cuda. If cuda is desired then add the package xorg-x11-drv-nvidia-390xx-cuda to the dnf command.

The real issue now is that support for x11 is rapidly fading and that driver only works with x11.

My recommendation would be to not install the nvidia driver at all, but instead allow the nouveau driver to be used. That will support wayland and it appears the only thing that would be missing would be hardware acceleration for graphics intensive apps.

@forking95
If you choose to do without the nvidia drivers so you can use wayland, then please also post here the output of cat /proc/cmdline and cat /etc/kernel/cmdline after doing the removal step 1 above. If it leaves the nouveau driver blacklisted in the kernel command line we may need to also fix that for you before you reboot.

2 Likes
10

I did some additional testing on my f42 system and found one command that will work for both removing the nvidia packages and installing new ones to avoid the conflicts when the 580 driver is already installed.
sudo dnf install akmod-nvidia-390xx xorg-x11-drv-nvidia-390xx-cuda --allowerasing --best

This does require that both the rpmfusion-nonfree and rpmfusion-nonfree-updates repo be enabled since the 390xx packages are not in the rpmfusion-nonfree-nvidia-driver repo.

1 Like
11

Hello! Sorry for taking this long to respond to your reply, I have been quite busy today. I ran the command lines and it’s solved the issue with the start-up screen errors… But unfortunately after the start-up screen it’s a just a black screen (The monitor doesn’t turn itself off, it’s still on but it’s just a black screen) + another thing to note is that the NVIDIA X Settings refuses to open when I click it. The only option that doesn’t have this issue from my Fedora boot menu (Is that the right term?) is the rescue one (Which I am assuming is equivalent to windows Safe Mode), is there any way I can check if I screwed anything up/ how to diagnose + fix it?

12

After booting and the black screen can you use ctrl>alt>f3 and open a text login window.?

If you can then please run `lsmod | grep -E ‘nvidia|nouveau’ and let us know what is shown there. It should be a group of nvidia modules or a group of nouveau modules.

I still stand by my recommendation to avoid the nvidia drivers (version 390 requires use of x11 only – (x11 is no longer supported on f43)) and to use the nouveau drivers which will work with wayland.

13

Linux support is mostly a community effort. Text is more efficient for back & forth troubleshooting, and is web searchable, which means others who have the same problem can find each other. A good way to gain fluency with command-line tools is https://linuxcommand.org, created by a real human. The internet has become overrun with AI clickbait sites offering “simple” solutions to common problems, so it is important to understand commands found on the web before running them.

When asking for help with an issue that doesn’t get a quick solution or a lot others with the same issue, you should a) provide enough detail (e.g., the output from running inxi -Fzxx in a terminal and posting as pre-formatted text) to allow others with access to similar hardware to reproduce the issue and b) ensure that Fedora and vendor firmware are fully updated so you aren’t chasing a solved problem.

14

Thank you for the advice, sir! I’m having quite a rocky road with preparing Fedora for my migration from Windows 10. I’m not quite used to having my issues not be solved in 5-10 minute Youtube / Firefox search and having to search forums for troubleshooting and solutions. Here’s my “inxi - Fzxx” if it will help:
System:
Kernel: 6.14.0-63.fc42.x86_64 arch: x86_64 bits: 64 compiler: gcc v: 15.0.1
Desktop: KDE Plasma v: 6.4.5 Distro: Fedora Linux 42 (KDE Plasma Desktop
Edition)
Machine:
Type: Desktop Mobo: ASRock model: H61M-VG3 serial:
UEFI: American Megatrends v: P1.90 date: 07/11/2013
CPU:
Info: quad core model: Intel Core i5-3350P bits: 64 type: MCP
arch: Ivy Bridge rev: 9 cache: L1: 256 KiB L2: 1024 KiB L3: 6 MiB
Speed (MHz): avg: 1597 min/max: 1600/3100 cores: 1: 1597 2: 1597 3: 1597
4: 1597 bogomips: 24744
Flags-basic: avx ht lm nx pae sse sse2 sse3 sse4_1 sse4_2 ssse3 vmx
Graphics:
Device-1: NVIDIA GF106 [GeForce GTS 450] vendor: LeadTek Research
driver: N/A arch: Fermi bus-ID: 01:00.0
Display: wayland server: X.org v: 1.21.1.18 with: Xwayland v: 24.1.8
compositor: kwin_wayland driver: N/A resolution: 1024x768~60Hz
API: EGL v: 1.5 drivers: swrast platforms:
active: wayland,x11,surfaceless,device inactive: gbm
API: OpenGL v: 4.5 vendor: mesa v: 25.1.9 glx-v: 1.4 direct-render: yes
renderer: llvmpipe (LLVM 20.1.8 256 bits)
API: Vulkan v: 1.4.313 drivers: llvmpipe surfaces: N/A devices: 1
Info: Tools: api: clinfo, eglinfo, glxinfo, vulkaninfo
de: kscreen-console,kscreen-doctor gpu: nvidia-settings,nvidia-smi
wl: wayland-info x11: xdriinfo, xdpyinfo, xprop, xrandr
Audio:
Device-1: Intel 6 Series/C200 Series Family High Definition Audio
vendor: ASRock driver: snd_hda_intel v: kernel bus-ID: 00:1b.0
Device-2: NVIDIA GF106 High Definition Audio vendor: LeadTek Research
driver: snd_hda_intel v: kernel bus-ID: 01:00.1
API: ALSA v: k6.14.0-63.fc42.x86_64 status: kernel-api
Server-1: PipeWire v: 1.4.9 status: active
Network:
Device-1: Realtek RTL8111/8168/8211/8411 PCI Express Gigabit Ethernet
vendor: ASRock driver: r8169 v: kernel port: d000 bus-ID: 03:00.0
IF: enp3s0 state: down mac:
Device-2: TP-Link TL-WN822N Version 4 RTL8192EU driver: rtl8xxxu type: USB
bus-ID: 1-1.4:5
IF: wlp0s29u1u4 state: up mac:
Drives:
Local Storage: total: 1.09 TiB used: 8.34 GiB (0.7%)
ID-1: /dev/sda vendor: Kingston model: SA400S37960G size: 894.25 GiB
ID-2: /dev/sdb vendor: Kingston model: SA400S37240G size: 223.57 GiB
Partition:
ID-1: / size: 892.67 GiB used: 7.71 GiB (0.9%) fs: btrfs dev: /dev/sda3
ID-2: /boot size: 973.4 MiB used: 618.4 MiB (63.5%) fs: ext4
dev: /dev/sda2
ID-3: /boot/efi size: 598.8 MiB used: 19.3 MiB (3.2%) fs: vfat
dev: /dev/sda1
ID-4: /home size: 892.67 GiB used: 7.71 GiB (0.9%) fs: btrfs
dev: /dev/sda3
Swap:
ID-1: swap-1 type: zram size: 7.7 GiB used: 11.7 MiB (0.1%) dev: /dev/zram0
Sensors:
System Temperatures: cpu: 33.0 C mobo: N/A
Fan Speeds (rpm): N/A
Info:
Memory: total: 8 GiB available: 7.7 GiB used: 2.68 GiB (34.8%)
Processes: 266 Uptime: 14m Init: systemd target: graphical (5)
Packages: N/A note: see --rpm Compilers: gcc: 15.2.1 Shell: Bash v: 5.2.37
inxi: 3.3.39

15

After booting and the black screen can you use ctrl>alt>f3 and open a text login window.?

-No, when I press ctrl>alt>f3 nothing appears on the black screen. Should I reinstall Fedora (Don’t worry, I don’t have any personal files yet since I’m preparing it for my migration from Windows 10) and try the nouveau drivers instead?

16

I would reinstall and not install the nvidia drivers.
As soon as you have completed the install you should update with sudo dnf upgrade done in a terminal window then reboot again. This will switch the kernel from 6.14.0 to the current 6.17.5 along with many other package updates.

Finally check which driver is loaded for the nvidia gpu with lsmod | grep -E "nvidia|nouveau".

That should show nouveau modules, and if so then try and see if your normal work flow is good.

Let us know the results.

Local Storage: total: 1.09 TiB used: 8.34 GiB (0.7%)
ID-1: /dev/sda vendor: Kingston model: SA400S37960G size: 894.25 GiB
ID-2: /dev/sdb vendor: Kingston model: SA400S37240G size: 223.57 GiB
Partition:
ID-1: / size: 892.67 GiB used: 7.71 GiB (0.9%) fs: btrfs dev: /dev/sda3
ID-2: /boot size: 973.4 MiB used: 618.4 MiB (63.5%) fs: ext4
dev: /dev/sda2
ID-3: /boot/efi size: 598.8 MiB used: 19.3 MiB (3.2%) fs: vfat
dev: /dev/sda1
ID-4: /home size: 892.67 GiB used: 7.71 GiB (0.9%) fs: btrfs
dev: /dev/sda3

BTW, I note that you only are using about 8GB of drive space for the entire system. I think that you probably should have at least 20GB for the install or you will quickly run out of space. The larger the drive space available the better overall.