"Missing nvidia kernel. Falling back on nouveau." message after updating to fedora version 44

Ask Fedora
, , ,
1

Hey, I’ve been getting this message after updating to version 44. My computer worked just fine before. I’m new to linux (switched once windows 10 support ended) and not an expert with computers. I’ve been trying many different fixes but honestly at this point I’m afraid I might have made things worse. So far there’s no evidence of that though.

I’ve tried uninstalling and reinstalling my nvidia drivers. I just recently followed this posts instructions by jeff and after rebooting my computer three times I now get the message “akmod: building /usr/src/akmods/nvidia-580xx-kmod.latest…”

Before this I’ve also tried updating my kernel, during which I updated from 6.19 which I understood to be the default kernel version 44 came with to 7.1.0. If this is a problem I’d like introductions on how to switch back. Trying to run nvdia-smi I get the following answer "error: module nvidia not found.

Before this I’ve also tried all of this posts instructions to no avail

output from typing in inxi -Fzxx
[1]

  1. System:
    Kernel: 7.1.0-0.rc4.328.vanilla.fc44.x86_64 arch: x86_64 bits: 64
    compiler: gcc v: 16.1.1
    Desktop: KDE Plasma v: 6.6.5 tk: Qt v: N/A wm: kwin_wayland dm: SDDM
    Distro: Fedora Linux 44 (KDE Plasma Desktop Edition)
    Machine:
    Type: Desktop Mobo: ASRock model: H110M-HDS serial:
    Firmware: UEFI-[Legacy] vendor: American Megatrends v: P1.20
    date: 01/25/2016
    CPU:
    Info: dual core model: Intel Core i3-6100 bits: 64 type: MT MCP
    arch: Skylake-S rev: 3 cache: L1: 128 KiB L2: 512 KiB L3: 3 MiB
    Speed (MHz): avg: 3700 min/max: 800/3700 cores: 1: 3700 2: 3700 3: 3700
    4: 3700 bogomips: 29598
    Flags-basic: avx avx2 ht lm nx pae sse sse2 sse3 sse4_1 sse4_2 ssse3 vmx
    Graphics:
    Device-1: NVIDIA GP107 [GeForce GTX 1050 Ti] vendor: ZOTAC driver: nouveau
    v: kernel arch: Pascal pcie: speed: 2.5 GT/s lanes: 16 ports:
    active: DVI-D-1 empty: DP-1,HDMI-A-1 bus-ID: 01:00.0 chip-ID: 10de:1c82
    temp: 37.0 C
    Display: wayland server: X.org v: 1.21.1.22 with: Xwayland v: 24.1.11
    compositor: kwin_wayland driver: gpu: nouveau display-ID: 0
    Monitor-1: DVI-D-1 model: AAA res: 1920x1080 hz: 60 dpi: 85
    diag: 660mm (25.96")
    API: EGL v: 1.5 platforms: device: 0 drv: nouveau device: 1 drv: swrast
    gbm: drv: nouveau surfaceless: drv: nouveau wayland: drv: nouveau x11:
    drv: nouveau
    API: OpenGL v: 4.5 compat-v: 4.3 vendor: mesa v: 26.0.6 glx-v: 1.4
    direct-render: yes renderer: NV137 device-ID: 10de:1c82 display-ID: :0.0
    API: Vulkan v: 1.4.341 surfaces: N/A device: 0 type: discrete-gpu
    driver: mesa nvk device-ID: 10de:1c82 device: 1 type: cpu
    driver: mesa llvmpipe device-ID: 10005:0000
    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 100 Series/C230 Series Family HD Audio vendor: ASRock
    driver: snd_hda_intel v: kernel bus-ID: 00:1f.3 chip-ID: 8086:a170
    Device-2: NVIDIA GP107GL High Definition Audio vendor: ZOTAC
    driver: snd_hda_intel v: kernel pcie: speed: 2.5 GT/s lanes: 16
    bus-ID: 01:00.1 chip-ID: 10de:0fb9
    Device-3: Tenx USB AUDIO driver: hid-generic,snd-usb-audio,usbhid
    type: USB rev: 1.1 speed: 12 Mb/s lanes: 1 bus-ID: 1-7:3 chip-ID: 1130:1620
    Device-4: Kingston HyperX QuadCast
    driver: hid-generic,snd-usb-audio,usbhid type: USB rev: 1.1 speed: 12 Mb/s
    lanes: 1 bus-ID: 1-9:5 chip-ID: 0951:16df
    API: ALSA v: k7.1.0-0.rc4.328.vanilla.fc44.x86_64 status: kernel-api
    Server-1: PipeWire v: 1.6.5 status: active with: 1: pipewire-pulse
    status: active 2: wireplumber status: active 3: pipewire-alsa type: plugin
    4: pw-jack type: plugin
    Network:
    Device-1: Intel Wireless 7265 driver: iwlwifi v: kernel pcie:
    speed: 2.5 GT/s lanes: 1 bus-ID: 02:00.0 chip-ID: 8086:095a
    IF: wlp2s0 state: up mac:
    Device-2: Realtek RTL8111/8168/8211/8411 PCI Express Gigabit Ethernet
    vendor: ASRock driver: r8169 v: kernel pcie: speed: 2.5 GT/s lanes: 1
    port: d000 bus-ID: 03:00.0 chip-ID: 10ec:8168
    IF: enp3s0 state: down mac:
    Drives:
    Local Storage: total: 931.51 GiB used: 203.66 GiB (21.9%)
    ID-1: /dev/sda vendor: Toshiba model: HDWD110 size: 931.51 GiB
    speed: 6.0 Gb/s serial:
    Partition:
    ID-1: / size: 930.51 GiB used: 202.9 GiB (21.8%) fs: btrfs dev: /dev/sda3
    ID-2: /boot size: 973.4 MiB used: 777.2 MiB (79.8%) fs: ext4
    dev: /dev/sda2
    ID-3: /home size: 930.51 GiB used: 202.9 GiB (21.8%) fs: btrfs
    dev: /dev/sda3
    Swap:
    ID-1: swap-1 type: zram size: 7.7 GiB used: 1.83 GiB (23.8%) priority: 100
    dev: /dev/zram0
    Sensors:
    System Temperatures: cpu: 40.0 C pch: 35.5 C mobo: N/A gpu: nouveau
    temp: 37.0 C
    Fan Speeds (rpm): N/A
    Info:
    Memory: total: 8 GiB available: 7.7 GiB used: 3.43 GiB (44.5%)
    Processes: 300 Power: uptime: 31m wakeups: 0 Init: systemd v: 259
    default: graphical
    Packages: pm: rpm pkgs: N/A note: see --rpm pm: flatpak pkgs: 46
    Compilers: gcc: 16.1.1 Shell: Bash v: 5.3.9 running-in: konsole inxi: 3.3.40 ↩︎

2

No harm done. Your GTX 1050 needs the 580xx legacy driver branch and the nvidia-580xx-kmod message means the correct driver is already in place.

The problem is your kernel. Seems you replaced the stock kernel with a 7.1.0-rc4 vanilla one? The Nvidia module can only compile against a kernel that has matching development headers, so the build can’t finish and you fall back to nouveau.

Reboot and at bootmenu choose Advanced options for Fedora, then pick an older 6.x kernel. That’s should be safe and bring your Nvidia driver back. Don’t reinstall drivers, that’s not the issue.

If you can, paste the output of these two read-only commands first:

rpm -qa | grep -E ‘kernel|nvidia|akmod’ | sort
uname -r

3

Rebooted and went to my boot menu, the only options were seemingly kernel 7.x. no option with a 6.x number on it.

Is it possible the older kernel got deleted at some point during all the troubleshooting and refeshing and such? If so how do I check and possibly reinstall the stock kernel?

Also result from the earlier command

$ rpm -qa | grep -E ‘kernel|nvidia|akmod’ | sort
uname -r
bash: akmod’: command not found..
bash: nvidia: command not found…
7.1.0-0.rc4.328.vanilla.fc44.x86_64

4

Boot into any 7.x option, open a terminal, and run these one at a time:

Which vanilla repo is enabled:
dnf copr list

Confirm the stock kernel is gone:
rpm -qa ‘kernel*’ | sort

Tells exactly which repo to disable and confirm the stock kernel is missing, makes reinstall the stock kernel easier without guessing :wink:

Btw:
$ rpm -qa | grep -E ‘kernel|nvidia|akmod’ | sort
uname -r
bash: akmod’: command not found..
bash: nvidia: command not found…
7.1.0-0.rc4.328.vanilla.fc44.x86_64

You have to run it with straight quotes.

5

dnf copr list
Making sure you're not a bot! [eternal_deps]
Making sure you're not a bot! (disabled)

rpm -qa 'kernel*' | sort
kernel-7.1.0-0.rc3.260514.e1914add.324.vanilla.fc44.x86_64
kernel-7.1.0-0.rc3.260516.6916d570.326.vanilla.fc44.x86_64
kernel-7.1.0-0.rc4.328.vanilla.fc44.x86_64
kernel-core-7.1.0-0.rc3.260514.e1914add.324.vanilla.fc44.x86_64
kernel-core-7.1.0-0.rc3.260516.6916d570.326.vanilla.fc44.x86_64
kernel-core-7.1.0-0.rc4.328.vanilla.fc44.x86_64
kernel-devel-7.1.0-0.rc3.260514.e1914add.324.vanilla.fc44.x86_64
kernel-devel-7.1.0-0.rc3.260516.6916d570.326.vanilla.fc44.x86_64
kernel-devel-7.1.0-0.rc4.328.vanilla.fc44.x86_64
kernel-devel-matched-7.1.0-0.rc4.328.vanilla.fc44.x86_64
kernel-headers-7.1.0-0.rc4.328.vanilla.fc44.x86_64
kernel-modules-7.1.0-0.rc3.260514.e1914add.324.vanilla.fc44.x86_64
kernel-modules-7.1.0-0.rc3.260516.6916d570.326.vanilla.fc44.x86_64
kernel-modules-7.1.0-0.rc4.328.vanilla.fc44.x86_64
kernel-modules-core-7.1.0-0.rc3.260514.e1914add.324.vanilla.fc44.x86_64
kernel-modules-core-7.1.0-0.rc3.260516.6916d570.326.vanilla.fc44.x86_64
kernel-modules-core-7.1.0-0.rc4.328.vanilla.fc44.x86_64
kernel-modules-extra-7.1.0-0.rc3.260514.e1914add.324.vanilla.fc44.x86_64
kernel-modules-extra-7.1.0-0.rc3.260516.6916d570.326.vanilla.fc44.x86_64
kernel-modules-extra-7.1.0-0.rc4.328.vanilla.fc44.x86_64
kernel-srpm-macros-1.0-28.fc44.noarch
kernel-tools-7.1.0-0.rc4.328.vanilla.fc44.x86_64
kernel-tools-libs-7.1.0-0.rc4.328.vanilla.fc44.x86_64

So only 7.x kernel installed apparently. I don’t remember intentionally deleting the older kernels but they probably got deleted during an update or a refresh.

6

Fedora defaults to an “installonly_limit” of “3”, so you have 3 different 7.1.0-0 kernels there:

7.1.0-0.rc3.260514.e1914add.324
7.1.0-0.rc3.260516.6916d570.326
7.1.0-0.rc4.328

so that 6.19* one was pushed out to keep the number of kernels “3”. You can either increase the “3” in a dnf config file and do it that way or just be more proactive at removing kernels before you get to 3.

7

Also

$ rpm -qa | grep -E ‘kernel|nvidia|akmod’ | sort
uname -r

[1]

  1. abrt-addon-kerneloops-2.17.8-3.fc44.x86_64
    akmod-nvidia-580xx-580.159.03-1.fc44.x86_64
    akmods-0.6.2-9.fc44.noarch
    kernel-7.1.0-0.rc3.260514.e1914add.324.vanilla.fc44.x86_64
    kernel-7.1.0-0.rc3.260516.6916d570.326.vanilla.fc44.x86_64
    kernel-7.1.0-0.rc4.328.vanilla.fc44.x86_64
    kernel-core-7.1.0-0.rc3.260514.e1914add.324.vanilla.fc44.x86_64
    kernel-core-7.1.0-0.rc3.260516.6916d570.326.vanilla.fc44.x86_64
    kernel-core-7.1.0-0.rc4.328.vanilla.fc44.x86_64
    kernel-devel-7.1.0-0.rc3.260514.e1914add.324.vanilla.fc44.x86_64
    kernel-devel-7.1.0-0.rc3.260516.6916d570.326.vanilla.fc44.x86_64
    kernel-devel-7.1.0-0.rc4.328.vanilla.fc44.x86_64
    kernel-devel-matched-7.1.0-0.rc4.328.vanilla.fc44.x86_64
    kernel-headers-7.1.0-0.rc4.328.vanilla.fc44.x86_64
    kernel-modules-7.1.0-0.rc3.260514.e1914add.324.vanilla.fc44.x86_64
    kernel-modules-7.1.0-0.rc3.260516.6916d570.326.vanilla.fc44.x86_64
    kernel-modules-7.1.0-0.rc4.328.vanilla.fc44.x86_64
    kernel-modules-core-7.1.0-0.rc3.260514.e1914add.324.vanilla.fc44.x86_64
    kernel-modules-core-7.1.0-0.rc3.260516.6916d570.326.vanilla.fc44.x86_64
    kernel-modules-core-7.1.0-0.rc4.328.vanilla.fc44.x86_64
    kernel-modules-extra-7.1.0-0.rc3.260514.e1914add.324.vanilla.fc44.x86_64
    kernel-modules-extra-7.1.0-0.rc3.260516.6916d570.326.vanilla.fc44.x86_64
    kernel-modules-extra-7.1.0-0.rc4.328.vanilla.fc44.x86_64
    kernel-srpm-macros-1.0-28.fc44.noarch
    kernel-tools-7.1.0-0.rc4.328.vanilla.fc44.x86_64
    kernel-tools-libs-7.1.0-0.rc4.328.vanilla.fc44.x86_64
    libreport-plugin-kerneloops-2.17.15-10.fc44.x86_64
    nvidia-gpu-firmware-20260410-1.fc44.noarch
    nvidia-modprobe-595.71.05-1.fc44.x86_64
    nvidia-persistenced-595.71.05-1.fc44.x86_64
    nvidia-settings-580xx-580.159.03-1.fc44.x86_64
    xorg-x11-drv-nvidia-580xx-580.159.03-1.fc44.x86_64
    xorg-x11-drv-nvidia-580xx-cuda-580.159.03-1.fc44.x86_64
    xorg-x11-drv-nvidia-580xx-cuda-libs-580.159.03-1.fc44.i686
    xorg-x11-drv-nvidia-580xx-cuda-libs-580.159.03-1.fc44.x86_64
    xorg-x11-drv-nvidia-580xx-kmodsrc-580.159.03-1.fc44.x86_64
    xorg-x11-drv-nvidia-580xx-libs-580.159.03-1.fc44.i686
    xorg-x11-drv-nvidia-580xx-libs-580.159.03-1.fc44.x86_64
    xorg-x11-drv-nvidia-580xx-power-580.159.03-1.fc44.x86_64
    xorg-x11-drv-nvidia-580xx-xorg-libs-580.159.03-1.fc44.x86_64
    7.1.0-0.rc4.328.vanilla.fc44.x86_64
    ↩︎

8

Disable the vanilla COPR:
sudo dnf copr remove @kernel-vanilla/mainline-wo-mergew

Reinstall the stock Fedora 44 kernel/matching headers:
sudo dnf --refresh install kernel kernel-core kernel-modules kernel-modules-core kernel-modules-extra kernel-devel

Output should be something like 6.x.x-xxx.fc44. Stop if you see vanilla or 7.10..

Rebuild the Nvidia module against the new stock kernel:
sudo akmods --force --kernels $(rpm -qa ‘kernel-core*’ | grep -v vanilla | sed ‘s/kernel-core-//’)
sudo dracut --force

Check the ’ in the command (straight quotes!)

If the first line gives trouble because of the substitution, a plain sudo akmods --force after the reboot also works, however, try this first.

Reboot and check with: nvidia-smi

See how this goes.

9

Fedora 44 has kernel 7.0.8-200 in the “updates” repo, you won’t see 6.19* ones anymore they’re gone due to EOL.

10

yeah vanilla kernel got deleted succesfully but the attempt to reinstall the stock kernel only came out with “kernel 7.1.0-0 is already installed” etc.

11

Absolutely true, my bad…

12

dnf will reject installing a kernel that has a lower version number than one that’s already installed. The “updates” repo only has 7.0.8-200 in it but you have those 7.1 kernels already installed. You have to explicitly put the version numbers to force it:

I think this might work (using curly globbing to reduce tediously repeating the version string):

sudo dnf --refresh install {kernel,kernel-core,kernel-modules,kernel-modules-core,kernel-modules-extra,kernel-devel}\*7.0.8-200\*

13

Run this:

sudo dnf --refresh install $(dnf repoquery --quiet --available --latest-limit=1 kernel kernel-core kernel-modules kernel-modules-core kernel-modules-extra kernel-devel)

(should be 7.0.8-200.fc44 packages, confirm with yes).

Check if it landed:
rpm -qa ‘kernel-core*’ | sort

Reboot, pick the kernel with no vanilla in it and run nvidia-smi

14

Yess that fixed it! Thanks both of you!