I recently installed Fedora 43 to get away from Windows. It has been great so far, getting Steam, Battle.net, and even 3rd party tools to enable my Corsair Keyboards Macros.
The only issue I’m having is newer Kernel versions created after system updates, KDE appears to fail when logging in.
However, on the newer Kernels, 6.18.8 and 6.18.9, the system will load to the login screen, all monitors working as expected, however the moment I login, all screens go black. My mouse still shows, and I can move it across all monitors, and if I take it to my left most monitor to the left top corner, the desktop feature activates, and I can select a new desktop or create a new one.
If I use ctrl + shift + f3, the terminal shows only on my GPU monitors as mirrored screens, my motherboard one goes black.
I followed the steps in the above guide again on 6.18.9 after login and using ctrl + shift + f3, however after reboot and login, the same issue happens again.
I’m not sure if this is a NVIDIA related since all the monitors light up at login screen, or KDE related with the newer kernels.
Any recommendations would be much appreciated!
P.S: I have Googled how to save 6.18.5 and I am just using this for now to keep Fedora working.
Post the URL it gives you here. The output will probably be more than 32K which is the maximum size of a post on this site, and as we don’t know the root cause, seeing the entire journal would be useful.
If the output shows us what the issue is, I’ll trim the output to the relevant part and post it in this thread as an addendum, to ensure it is preserved.
You appear to have an amd gpu in this machine and an nvidia gpu ( the driver of which is failing to load and falling back to nouveau… which is blacklisted and possibly the cause of the back screens… maybe!)
Is this the case, and could you post the output from inxi -Fzxx, whilst also confirming which cards you expect to be using. If you say “all of them” I’m going to have an aneurysm.
Assuming you want to use the amdgpu AND nvidia (given that you have 5 monitors) then I feel the issue is probably the fact that your nvidia drivers are not loading cleanly, falling back to nouveau and that’s causing Kwin to crash and showing what the new black looks like.
Ergo, making sure that you have the nvidia drivers built, keys enrolled as you have Secure boot enabled and I’d also suspect you are wanting to use the rpmfusion drivers. Are you ok with rebuilding the nv drivers yourself or want a hand with that?
The AMD GPU you’re seeing would be the 9800X3D I assume. Apologies for all the monitors I’ve just always liked having more screens for less minimising.
When loaded into 6.18.9-200 I re-ran the steps in the guide I linked earlier. Most commands returned already loaded nothing to do etc. I used --force to regenerate a new Key, the rebooted and enrolled the MOK Key with the password. If there are more steps I need to rebuild and work with secure boot, please let me know!
I’m also quite happy to stay on 6.18.5-200 if it isn’t urgent to be on the latest kernel. All my games and everything are still running fine and I don’t have problems, but I guess knowing what to do when it’s important to be on the latest kernel would be good.
The nvidia GPU seems to be working well with 4 monitors and the nvidia driver.
If I understand correctly the amd gpu is on the motherboard?
With that arrangement I suspect that forcing one monitor to be used by the onboard GPU and having 4 on the nvidia gpu might contribute to this situation.
What about trying all monitors on the same (nvidia) gpu and not use the amd ? If you did this you could also possibly disable the amdgpu in bios and avoid the conflicts.
So the top left is plugged into the Motherboard using the 9800X3D, and the others are all in the 5080.
I’m assuming you are correct that using both is causing the issue on the newer Kernel’s, so I’m just wondering why that’s the case on 6.18.9-200 and not on 6.18.9-200.
My concern with staying on 6.18.5 is that dnf commands to pin a specific kernel version don’t do what you expect them to - even though you’ve said “pin this version and never let it go, like I did with my first love to my eternal regret”, they’ll still get blown out of the water leaving you with 3 options on the grub screen, none of which give you a working graphical desktop.
Yes - I think it’s insane too, but apparently that’s what happens.
Just take heed when applying any updates for a while - you really don’t want to update enough kernels to force your remaining kernel off the grub list, until we have this cleaned up.
Re: the nvidia rpmfusion drivers: ensure that they are built for all existing kernels (sudo akmods --rebuild --force) and then build the initrd for a version of the kernel you want to test… in your case, you can immediately use version 6.18.9.-200 as that’s the kernel version you were running when you rebuilt the drivers. sudo dracut --kver 6.18.9-200.fc43.x64_64 --force
The instructions about leaving it a few minutes beofre rebooting when faffing about with drivers and sucklike aren’t voodoo - the prompt returns and looks finished whilst the drivers are being compiled in the background. Watch htop and you’ll see behind the curtain. Only reboot when it goes quiet and the fans spin down!)
If that dracut command comes back with no errors then reboot into that kernel and see if you get some screens. If you do and it all looks good we have a fix.
I’m not entirely sure WHY the nvidia driver are not loading and it’s falling back to nouveau though - maybe I’m missing something and someone else will spot the smoking gun
I did try this, watched top to see when the usage dropped, basically right after the fans spun down after about 20 seconds.
But on reboot this killed the GPU monitors on 6.18.5, and didn’t help 6.18.9 even though the command said it was building for 6.18.9 first, then 6.18.5.
I went back to the earlier guide, and tried to get 6.18.5 working by doing the secure boot keys and booting into MOK to enter the password etc, but re-following on still left only the motherboard monitor working on 6.18.5, and the same issue on 6.18.9.
So I’ve just disabled secure boot, and booting straight into 6.18.9 has worked. I did leave secure boot going just encase I wanted to go back to play BF6 on Windows, but honestly I’m kinda over it and everything else on Steam has been running great on Linux from my tests.
If another game that doesn’t work on Linux appears down the road I might have to look back into this, but until then, secure boot being off is my answer.
hmm - I must admit, I don’t bother to dick about with secure boot and I leave it off, as I no longer can be arsed with windows. If it works, don’t break it.
Then again, this should work with secure boot enabled too, with no dicking around - if you have a safe and secure (for you) way to get a working, multi-monitor, aneurysm and envy inducing way of getting it to chooch for you… ripper.
On the other hand… if you fancy arsing about with things to see why it soils the duvet, shout up - there are several people on here who I suspect would like to add to their notebooks on how this code operates in oddball situations like this and the more, the merrier.
Yeah if anybody wants me to try something else, happy to re-enable Secure boot and run through the steps
The fact it lit straight up on the latest kernel as soon as secure boot kicked the bucket means the drivers are fine, and just seems to be a matter of why secure boot hates me.
In addition, for the record, here is an excerpt of the fpaste you did of a failed boot.
This shows a fallback to the nouveau driver.
grep -Ei 'kernel: Linux version|nvidia|fail|error' e3742659
Feb 14 21:30:29 kernel: Linux version 6.18.9-200.fc43.x86_64 (mockbuild@facc52a69fa549959e04aa067a88bdf9) (gcc (GCC) 15.2.1 20260123 (Red Hat 15.2.1-7), GNU ld version 2.45.1-4.fc43) #1 SMP PREEMPT_DYNAMIC Fri Feb 6 21:43:09 UTC 2026
Feb 14 21:30:29 kernel: pci 0000:01:00.0: VF BAR 2 [mem size 0x10000000 64bit pref]: failed to assign
Feb 14 21:30:29 kernel: pci 0000:01:00.0: VF BAR 4 [mem size 0x02000000 64bit pref]: failed to assign
Feb 14 21:30:29 kernel: integrity: Loaded X.509 cert 'My NVIDIA MOK: d976435d968462f53bf824113641c5490b1f82ba'
Feb 14 21:30:29 kernel: RAS: Correctable Errors collector initialized.
Feb 14 21:30:29 systemd-vconsole-setup[324]: /usr/bin/setfont failed with a "system error" (EX_OSERR), ignoring.
Feb 14 21:30:29 systemd-vconsole-setup[332]: setfont: ERROR kdfontop.c:212 put_font_kdfontop: Unable to load such font with such kernel version
Feb 14 21:30:29 systemd-vconsole-setup[324]: Configuration of first virtual console failed, ignoring remaining ones.
Feb 14 21:30:30 systemd-vconsole-setup[504]: setfont: ERROR kdfontop.c:212 put_font_kdfontop: Unable to load such font with such kernel version
Feb 14 21:30:30 systemd-vconsole-setup[500]: /usr/bin/setfont failed with a "system error" (EX_OSERR), ignoring.
Feb 14 21:30:30 systemd-vconsole-setup[500]: Configuration of first virtual console failed, ignoring remaining ones.
Feb 14 21:30:31 kernel: amdgpu 0000:74:00.0: amdgpu: [drm] Failed to setup vendor infoframe on connector HDMI-A-1: -22
Feb 14 10:30:36 systemd[1]: systemd-boot-clear-sysfail.service - Clear SysFail Entry If The Boot Is Successful skipped, unmet condition check ConditionPathExists=/sys/firmware/efi/efivars/LoaderEntrySysFail-4a67b082-0a4c-41cf-b6c7-440b29bb8c4f
Feb 14 10:30:36 systemd[1]: systemd-boot-clear-sysfail.service - Clear SysFail Entry If The Boot Is Successful skipped, unmet condition check ConditionPathExists=/sys/firmware/efi/efivars/LoaderEntrySysFail-4a67b082-0a4c-41cf-b6c7-440b29bb8c4f
Feb 14 10:30:36 systemd[1]: systemd-boot-clear-sysfail.service - Clear SysFail Entry If The Boot Is Successful skipped, unmet condition check ConditionPathExists=/sys/firmware/efi/efivars/LoaderEntrySysFail-4a67b082-0a4c-41cf-b6c7-440b29bb8c4f
Feb 14 10:30:37 systemd-vconsole-setup[820]: setfont: ERROR kdfontop.c:212 put_font_kdfontop: Unable to load such font with such kernel version
Feb 14 10:30:37 systemd-vconsole-setup[818]: /usr/bin/setfont failed with a "system error" (EX_OSERR), ignoring.
Feb 14 10:30:37 systemd-vconsole-setup[818]: Configuration of first virtual console failed, ignoring remaining ones.
Feb 14 10:30:37 kernel: asus_wmi: failed to register LPS0 sleep handler in asus-wmi
Feb 14 10:30:37 kernel: Bluetooth: hci0: Opcode 0x0c03 failed: -16
Feb 14 10:30:37 kernel: input: HDA NVidia HDMI/DP,pcm=3 as /devices/pci0000:00/0000:00:01.1/0000:01:00.1/sound/card0/input26
Feb 14 10:30:37 kernel: input: HDA NVidia HDMI/DP,pcm=7 as /devices/pci0000:00/0000:00:01.1/0000:01:00.1/sound/card0/input27
Feb 14 10:30:37 kernel: input: HDA NVidia HDMI/DP,pcm=8 as /devices/pci0000:00/0000:00:01.1/0000:01:00.1/sound/card0/input28
Feb 14 10:30:37 kernel: input: HDA NVidia HDMI/DP,pcm=9 as /devices/pci0000:00/0000:00:01.1/0000:01:00.1/sound/card0/input29
Feb 14 10:30:37 systemd[1]: systemd-boot-clear-sysfail.service - Clear SysFail Entry If The Boot Is Successful skipped, unmet condition check ConditionPathExists=/sys/firmware/efi/efivars/LoaderEntrySysFail-4a67b082-0a4c-41cf-b6c7-440b29bb8c4f
Feb 14 10:30:37 systemd[1]: systemd-boot-clear-sysfail.service - Clear SysFail Entry If The Boot Is Successful skipped, unmet condition check ConditionPathExists=/sys/firmware/efi/efivars/LoaderEntrySysFail-4a67b082-0a4c-41cf-b6c7-440b29bb8c4f
Feb 14 10:30:37 systemd[1]: Started nvidia-powerd.service - nvidia-powerd service.
Feb 14 10:30:37 audit[1]: SERVICE_START pid=1 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:init_t:s0 msg='unit=nvidia-powerd comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
Feb 14 10:30:37 nvidia-powerd[976]: nvidia-powerd version:2.0 (build 1)
Feb 14 10:30:37 boltd[1092]: [32154c17-0010-domain0 ] udev: failed to determine if uid is stable: unknown NHI PCI id '0x2425'
Feb 14 10:30:38 systemd[1]: Starting nvidia-fallback.service - Fallback to nouveau as nvidia did not load...
Feb 14 10:30:38 kernel: nouveau 0000:01:00.0: NVIDIA GB203 (1b3000a1)
Feb 14 10:30:38 nvidia-powerd[976]: ERROR! Allocate Root client failed 0x59
Feb 14 10:30:38 systemd[1]: nvidia-powerd.service: Deactivated successfully.
Feb 14 10:30:38 audit[1]: SERVICE_STOP pid=1 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:init_t:s0 msg='unit=nvidia-powerd comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
Feb 14 10:30:40 systemd[1]: Finished nvidia-fallback.service - Fallback to nouveau as nvidia did not load.
Feb 14 10:30:40 audit[1]: SERVICE_START pid=1 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:init_t:s0 msg='unit=nvidia-fallback comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
Feb 14 10:30:40 dbus-broker-launch[958]: Activation request for 'org.freedesktop.home1' failed: The systemd unit 'dbus-org.freedesktop.home1.service' could not be found.
Feb 14 10:30:40 audit[1632]: AUDIT1103 pid=1632 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:init_t:s0 msg='op=PAM:setcred grantors=? acct="sddm" exe="/usr/lib/systemd/systemd-executor" hostname=? addr=? terminal=? res=failed'
Feb 14 10:30:40 sddm-helper-start-wayland[1651]: "kwin_core: Failed to open /dev/dri/card2 device (No such device)\n"
Feb 14 10:30:40 sddm-helper-start-wayland[1651]: "kwin_wayland_drm: Failed to open drm device /dev/dri/card2\n"
Feb 14 10:30:40 sddm-helper-start-wayland[1651]: "kwin_core: Failed to open /dev/input/event7 device (No such device)\n"
Feb 14 10:30:41 sddm-helper-start-wayland[1651]: "kwin_wayland_drm: drmSetClientCap for Atomic Mode Setting failed. Using legacy mode on GPU \"/dev/dri/card2\"\n"
Feb 14 10:30:41 sddm-helper-start-wayland[1651]: "kwin_wayland_drm: Failed to open drm device \n"
Feb 14 10:30:41 sddm-helper-start-wayland[1651]: "kwin_wayland_drm: Failed to open drm device \n"
Feb 14 10:30:41 sddm-helper-start-wayland[1651]: "kwin_wayland_drm: Failed to open drm device \n"
Feb 14 10:30:41 sddm-helper-start-wayland[1651]: "kwin_wayland_drm: Failed to open drm device \n"
Feb 14 10:30:42 sddm-helper-start-wayland[1651]: "QSoundEffect(pulseaudio): Error decoding source file:///usr/share/maliit/keyboard2/sounds/key_tick2_quiet.wav\n"
Feb 14 10:30:42 sddm-helper-start-wayland[1651]: "QSGContext::initialize: depth buffer support missing, expect rendering errors\nQSGContext::initialize: stencil buffer support missing, expect rendering errors\n"
Feb 14 10:30:44 sddm-helper-start-wayland[1651]: "kwin_core: Failed to open /dev/input/event7 device (No such device)\n"
Feb 14 10:30:44 sddm-helper-start-wayland[1651]: "kwin_core: Failed to open /dev/input/event15 device (Device already taken)\n"
Feb 14 10:30:44 sddm-helper-start-wayland[1651]: "kwin_core: Failed to open /dev/input/event16 device (Device already taken)\n"
Feb 14 10:30:44 sddm-helper-start-wayland[1651]: "kwin_core: Failed to open /dev/input/event9 device (Device already taken)\n"
Feb 14 10:30:44 sddm-helper-start-wayland[1651]: "kwin_core: Failed to open /dev/input/event14 device (Device already taken)\n"
Feb 14 10:30:49 ckb-next-daemon[964]: [E] _usbrecv (via firmware.c:45): Too many send/recv failures. Dropping.
Feb 14 10:30:49 ckb-next-daemon[964]: [E] _usbrecv (via firmware.c:45): Too many send/recv failures. Dropping.
Feb 14 10:31:54 audit[2124]: AUDIT1103 pid=2124 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:init_t:s0 msg='op=PAM:setcred grantors=? acct="acans" exe="/usr/lib/systemd/systemd-executor" hostname=? addr=? terminal=? res=failed'
Feb 14 10:31:54 pipewire[1702]: spa.alsa: 'hdmi:1': playback open failed: Permission denied
Feb 14 10:31:54 wireplumber[1703]: s-monitors: Failed to create alsa_output.pci-0000_74_00.1.hdmi-stereo: Object activation aborted: PipeWire proxy destroyed
Feb 14 10:31:54 wireplumber[1703]: wp-event-dispatcher: <WpAsyncEventHook:0x5574988e9d80> failed: failed to enum routes: pipewire proxy destroyed before finishing
Feb 14 10:31:55 kwin_wayland[2530]: drmSetClientCap for Atomic Mode Setting failed. Using legacy mode on GPU "/dev/dri/card2"
Feb 14 10:32:20 /usr/libexec/xdg-desktop-portal[2551]: Failed to create settings proxy: Error calling StartServiceByName for org.freedesktop.impl.portal.desktop.kde: Timeout was reached
Feb 14 10:32:20 kwin_wayland[2530]: Failed to register with host portal QDBusError("org.freedesktop.DBus.Error.NoReply", "Did not receive a reply. Possible causes include: the remote application did not send a reply, the message bus security policy blocked the reply, the reply timeout expired, or the network connection was broken.")
Feb 14 10:32:25 sddm[1619]: Authentication error: SDDM::Auth::ERROR_INTERNAL "Process crashed"
Feb 14 10:32:25 sddm[1619]: Authentication error: SDDM::Auth::ERROR_INTERNAL "Process crashed"
Feb 14 10:32:25 dbus-broker-launch[958]: Activation request for 'org.freedesktop.Avahi' failed.
Feb 14 10:32:25 dbus-broker-launch[958]: Activation request for 'org.bluez' failed.
Feb 14 10:32:25 systemd[1]: nvidia-fallback.service: Deactivated successfully.
Feb 14 10:32:25 systemd[1]: Stopped nvidia-fallback.service - Fallback to nouveau as nvidia did not load.
Feb 14 10:32:25 audit[1]: SERVICE_STOP pid=1 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:init_t:s0 msg='unit=nvidia-fallback comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
Feb 14 10:32:25 dbus-broker-launch[958]: Activation request for 'org.bluez' failed.
Feb 14 10:32:25 systemd[2124]: dbus-:1.2-org.kde.KSplash@0.service: Failed with result 'exit-code'.
Feb 14 10:32:25 systemd[2124]: plasma-ksplash.service: Failed with result 'signal'.
Feb 14 10:32:25 dbus-broker-launch[2238]: Activation request for 'org.freedesktop.impl.portal.desktop.kde' failed.
Feb 14 10:32:25 systemd[2124]: sys-devices-pci0000:00-0000:00:01.1-0000:01:00.1-sound-card0-controlC0.device: Failed to enqueue SYSTEMD_USER_WANTS job, ignoring: Transaction for sound.target/start is destructive (shutdown.target has 'start' job queued, but 'stop' is included in transaction).
Feb 14 10:32:25 systemd[1]: sys-devices-pci0000:00-0000:00:01.1-0000:01:00.1-sound-card0-controlC0.device: Failed to enqueue SYSTEMD_WANTS job, ignoring: Transaction for sound.target/start is destructive (shutdown.target has 'start' job queued, but 'stop' is included in transaction).
Feb 14 10:32:25 systemd[1]: sys-devices-virtual-misc-rfkill.device: Failed to enqueue SYSTEMD_WANTS job, ignoring: Transaction for systemd-rfkill.socket/start is destructive (shutdown.target has 'start' job queued, but 'stop' is included in transaction).
Feb 14 10:32:25 systemd[1]: sys-devices-pci0000:00-0000:00:02.1-0000:04:00.0-0000:05:0c.0-0000:0a:00.0-usb1-1\x2d4-1\x2d4.2-1\x2d4.2:1.0-sound-card2-controlC2.device: Failed to enqueue SYSTEMD_WANTS job, ignoring: Transaction for sound.target/start is destructive (shutdown.target has 'start' job queued, but 'stop' is included in transaction).
Feb 14 10:32:25 systemd[1]: sys-devices-pci0000:00-0000:00:02.1-0000:04:00.0-0000:05:0c.0-0000:0a:00.0-usb1-1\x2d6-1\x2d6:1.0-sound-card3-controlC3.device: Failed to enqueue SYSTEMD_WANTS job, ignoring: Transaction for sound.target/start is destructive (systemd-poweroff.service has 'start' job queued, but 'stop' is included in transaction).
Feb 14 10:32:25 systemd[1]: sys-devices-pci0000:00-0000:00:02.1-0000:04:00.0-0000:05:0c.0-0000:0a:00.0-usb1-1\x2d4-1\x2d4.3-1\x2d4.3.1-1\x2d4.3.1.2-1\x2d4.3.1.2:1.2-sound-card4-controlC4.device: Failed to enqueue SYSTEMD_WANTS job, ignoring: Transaction for sound.target/start is destructive (systemd-poweroff.service has 'start' job queued, but 'stop' is included in transaction).
Feb 14 10:32:25 systemd[1]: sys-devices-pci0000:00-0000:00:08.1-0000:74:00.1-sound-card1-controlC1.device: Failed to enqueue SYSTEMD_WANTS job, ignoring: Transaction for sound.target/start is destructive (systemd-poweroff.service has 'start' job queued, but 'stop' is included in transaction).
Feb 14 10:32:25 systemd[2124]: sys-devices-pci0000:00-0000:00:02.1-0000:04:00.0-0000:05:0c.0-0000:0a:00.0-usb1-1\x2d4-1\x2d4.2-1\x2d4.2:1.0-sound-card2-controlC2.device: Failed to enqueue SYSTEMD_USER_WANTS job, ignoring: Transaction for sound.target/start is destructive (exit.target has 'start' job queued, but 'stop' is included in transaction).
Feb 14 10:32:25 systemd[2124]: sys-devices-pci0000:00-0000:00:02.1-0000:04:00.0-0000:05:0c.0-0000:0a:00.0-usb1-1\x2d6-1\x2d6:1.0-sound-card3-controlC3.device: Failed to enqueue SYSTEMD_USER_WANTS job, ignoring: Transaction for sound.target/start is destructive (shutdown.target has 'start' job queued, but 'stop' is included in transaction).
Feb 14 10:32:25 systemd[2124]: sys-devices-pci0000:00-0000:00:02.1-0000:04:00.0-0000:05:0c.0-0000:0a:00.0-usb1-1\x2d4-1\x2d4.3-1\x2d4.3.1-1\x2d4.3.1.2-1\x2d4.3.1.2:1.2-sound-card4-controlC4.device: Failed to enqueue SYSTEMD_USER_WANTS job, ignoring: Transaction for sound.target/start is destructive (systemd-exit.service has 'start' job queued, but 'stop' is included in transaction).
Feb 14 10:32:25 systemd[2124]: sys-devices-pci0000:00-0000:00:08.1-0000:74:00.1-sound-card1-controlC1.device: Failed to enqueue SYSTEMD_USER_WANTS job, ignoring: Transaction for sound.target/start is destructive (shutdown.target has 'start' job queued, but 'stop' is included in transaction).
Feb 14 10:32:30 dbus-broker-launch[958]: Activation request for 'org.freedesktop.nm_dispatcher' failed.
Feb 14 10:32:30 (udev-worker)[3067]: zram0: Failed to process device, ignoring: Device or resource busy
I’ve checked and confirmed the following file does not exist on my system
/etc/dracut.conf.d/nvidia.conf
I checked the other post, but I’m not sure it’s the same. It talks about the black screen before FDE password, but the black screen happens after I enter my password.
Ok, then no need to rebuild the initramfs (with sudo dracut ...) after akmods has
built the nvidia kernel module.
Right, but I was fearing that having the nvidia kernel module in the initramfs may explain
the fallback to the nouveau driver.
I wonder thus if the 6.18.9 kernel nvidia module has been signed with the correct key.
Can you start a root shell with: sudo bash --login
then execute from it the following and paste here the result ?
Why are you messing with the keys? Have you generated a new key every time? You should generate the key only once! Then skip this part and make sure all modules are signed with the same key.
check whether current key is enrolled: sudo mokutil --test-key /etc/pki/akmods/certs/public_key.der
and make sure all nvidia kernel modules are signed with that key. Otherwise akmods --rebuild --force --kernels <kernel version>
you can also list all enrolled keys with sudo mokutil --list-enrolled --short
Since I’ve generated multiple, do I need to delete the keys and start from scratch? I did select clear keys when disabling Secure boot, but I’m assuming that’s only in the BIOS and I might also need to delete them in Fedora?
+ ls --color=auto -l /etc/pki/akmods/private/
total 28
-rw-r-----. 1 root akmods 3272 Feb 1 21:57 Acans-PC_1769943420_23800e61.priv
-rw-r-----. 1 root akmods 3272 Feb 14 00:00 Acans-PC_1770987621_9ba6daea.priv
-rw-r-----. 1 root akmods 3272 Jan 18 13:34 fedora_1768703685_c1a8c53c.priv
-rw-r-----. 1 root akmods 3272 Feb 1 22:06 fedora_1769943964_2d4ba3bc.priv
-rw-r-----. 1 root akmods 3268 Feb 14 13:11 fedora_1771035061_167217ea.priv
-rw-r-----. 1 root akmods 3272 Feb 14 13:25 fedora_1771035932_9f76397a.priv
lrwxrwxrwx. 1 root akmods 55 Feb 14 13:25 private_key.priv -> /etc/pki/akmods/private/fedora_1771035932_9f76397a.priv
+ modinfo -F signer /lib/modules/6.18.5-200.fc43.x86_64/extra/nvidia/nvidia.ko
+ modinfo -F signer /lib/modules/6.18.9-200.fc43.x86_64/extra/nvidia/nvidia.ko
+ rpm -qi kmod-nvidia-6.18.5-200.fc43.x86_64
+ grep --color=auto 'Build Date'
Build Date : Sat 14 Feb 2026 12:43:22
+ rpm -qi kmod-nvidia-6.18.9-200.fc43.x86_64
+ grep --color=auto 'Build Date'
Build Date : Sat 14 Feb 2026 12:42:50
It returned the key is enrolled. I ran rebuild, and it successfully rebuilt for kernel 6.18.9-200, but upon restart, the same issue occurs. It also gave an error saying it couldn’t find the files for 6.18-5-200.
pls post journalctl -b -1 -g NVRM of that failed attempt. adjust -1= previous boot , -2 the boot attempt before that etc and the akmods error for 6.18-5-200.
pls also post output for rpm -qa kernel\* kmod-nvidia\* |sort
and
what is the output of
modinfo -F signer /lib/modules/6.18.9-200.fc43.x86_64/extra/nvidia/nvidia.ko
and modinfo -F signer /lib/modules/6.18.5-200.fc43.x86_64/extra/nvidia/nvidia.ko
Could not find files needed to compile modules for 6.18.5-200.fc43.x86_64
Are the development files for kernel 6.18.5-200.fc43.x86_64 or the appropriate kernel-devel package installed?
Checking kmods exist for 6.18.9-200.fc43.x86_64 [ OK ]
Building and installing nvidia-kmod [ OK ]
I’ve just always liked having more screens for less minimising.