Looks like the video drivers are probably the cause.
> systemd-modules-load[253]: Failed to insert module 'nvidia' . . . > systemd-modules-load[253]: Failed to insert module 'nvidia_drm' . . . > Failed to start Load Kernel Modules. > Failed to start Light Display Manager.
The output of inxi -b shows kernel 5.4.80 but also says no driver no installed is for my graphics card (GTX 760).
Device-2: NVIDIA GK104 [GeForce GTX 760] driver: N/A
With yesterday’s stable update, on my XFCE manjaro with kernel 5.10 and 5.4 (both with nvidia 340xx drivers), with kernel 5.10.19.1 it works perfect, but with 5.4.101.1 manjaro does not work, it does the disk check and then the screen is black with the cursor (until version 5.4.100.1 it worked perfect), apparently it is an incompatibility of the nvidia driver with the kernel, so I think we will have to wait for a new version of this kernel. Luckily I have 2 kernels installed (plus a backup with timeshift) so I don’t have to reinstall anything again
Yeah, I had a backup in timeshift too, but when I tried to revert to the backup I did just before the upgrade, everything went to hell, lol.
Not to be a help vampire, but can anyone give me an idea of how to proceed with this?
You’ve missed 6-7 stable updates already. The more you don’t update, the more likely your installation will be broken after the next update. There were also a lot of security updates between then and now.
You have to update every package installed to prevent partial upgrades, which can easily break your installation.
If you do not like to update often, I recommend going to a non-rolling release distro.
Instead of being a help vampire, looking up “Failed to start Load Kernel Modules” on the forum is a good start.
There is a reason why on the main page of the forum it says:
Please use the search function of the forum before posting any help requests.
This has been solved many many times. Majority of them are NVIDIA related due to the update in January, but posts solved the issue within the same day.
I did search it first, and there are a bunch of threads with different proposed solutions. So rather than blindly entering commands and hoping nothing else craters, I posted information specific to my system hoping to get specific information to fix it.
It’s not that I don’t like to update often; I’ve been running Manjaro for a couple years and the Dec update was the first one I’ve put off for any reason.
mhwd -a pci free 0300 output:
Warning: config '/var/lib/mhwd/local/pci/video-hybrid-amd-nvidia-prime/MHWDCONFIG' is invalid!
Skipping already installed config 'video-linux' for device: 0000:03:00.0 (0300:10de:1187) Display controller nVidia Corporation GK104 [GeForce GTX 760]
Skipping already installed config 'video-linux' for device: 0000:02:00.0 (0300:1002:683d) Display cotroller ATI Technologies Inc Cape Verde XT [Radeon HD 7770/8760 / /R7 250X]
pamac search -r nvidia flags these as installed:
> mhwd-nvidia-390xx [INSTALLED] 390.132-1 core
> mhwd-nvidia [INSTALLED] 460.39-1 core
> linux54-nvidia [INSTALLED] 460.39-9 extra
> linux419-nvidia [INSTALLED] 460.39-8 extra
> lib32-nvidia-utils [INSTALLED] 460.39-1 multilib
> nvidia-390xx-utils [INSTALLED] 390.141-1 extra
> nvidia-prime [INSTALLED] 1.0-4 extra
inxi -b relevant output:
Graphics:
Device-2: NVIDIA GK104 [GeForce GTX760] driver: N/A
No change when rebooting. Still getting Failed to start Load Kernel Modules
Do pacman -Syu and then try to re-install your kernel modules with sudo pacman -S linux54-nvidia linux419-nvidia that will in turn initiate a regeneration of initrd images.
Also tell us if your machine is an Optimus laptop or a desktop.
Thank you. It is a desktop. pacman is unable to sync the databases. Does this need to be done via a Live DVD & chroot so I have internet connectivity? I’m currently in TTY.
There are issues with mirrors lately.
What is the output of pacman-mirrors ?
To be sure, you may reset your mirrors with sudo pacman-mirrors -c all then sudo pacman -Syyu
It shows:
Pacman-mirrors version 4.16.4 : :Downloading status failed! : :Please check you network connection ...
I don’t know the TTY commands but obviously one has to fix network connection first.
Yeah, and I’ve no idea how to connect to WiFi from TTY, or if that’s even possible.
I searched and found the commands to connect to WiFi but it fails, telling me it can’t find a network with that SSID (which I know for a fact exists). If I have to drag everything to another room so I can cable it to ethernet, I will.
Maybe the best approcah would be to boot from a live DVD, be sure to connect to Internet, use sudo manjaro-chroot -a and check again from the pacman mirrors.
And maybe reset the mirror list and retry a full update.
What is done is done, but, these past 3 months were particularly rich in updates/new functionalities.
More frequent updates are recommended.
If I remember correctly, nm-cli can help you to establish a connection to your WiFi network.
I believe that was the one I tried using. I’m back in the Live environment now with Internet connectivity, I’m trying to work out some package dependency conflicts. I guess I’ll see how that goes. If it comes down to it, I suppose I could do a fresh install. My /home is on a physically separate disk, at least.
I’ve resolved the dependency conflicts. Now there are a bunch of warnings that all appear related to the nvidia driver. However, there’s a hook to trigger a timeshift backup that stops the upgrade from running. I’ve googled it and looked in the filesystem to try to figure out how to disable that hook for now, but I can’t find it. I assume I need to edit a .conf file somewhere?
warning: could not get file information for var/lib/mhwd/ids/pci/nvidia.ids
warning: could not get file information for usr/lib32/libEGL_nvidia.so.460.39
warning: could not get file information for usr/lib32/libGLESv1_CM_nvidia.so.460.39
warning: could not get file information for usr/lib32/libGLESv2_nvidia.so.460.39
warning: could not get file information for usr/lib32/libGLX_nvidia.so.460.39
warning: could not get file information for usr/lib32/libcuda.so.460.39
warning: could not get file information for usr/lib32/libnvcuvid.so.460.39
warning: could not get file information for usr/lib32/libnvidia-encode.so.460.39
warning: could not get file information for usr/lib32/libnvidia-fbc.so.460.39
warning: could not get file information for usr/lib32/libnvidia-ifr.so.460.39
warning: could not get file information for usr/lib32/libnvidia-ml.so.460.39
warning: could not get file information for usr/lib32/libnvidia-opticalflow.so
warning: could not get file information for usr/lib32/libnvidia-opticalflow.so.460.39
warning: could not get file information for usr/lib32/libnvidia-ptxjitcompiler.so.460.39
warning: could not get file information for usr/lib32/vdpau/libvdpau_nvidia.so.460.39
warning: could not get file information for usr/share/licenses/lib32-nvidia-utils
:: Running pre-transaction hooks...
(1/4) Creating Timeshift snapshot before upgrade...
Select backup device:
Num Device Size Type Label
------------------------------------------------------------------------------
Enter device name or number (a=Abort): E: Invalid input
Enter device name or number (a=Abort): E: Invalid input
Enter device name or number (a=Abort): E: Invalid input
E: Failed to get input from user in 3 attempts
Aborted.
Unable to run timeshift-autosnap! Please close Timeshift and try again. Script will now exit...
error: command failed to execute correctly
error: failed to commit transaction (failed to run transaction hooks)
Errors occurred, no packages were upgraded.Search it in /etc/pacman.d/hooks or /use/share/libalpm/hooks. Rename it with adding .bak to the end of its name to disable it.