Yesterday there was an update of 258 packages for Fedora 43 and after updating it failed to boot stopping before the login page. As one of the updates was a new kernel I tried the older kernels including the recovery kernel with the same result. Luckily I always backup on a regular basis and am back up running on a backup from the previous day. What is the best way to investigate this problem?
Look in the output from journalctl for the failed boot so see what was happening when the boot âstoppedâ.
By âstoppedâ, I presume you meant you were left with a black screen and you could not switch to an alternative TTY?
Yes it was a black screen and I could not switch to an alternative TTY but as I am now running on a backup therefore journalert shows nothing revelant.
Youâd need to run journalctl for the failed boot - see the -b flag to specify how many
âboots agoâ you want to view.
-b 0 is the current boot,
-b -1 is the previous boot and so on.
Sounds like the root partition was restored from a backup, so any evidence has been overwritten.
To really investigate this, youâd need to run the update again and âhopeâ it gets the system back into the broken state it was before.
Ahhh - an entire restore rather than just changing back to the previous working kernelâŚ
A large hammer to tap in a small nail, but if it worksâŚ
To actually investigate it, perform the same upgrade, boot the failing kernel and then revert to a working kernel and view the logs from the previous boot, with -b -1. As @pg-tips says, youâve wiped out the evidence so thereâs nothing to investigate until you replicate the issue.
tbf it sounds like just reverting to the previous kernel didnât work. So it was something in the userspace update that broke it, rather than the kernel.
thatâll teach me to actually read the question properly eh! You did state all of this in your question, and I glossed over it - my apologies.
Just based on that, Iâd be inclined to suspect an interaction between Plasma and the Nvidia drivers. (I guess youâre using the 470xx drivers with that card?)
How can I check what Nvidia Drivers I have installed as I donât remember adding any.
lspci | grep -i nvidia
01:00.0 VGA compatible controller: NVIDIA Corporation GK208B [GeForce GT 710] (rev a1)
01:00.1 Audio device: NVIDIA Corporation GK208 HDMI/DP Audio
You can check to see if youâre using the nouveau drivers with inxi -Gzxx
If that states youâre using the Nvidia drivers you can look at nvidia-smi to get the version. Inxi might also tell you the version in use, but I canât check as I no longer have any Nvidia cards installed.
$ inxi -Gzxx
bash: inxi: command not foundâŚ
Install package âinxiâ to provide command âinxiâ? [N/y]
Does not having inxi installed mean I have no nouveau drivers if so I will install it
inxi is a tool you can install that will report various hardware and driver information. Itâs not present on a standard Fedora install (regardless of whether nouveau is in use), but you can install it from the prompt you show.
Alternatively, lspci -vs 01:00.0 should also tell you which drivers are in use for your card.
I installed inxi and results follows
phil@fedora:~$ inxi -Gzxx
Graphics:
Device-1: NVIDIA GK208B [GeForce GT 710] vendor: ZOTAC driver: nouveau
v: kernel arch: Fermi-2 pcie: speed: 5 GT/s lanes: 8 ports: active: HDMI-A-1
empty: DVI-D-1,VGA-1 bus-ID: 01:00.0 chip-ID: 10de:128b temp: 31.0 C
Display: x11 server: X.Org v: 21.1.23 with: Xwayland v: 24.1.12
compositor: kwin_x11 driver: X: loaded: modesetting alternate: fbdev,vesa
dri: nouveau gpu: nouveau display-ID: :0 screens: 1
Screen-1: 0 s-res: 1920x1080 s-dpi: 96
Monitor-1: HDMI-A-1 mapped: HDMI-1 model: Dell ST2220L res: 1920x1080
hz: 60 dpi: 102 diag: 547mm (21.5")
API: EGL v: 1.5 platforms: device: 0 drv: nouveau device: 1 drv: swrast
gbm: drv: nouveau surfaceless: drv: nouveau x11: drv: nouveau
inactive: wayland
API: OpenGL v: 4.5 compat-v: 4.3 vendor: mesa v: 25.3.6 glx-v: 1.4
direct-render: yes renderer: NV106 device-ID: 10de:128b
API: Vulkan v: 1.4.341 surfaces: N/A device: 0 type: discrete-gpu
driver: mesa nvk device-ID: 10de:128b device: 1 type: cpu
driver: mesa llvmpipe device-ID: 10005:0000
Info: Tools: api: clinfo, eglinfo, glxinfo, vulkaninfo
de: kscreen-console,kscreen-doctor wl: wayland-info x11: xdriinfo,
xdpyinfo, xprop, xrandr
phil@fedora:~$ lspci -vs 01:00.0
01:00.0 VGA compatible controller: NVIDIA Corporation GK208B [GeForce GT 710] (rev a1) (prog-if 00 [VGA contro
ller])
Subsystem: ZOTAC International (MCO) Ltd. Device 7326
Flags: bus master, fast devsel, latency 0, IRQ 99, IOMMU group 9
Memory at fb000000 (32-bit, non-prefetchable) [size=16M]
Memory at 7ff0000000 (64-bit, prefetchable) [size=128M]
Memory at 7ff8000000 (64-bit, prefetchable) [size=32M]
I/O ports at f000 [size=128]
Expansion ROM at fc000000 [disabled] [size=512K]
Capabilities:
Kernel driver in use: nouveau
Kernel modules: nouveau
Youâre using the nouveau drivers and not any version of the Nvidia drivers.
Might want to try the Nvidia drivers from rpmfusion - considerably more performant that the nouveau drivers, and it may well clean up your issue. Canât make it any worse, unless youi have some reason for sticking with nouveau
Itâs worth a try to see if it fixes this. But there is a good reason to stick with nouveau - this card only supports the 470xx drivers, which donât support Wayland.
Right now thatâs not a problem because OP is running the X11 session in Plasma, but when Plasma 6.8 arrives, there will be no X11 support.
NOT
Installing those drivers would mean that he needs to use x11 instead of wayland and that would present other major issues
As I am running a X11 session in Plasma ( Wayland continually crashed ) it might be time to think of moving to another Desktop Session as there is no future for X11 support in Plasma. Would my Nvidia card cause problems on other Desktop Sessions?
Could always put a cheap second-hand AMD card in there - no faffing with drivers required, no issues running Wayland so you can use any desktop you want, no need to reinstall anything and you future-proof this tin for a few more years.