I am encountering an issue with my Fedora 40 system. It no longer boots properly.
The only notable changes I made before this problem started were:
Installing Wireshark using the package manager sudo dnf install wireshark
Adding my user to the wireshark group
su - myUser to use wireshark without having to login again to load the new group
Now when I try to boot my laptop the following happens:
I get prompted for my partition encryption password
After entering my password a black screen with Booting Fedora... appears and that’s where I get stuck. Usually that black screen only lasted a few seconds.
I tried using the grub menu to boot from the previous version and from the rescue version. They also get stuck on the black screen.
I used a live image to boot my laptop and access my file system. I extracted the boot.log file. It’s accessible on pastebin. I tried to include it as a details but it was too long for the platform.
Am’I right that the process seems to be stuck at Starting e[0;1;39msystemd-user-sessions.servicee[0m - Permit User Sessions... ?
Can anyone provide help on restoring access to my installation or help debug the problem ?
I’ll will probably have to re-install tonight if I can’t fix the problem (this is my work laptop).
PS: If I made any mistake while creating this post I’m more than willing to fix them (it’s my first time writing here).
I don’t see any obvious errors in the log you shared. However, Started mdatp.service - Microsoft Defender. and Starting docker.socket - Docker Socket for the API... might be somewhat uncommon additions to a Fedora Linux installation.
If all you see is a “black screen”, it might just be the login manager that is failing. Can you access a virtual terminal by pressing Ctrl+Alt+F3?
Hmm, well, the status report says it is running. I’m not sure if that locale error could cause the black screen issue you are seeing.
I guess you could try setting your locale back to the normal C.UTF-8 and restart the sddm.service to see if that clears the error message and maybe also makes the display manager work?
Here is what my system shows:
$ localectl status
System Locale: LANG=C.UTF-8
VC Keymap: (unset)
X11 Layout: (unset)
You might try sudo localectl set-locale C.UTF-8 to set your locale back to C.UTF-8 and then sudo systemctl restart sddm.service to restart your display manager. Then, Ctrl+Alt+F1 should switch you back to the graphical login screen if it works.
If it helps, I checked the output of sudo systemctl status display-manager.service on my system (I have no bootup or login issues, so consider this a “healthy control”).
I get the same locale error, so that likely isn’t relevant to @Driky’s problem here.
The difference I see is that I get “pam_unix … session opened” but I don’t get the “pam_unix … session closed” immediately afterward. Perhaps that’s relevant?
OK. P G’s helpful reply suggests the problem could be something in the PAM stack (Fedora Linux’s authentication subsystem). That could be difficult to troubleshoot and mistakes when editing it could easily lock you out of your system completely.
Perhaps installing and using a different display manager would be a reasonable workaround?
If that still doesn’t do the trick, you can try switching to the GNOME Desktop Manager with the following (it might pull down a lot of extra GNOME-related packages, unfortunately).
No, that link is for switching the Desktop Environment (what you see after you sign in). What you are trying to change right now is the Desktop Manager (the sign-in screen itself).
Edit: I see it has instructions for changing the Desktop Manager as well further down the page, however, I think those instructions might be outdated.
Tried Ctrl+Alt+F1, still a black screen. I tried to reboot, now I’m frozen on the splash screen after entering my decryption key. I don’t have access to the virtual terminal anymore.
You should be able to boot straight to the terminal by adding systemd.unit=mulit-user.target to the kernel command line.
Since the GNOME Display Manager didn’t work out, the first thing you might want to do after you get back into your system is sudo dnf history undo last. You will probably also need to run systemctl enable --force sddm.service.
By kernel command line do you mean the commands displayed when pressing e in grub ?
Do I put systemd.unit=mulit-user.target at the end or on the line starting with linux ?
The downside to that is that you won’t have networking when you get to your terminal. Oh, also, I think you might have to sign in as the root user. If you don’t have a password set for your root account, then that might not work.
That might have a decent chance of correcting whatever is wrong. I would undo that GDM installation first.
Edit: If you want to do the update from the command line, try sudo dnf upgrade --releasever=42. That command will give you a better chance at recovering from a situation where the update might fail. (Typically by giving you the chance to uninstall a package that fails to update and then re-trying the upgrade.)
Edit2: You might also want to take this opportunity to set your root account password just in case you do end up needing to use that rescue kernel option later on. Use sudo passwd root to set the root account’s password.
