Stuck in Blackscreen after Reboot

General system Kernel, boot, graphics & hardware
1

Hey party people, emberassing as it is I used perplexity AI for some troubleshooting after not being able to open my games via heroic (+ protonGE) post new gpu (amd rx 5600 xt) and after a reboot am now stuck in blackscreen, from which I can now access TTY, but am not getting further, I managed to remove additions to the grub config, that were meant to help with the original issue. However after rebooting via TTY after the config changes the blackscreen stayed. I have tried restarting sddm which triggered some processes but ended up back in black.

I really hope somebody has mercy to help me out with this dumpsterfire I put myself into :slight_smile:

2

Without knowing what you have actually done it is going to be hard for anyone to give you any advise on what to do.

3

If you are using GRUB and EOS then go and select the fallback initramfs kernel option. It will be something like
EndeavourOS Linux, with Linux linux (fallback initramfs)
or
EndeavourOS Linux, with Linux linux-lts (fallback initramfs)

Also what did you change? Did you put some boot parameters or modified something in kernel?

4

Hi, thanks for responding. I currently have no access to the device, but I know what you are talking of and that there are multiple options all with very similar names. Could I later on give you that list and you help me choose with of those to boot?

Also I think the issue may be connected to this: sudo pacman -Rns vulkan-radeon lib32-vulkan-radeon vulkan-icd-loader lib32-vulkan-icd-loader libdisplay-info lib32-libdisplay-info 2>/dev/null || true
sudo rm -rf /usr/share/vulkan/icd.d/*
rm -rf ~/.cache/vulkan* /tmp/vulkan*

I asked to do a clean sweep of the modifications relating to the JSONs that were made before because I already had the feeling this was best not fixed by AI. However this seemed to break a lot of dependencies, but i think I reinstalled those back before rebooting (if I haven’t, would that most likely be the reason for the TTY-only situation?).

After that I tried restarting sddm, no change. Then I added, “nomodeset” to the line “GRUB_CMDLINE_LINUX_DEFAULT="nowatchdog nvme_load=YES loglevel=3 “ in the grub config, which was supposed to bring Plasma back I guess. I didn’t work and I removed it from the grub . Next I tried adding “amdgpu.runpm=0” rebooted and still no change.

From what I can tell booting the fallback initramfs is the best option. Is there one on the list, that is the go to?

Thanks again

5

You did a sudo pacman -Rns ..... 2>/dev/null that redirected the error output to /dev/null?

What did you search? Where did you search? Please do not tell me ChatGPT.

1 Like
6

Well it wasn’t on ChatGPT, however it may very well have been GPT, since I used perplexity. I know it is bad, I knew before and I know even more now..

I tried the fallback initramfs, however am still stuck in tty.

I don’t understand what exactly the ramifications of what specifically I did with command you mentioned. However I would like to ask you if you please could help me revert it, or if that is possible at all.

Worst Case but also acceptable because there wasnt anything major on the device yet I could do a fresh OS install but would appreciate some help with that as well, as I have a dual boot with a small win10 partition, that I dont know if I could leave? Like I know I had to start creating the win10 partition to prevent it from activating safebios and blocking the linux partition created before.

7

First of all use a live EOS USB or something equivalent like SystemRescueCD or Rescuezilla or anything else that you have used and boot the system. Copy out your user data into an external drive or upload it into cloud. User data refers to all the docs, images, pdfs, spreadsheets, coding projects, etc which typically reside in /home

If you are able to get to TTY/Virtual Terminal, and assuming that you are using EOS, then try to login with normal username and password. Once you are in, try to start your DE. Assuming that you were using KDE the command will either be startplasma-wayland or startplasma-x11. If you were not using KDE then give the command startx. Check if this works.

The steps and command outlined below can be executed in a terminal once you start your Desktop Environment as given above or from TTY/Virtual Terminal post your login.

What we are looking for are the boot logs. We have to see where it the boot failing. I am assuming that you have systemd journal installed and configured. So after you have logged in give the command
sudo journalctl --list-boots >> $HOME/ListOfBoots.log
sudo journalctl --list-boots

From this look at the last boot that was successful, For example look for boots prior to Feb-26. I am assuming that 2 days ago the boot happened successfully for you. If not then look at earlier boots. The second command will give you and output similar to the one given below. It will not be exactly the same. There might be more entries or less entries.

IDX BOOT ID                          FIRST ENTRY                 LAST ENTRY
-4  312f36ccc2ff4f56875585f5612531ef Mon 2026-02-24 23:12:14  Mon 2026-02-24 23:25:06 
-3  bfea0812954b45018c8b65c634c38b77 Mon 2026-02-24 23:25:35  Mon 2026-02-24 23:54:04 
-2  3068beb397984b9d890cd554b64cc732 Tue 2026-02-26 05:32:47  Tue 2026-02-26 05:19:50 
-1  c9a69fa43c6d44a2be1b34a1707b7ca6 Thu 2026-01-26 10:22:29  Mon 2026-02-26 10:20:01 
-0  9757e81609c94d438196555447c63b6d Mon 2026-02-27 08:20:38  Mon 2026-02-27 08:32:46

Look at the column BOOT ID. The value for that you will get for BOOT ID on your system will be different In the sample given above. Now if we are looking for logs for 26-Feb-2028 which booted successfully you will need to look at the boot logs for c9a69fa43c6d44a2be1b34a1707b7ca6.
So look at the boot log
sudo journalctl --boot=c9a69fa43c6d44a2be1b34a1707b7ca6 >> $HOME/bootlogs_working.log
sudo journalctl --boot=c9a69fa43c6d44a2be1b34a1707b7ca6
In your case the c9a69fa43c6d44a2be1b34a1707b7ca6 value will be different and you will have to replace c9a69fa43c6d44a2be1b34a1707b7ca6 with something equivalent.

Now look at the latest boot log by giving the command
sudo journalctl -b>> $HOME/bootlogs_LatestNotWorking.log
sudo journalctl -b
Compare both of the boot logs and see where it is failing and where the errors are coming from search google for the fixes. There might be multiple errors and you will have to resolve all of them one by one. It is going to take time.
After this look at the KErnel message, dmesg, for errors and critical messages. Use the following command
dmesg -T --level=crit,err,emerg,alert
Identify the messages and search google for the fixes.

DO NOT USE CHATGPT/PERPLEXITY OR ANY OTHER LLM FOR SEARCHING OF FIXES. ALSO KEEP A RECORD OR DIARY FOR WHAT YOU HAVE DONE.

The diary or record can be maintained in a simple text file. vi or kwrite are your friends use them.

Now in the worst case that you are not able to log into TTY/Virtual Terminal or launch your Desktop environment then install SystemRescueCD or EOS Live USB on a USB Memory Stick. Boot from the USB memory stick and mount the partitions of your system which are not working. systemd logs are typically kept in the ‘/var/log/journal/’. Assuming that you have mounted you partition into /mnt/nonworkingsys directory then give the following commands
sudo journalctl -D /mnt/nonworkingsys/var/log/journal --list-boots
sudo journalctl -D /mnt/nonworkingsys/var/log/journal --boot=c9a69fa43c6d44a2be1b34a1707b7ca6
sudo journalctl -D /mnt/nonworkingsys/var/log/journal -b
Replace c9a69fa43c6d44a2be1b34a1707b7ca6 with a value from your case.
Then look for all the errors and search in google for fixes.

8

See my response posted just a few minutes ago.
Going forward for all changes that you do, especially those taken from ChatGPT or Preplexity or some other LLM, please keep a record or diary. What you did and why you did. This can be a in a simple text file or a markdown file or anything else, but which you can access in times of crisis. Text files are the best as they do not require any special software to view them. Pen and paper also works.

Take a image of the Windows 10 OS partition. You can use RescueZilla or anything else that you have.

IT is possible that you will have to re-install your Linux system.

9

Thank you so much for taking your time to help me out. I will try my best to follow your steps and solve the issues independently and with guidance of the community. Thank you again :folded_hands:t4::folded_hands:t4:

10

Would be helpfull to know on what you did to fix this issue. It will help others out in future.

2 Likes