Hardware • Suddenly window manager won't load, driver errors

Hi all, have been a Linux user close to 20 years now, a Debian user for 16 of those I'd say, and I'm currently a Linux SysAdmin for a department at the University for about 8 years now, so I know a thing or two about Linux and Debian, but there's even more that I don't know.  Here's one example where yesterday I wiggled the mouse to wake up the screen on my workstation and login as usual, except after I entered my password, the login window froze.  

This system is running Debian 11 still and, at the time, it was running Linux 5.10.0-21-amd64, Debian 5.10.162-1 (2023-01-21).

Keyboard and mouse were fine and I switched to a console to login.  There was nothing new in the log files, so I just reboot'd the system since it had some updates recently.  It booted up fine as I saw all the green OK messages as it loaded up everything. One of the last ones I see on this screen is [ OK ] Starting GNOME Display Manager and screen goes black and login window opens.  Except this time I was greeted with the built-in autoblinder mechanism, on dual monitors, as depicted here https://redhat.discourse-cdn.com/fedora ... 45d7a0.png (sorry for tainting the forum with redhat material in my first post, but that's what ya get for attacking my eyes!) telling me there's a problem it can't recover from and to contact a sys admin. Egads computer, that's me!

I switched back to a nice, black console and I was then shown these messages:Well, I can see it doesn't like 084 anymore, but why?  BTW, 01:00.0 is a:
01:00.0 VGA compatible controller: NVIDIA Corporation GK107 [GeForce GTX 650] (rev a1)

So far, I've tried:

• Upgrade xserver-xorg-video-nouveau:amd64 (1:1.0.17-1, 1:1.0.17-2)
• Upgrade: linux-image-amd64:amd64 (5.10.162-1, 6.1.99-1) (2024-07-15)
• Switched to the nvidia-tesla-470-driver packaged firmware

Now, the login window doesn't even load, but there are no errors shown like with the nouveau firmware, but the console font and resolution have turned to sh*te, maybe 640x480 or less and Instead, these are the latest from dmesg:I'm not sure what else to try to get it working.  I found someone out there that had quite a technical approach with manual archive downloads from a vendor? somewhere and manually extracting code into the kernel source tree?  I can find that again probably but that seemed way to technical and frakendebian as a viable solution.

It seems wrong that this would suddenly start happening, and both updated nouveau and Nvidia drivers fail to ... well I don't know exactly, what happens when the display changes from the low-res text-based console to high-res graphics environment?  Is the difference between the video modes used for the consoles and during booting compared to what something like Gnome uses able to be caused from a hardware issue??  It has to be hardware related I'm guessing, but not sure how the initial video during POST and boot and for the consoles is working if there's a hardware issue with the card. So, as usual need to figure out what's broken and why and how to fix it. Would appreciate your feedback and insights to figure it out ASAP. Thanks!

Statistics: Posted by agi-sysadmin — 2024-08-16 23:31

---