tl;dr X11 hangs the system on 5.15.89 but runs fine on 5.10.90. No lines in the Xorg log contain (EE) and I cannot switch ttys or send SIGINT to startx after attempting.
Hello! I was able to successfully install Qubes OS and boot to it several times. However once I updated dom0 and rebooted, the system failed to start. Looking at the dmesg output as the system hung, I noticed that everything that had a Starting… message had a “[OK] Started” or “[OK] Finished” message except the last two lines: “Starting Light Display Manager…” and “Starting Hold until boot process finishes…”
Experimentally, I rebooted to a Fedora ISO on a Ventoy USB, mounted my fs, and renamed /etc/systemd/system/display-manager.service to display-manager.service.disabled. I rebooted to Qubes and, as expected, was met with the tty. After logging in successfully, I reenabled display-manager.service, ran systemctl daemon-reload, systemctl start display-manager, aaaaand… screen’t. Black screen, white underscore in top left corner. I tried switching ttys to no avail. I held the power button until the system turned off.
After disabling the display-manager service again, I looked around for any relevant logs from lightdm but found nothing. Since I’d already installed dwm, I decided to create a simple .xinitrc with “exec /usr/local/bin/launchdwm”. After running startx, the system hung, with the beginnings of the Xorg log visible at the bottom of the screen, and no ability to switch ttys. This led me to believe the issue was unrelated to lightdm. There were no (EE) messages in the Xorg log, and with nothing else to do, I held the power button once more.
I don’t remember why I did this, but I tried booting to kernel 5.10 in the advanced grub options. After reenabling the lightdm service, I found that I could start X11 on kernel 5.10 just fine.
I guess I can use my system like this, but I feel obligated to report the bug, if it is a bug and not my own stupidity. What information, if any, would be useful to provide?