[SOLVED] Black screen on bootup, fresh install works

From some unknown update or system change my multi-boot notebook no longer boots Manjaro XFCE w/kernel 4.14. I get a black/frozen screen.
So I created a new separate partition and installed the latest Manjaro Xfce which boots fine from the rEFInd menu.
I then used mhwd-chroot to install a the newer 4.17 kernel. But this has no change. I get what looks like it booting up, where a flash of green status logging followed by black screen and blinking cursor at the next phase of booting up. Any ideas? How can I see dmesg or other logs to try to find out what is failing during the bootup? Clearly this is some config issue with my old/original system partition.

Here is my overly verbose record of what I did. The first half is my re-installing, and the second half below “## PARTITION INFO ##” shows my partition and UUID config info.

One possibility is I remember monkeying with the video driver config - might have installed a different driver. I need to see how I can return that setting to default.

I found this “no screens found” error at the bottom of the Xorg.0.log file. Maybe it’s just the wrong video driver? How do I set the drivers back to default when I cannot boot it up?

[ 10.207] (EE) VESA(0): V_BIOS address 0x0 out of range
[ 10.207] (II) UnloadModule: “vesa”
[ 10.207] (II) UnloadSubModule: “int10”
[ 10.207] (II) Unloading int10
[ 10.208] (II) UnloadSubModule: “vbe”
[ 10.208] (II) Unloading vbe
[ 10.208] (EE) Screen(s) found, but none have a usable configuration.
[ 10.208] (EE)
Fatal server error:
[ 10.208] (EE) no screens found(EE)
[ 10.208] (EE)
Please consult the The X.Org Foundation support
at http://wiki.x.org
for help.
[ 10.208] (EE) Please also check the log file at “/var/log/Xorg.0.log” for additional information.
[ 10.208] (EE)
[ 10.214] (EE) Server terminated with error (1). Closing log file.

[ 10.193] (II) systemd-logind: logind integration requires -keeptty and -keeptty was not provided, disabling logind integration
[ 10.194] (II) xfree86: Adding drm device (/dev/dri/card0)
[ 10.202] (–) PCI:*(0:0:1:0) 1002:9851:17aa:3801 rev 0, Mem @ 0xe0000000/268435456, 0xf0000000/8388608, 0xf0c00000/262144, I/O @ 0x00004000/256, BIOS @ 0x???/131072
[ 10.202] (WW) Open ACPI failed (/var/run/acpid.socket) (No such file or directory)
[ 10.202] (II) LoadModule: “glx”
[ 10.202] (II) Loading /usr/lib/xorg/modules/extensions/libglx.so
[ 10.205] (II) Module glx: vendor=“X.Org Foundation”
[ 10.205] compiled for 1.19.6, module version = 1.0.0
[ 10.205] ABI class: X.Org Server Extension, version 10.0
[ 10.205] (II) LoadModule: “vesa”
[ 10.205] (II) Loading /usr/lib/xorg/modules/drivers/vesa_drv.so
[ 10.205] (II) Module vesa: vendor=“X.Org Foundation”
[ 10.206] compiled for 1.19.6, module version = 2.4.0
[ 10.206] Module class: X.Org Video Driver
[ 10.206] ABI class: X.Org Video Driver, version 23.0
[ 10.206] (II) VESA: driver for VESA chipsets: vesa
[ 10.206] (II) Loading sub module “vbe”
[ 10.206] (II) LoadModule: “vbe”
[ 10.206] (II) Loading /usr/lib/xorg/modules/libvbe.so
[ 10.206] (II) Module vbe: vendor=“X.Org Foundation”
[ 10.206] compiled for 1.19.6, module version = 1.1.0
[ 10.206] ABI class: X.Org Video Driver, version 23.0
[ 10.206] (II) Loading sub module “int10”
[ 10.206] (II) LoadModule: “int10”
[ 10.206] (II) Loading /usr/lib/xorg/modules/libint10.so
[ 10.207] (II) Module int10: vendor=“X.Org Foundation”
[ 10.207] compiled for 1.19.6, module version = 1.0.0
[ 10.207] ABI class: X.Org Video Driver, version 23.0
[ 10.207] (II) VESA(0): initializing int10
[ 10.207] (EE) VESA(0): V_BIOS address 0x0 out of range
[ 10.207] (II) UnloadModule: “vesa”
[ 10.207] (II) UnloadSubModule: “int10”
[ 10.207] (II) Unloading int10
[ 10.208] (II) UnloadSubModule: “vbe”
[ 10.208] (II) Unloading vbe
[ 10.208] (EE) Screen(s) found, but none have a usable configuration.
[ 10.208] (EE)
Fatal server error:
[ 10.208] (EE) no screens found(EE)
[ 10.208] (EE)

inxi -Fxzc0 shows this when booted from working partition sda7 install

Graphics:
Card-1: AMD Mullins [Radeon R4/R5 Graphics] driver: radeon v: kernel
bus ID: 00:01.0
Display: x11 server: X.Org 1.19.6 driver: ati,radeon
unloaded: fbdev,modesetting,vesa resolution: 1366x768~60Hz, 1920x1080~60Hz
OpenGL: renderer: AMD MULLINS (DRM 2.50.0 / 4.14.39-1-MANJARO LLVM 6.0.0)
v: 4.5 Mesa 18.0.3 direct render: Yes

inxi -Fxzc0 shows this when chrooted into not working partition sda8

Graphics:
Card-1: AMD Mullins [Radeon R4/R5 Graphics] driver: radeon v: kernel
bus ID: 00:01.0
Display: server: X.org 1.19.6 driver: none tty: 80x24
Message: Advanced graphics data unavailable for root. Old System?

This should be your start point to find out what caused your problem, IMHO.
You should inspect /var/log/pacman.conf
I guess you are comfortable with rEFInd updating and you have taken care of this, since it’s the most obvious possible reason for a broken system and you don’t mention anything.
In Archwiki refind there are a lot of things to checkout, for troubleshooting and/or proper configuration and update.

Note: I have no experience with rEFInd, so I can just give some ideas. Nevertheless, it seems you use GRUB as well, IDK in what way in relation to your broken system.

I’m using refind to boot the working sd7 partition so it’s not that. I’ve been using rEFInd for years with many different distro and kernel installs, am familiar with it. I don’t use grub usually, maybe to boot a different kernel.
I am aware that my changing something broke it - hence my stating that.
I’m thinking it’s likely I did a bone head move and changed the video driver. The notebook used the Radeon R4 AMD GPU. I bought a new big monitor and thought the other driver would work/perform better. Lets see if I can change it back.

The working sda7 does not use the video-vesa driver but the non-working sda8 does (below), so uninstalling it…
brad@G50 ~]$ mhwd-chroot # choosing sda8
[G50 /]# mhwd --listinstalled

Installed PCI configs:


              NAME               VERSION          FREEDRIVER           TYPE

        video-vesa            2017.03.12                true            PCI
       video-linux            2018.05.04                true            PCI
     network-r8168            2016.04.20                true            PCI

[brad@G50 ~]$ mhwd --remove pci video-vesa

Removing video-vesa…
Using default
Has lib32 support: true
Sourcing /var/lib/mhwd/local/pci/video-vesa/MHWDCONFIG
Processing classid: 0300
Sourcing /var/lib/mhwd/scripts/include/0300
checking dependencies…

Packages (1) xf86-video-vesa-2.4.0-1

Total Removed Size: 0,07 MiB

:: Do you want to remove these packages? [Y/n]
:: Processing package changes…
removing xf86-video-vesa…
:: Running post-transaction hooks…
(1/1) Arming ConditionNeedsUpdate…
‘/etc/X11/xorg.conf.d/90-mhwd.conf’ symlink is invalid! Removing it…

I cannot say I have a solution to get your graphic drivers back in order.
But generally the way to go about doing this is to first boot to prompt (no gui, so no graphics involved) and then fix the graphics from the prompt line; which is using mhwd commands to remove all graphic drivers and then reinstalling the correct ones. [1]

That (booting to prompt) is easily done in grub (add ‘3’ to linux line) but since you are using rEFInd, you can try to see if press F2 and use one of the boot options with ‘single’ in it. If that does not boot [2], try pressing F2 again and you can edit the boot and add a ‘single’ to that line.

Of course you can still boot (to prompt) using grub from the install media (careful with the right partition as you now have 2 manjaro’s - use sdax instead).

And you can still chroot to it from the new installation and try to fix graphics drivers instead of installing a new kernel.

Good luck.

[1] - I don’t understand how changing kernel (your way, installing 417 kernel chrooted to it) can help resolve a graphic driver issue.
[2] - My experience with rEFInd is that booting with /boot/refind_linux.conf is not reliable and is ‘fickle’, however manual boots with /boot/efi/refind/refind.conf is a bit better.
[3] - rEFInd does not boot intel-ucode in Manjaro (unless kernel is modified manually by user) - hope that is not the cause of your problem, though seemingly unlikely.
[4] - If you have grub installed (seems that way), you can boot up using grub through rEFInd in its ‘detect efi files’ alternatives. and use the ‘3’ outlined above to use grub to boot to prompt.

Yep, uninstalling the video-vesa driver as I did above w/chroot fixed it. What a relief! But gawd I spent a long time troubleshooting this. I think originally I had more than one problem. Gonna be more aware of what I change, and not changing stuff… and making a backup!

4 Likes

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.