No login screen (black screen after boot) after updating leap on 2016-09-27.

English Install/Boot/Login
#1

Env : leap 42.1 up to date.

Card : Nvidia 7200, nvidia driver.

After updating yesterday, the system every time the screen lock mechanism kicked in, gave me a black screen saying that something was messed up and that I needed to do a “loginctl unlock-sessions” to log back in. I figured the update needed to reboot so I did.

After rebooting (I see the graphical splash screen), the system boots to a “black screen”. I can ctrl-alt-backspace and the X server seems to reboot, but I can’t get a GUI login at all.

I can ctrl-alt-FX and go to text login.

Any help?

#2

Video driver??

If you installed the driver the hardway you need to reinstll

You can try the fall back driver by using nomodeset at the boot.

Press e then find line starting linux or linuxefi go to true end (it wraps) enter a space then nomodeset press F10 to continue boot. This will use one of the fall back drivers and it may not have full resolution but should allow the repair to proceed

#3

ok, I’ll try it in a couple of hs when I get back to the system. The driver is the official nvidia driver loaded from suse/nvidia repositories.

#4

ok, here are the results

Tried the boot option, didn’t work. Still getting a black screen.

However, I’m seeing this on journalctl -b

ep 28 20:48:29 lxserver udisksd[2042]: Acquired the name org.freedesktop.UDisks2 on the system message bus
Sep 28 20:48:29 lxserver systemd[1]: Starting Daemon for power management…
Sep 28 20:48:29 lxserver dbus[1334]: [system] Successfully activated service ‘org.freedesktop.UPower’
Sep 28 20:48:29 lxserver systemd[1]: Started Daemon for power management.
Sep 28 20:48:30 lxserver sddm-greeter[2033]: Adding view for “VGA-1” QRect(0,0 1920x1080)
Sep 28 20:48:30 lxserver sddm-greeter[2033]: QDBusConnection: name ‘org.freedesktop.UDisks2’ had owner ‘’ but we thought it was ‘:1.24’
Sep 28 20:48:30 lxserver sddm-greeter[2033]: Message received from daemon: Capabilities
Sep 28 20:48:30 lxserver sddm-greeter[2033]: Message received from daemon: HostName
Sep 28 20:48:30 lxserver sddm-greeter[2033]: file:///usr/share/sddm/themes/breeze/components/InfoPane.qml:52:22: Unable to assign [undefined] to int
Sep 28 20:48:30 lxserver sddm-greeter[2033]: Failed to create OpenGL context for format QSurfaceFormat(version 2.0, options QFlags(), depthBufferSize 24, redBufferSize -1, greenBufferSize -1, blueBufferSize -1, alphaBufferSize -1, stencilBufferSize 8, samples -1, swapBehavior 2, swapInterval 1, profile 0)
Sep 28 20:48:30 lxserver sddm-helper[2029]: [PAM] Ended.
Sep 28 20:48:30 lxserver sddm[1989]: Auth: sddm-helper exited with 6
Sep 28 20:48:30 lxserver sddm[1989]: Greeter stopped.

I saw some hits on google about the sddm user having to be on the video group, but that didn’t work either.

#5

There are similar hits about this issue:

https://forum.kde.org/viewtopic.php?f=67&t=130624

#6

Yup, for the fourth time, Nvidia has borked my system. I will never buy an Nvidia GPU again. They have ruined my confidence in them. One time is a programmer error…4 times is intended to weed out older gpu’s . The repository should know by now at this point in time what gpu is installed, and to stop switching to a driver that totally blackscreen borks the **** out of the OS… I have researched recovery options…none of them work. So now I’ll spend another 2 weeks tweaking my OS again because Nvidia is just another greedy company…Tired of getting shafted…

#7

Hey

I’m running the G02 version of the driver. However, I tried de-installing it and I’m still getting a black screen without the nvidia driver (but still the nvidia card).

I just checked a laptop running leap and the thing is, I find no “sddm” on journalctl… on a second view, the thing is that laptop is not running breeze, but another theme.

#8

For the GTX 650 you should be using the G04 drivers (367.44 is current at the moment).

I’ve run the G04 series for my GT610 since day one - never had a problem.

Here is the Nvidia Support page for the G04 drivers. Click the “Supported Products” tab on this page to see the list of all supported devices.
The GTX 650 is in that list - along with most other GeForce and Quadro products.
NOTE - DO NOT install from this web page, install the G04 drivers from the openSuse Nvidia repos instead.

http://www.nvidia.com/download/driverResults.aspx/106780/en-us

#9

frankensuse:

I forgot to mention…
I provided the info above because the signature section of your original post states you are using a GTX 650 with the G03 driver…

#10

[HR][/HR]ok, got it work somewhat. Went to yast, sysconfig editor, and replaced the Display MAnager from sddm to kdm, and now I’m able to log in using icewm.

Plasma is giving errors and I can’t get to it : ] kactivitymanage[2242]: segfault at 7fa426571d10 ip 00007fa42658869a sp 00007ffdc558e528 error 4 in libQt5Sql.so.5.5.1[7fa426573000+3f000]

#11

ok, this thing seems solved for me.

I tried a couple of more things on icewm : glxinfo and glxgears wouldn’t work, crashed, but worked if I ran them as root. When it crashed, it gave some gl/glx errors. Even nvidia-settings or good old xscreensaver wouldn’t do the gl savers.

So called Nvidia and they gave me a link to download 304.131 drivers, instead of the latest 304.132. Deleted 304.132 from the system , reboot, ran nvidia script to install the old ones, rebooted, and now sddm works, plasma works, glxgears/glxinfo works, nvidia-settings works.

The guy finally told me not to update at least for a few days to 304.132, or stay with the 131. It seems there might be newer drivers coming. Maybe they are aware of the issue. Not sure if this is hitting owners of newer cards, but certainly caused problems on my system with a 7xxx card.

Regards!

#12

**xlis, **

That is correct I am using the G03, every time I installed the G04 including just the other day I get the black screen of death. The only way I have found to recover from this is to reinstall the OS, load the G03 then remove the repository so it won’t send me that update to G04 anymore.

I have no clue why G04 borks my system, but it does, and I would not go near it with a ten foot pole anymore.

Luckily I have my home folder and most of my settings on another drive so it is fairly easy to reinstall, I just have to reinstall the apps I had, and do some tweaks…But I am back up and running, nice and smooth with G03…

#13

GO3 will not be updated GO4 unless GO4 is already installed.

You can recover by booting to a terminal and running yast from command line (note may need nomodeset in some situations). No need to reinstall

To boot to terminal press 3 at boot screen find line starting linux or linuxefi go to true end of line (it wraps) add a space and a 3

#14

You would think that G03 would not be updated to G04, unless 04 was already installed, but that is just not the case. Now it won’t because I pulled the repository… like I said this is not the first time this happened to me, As for getting to a terminal at boot by pressing 3 that didn’t work either. I tried that… Nothing but black screen… My only recourse was to reinstall the OS.

#15

oops typo not 3 press e at boot

GO3 will not become GO4 by magic. Note it is always best to remove the NVIDIA driver before installing another since junk can be left behind if you just overwrite it

You absolutely do not need to reinstall

#16

Hi Folks,

I am having a similar issue here. I have tried to find the problem, but it took me so much effort to put the system working again that I am a little bit lazy about installing G04 drivers again and copying the messages by hand, but I might do that if it is really needed. I am currently using nouveau.

What I found out was that the system enters a loop trying to load the two nvidia kernel modules (nvidia and nvidia-uvm). If the grub bootline includes the option “quiet”, the user is not able to see what is happening, and the system looks frozen, when in fact it is looping.

You really do not need to reinstall the system, but you do need a rescue disk, do some chroot magic and remove nvidia rpms by hand using rpm, not zypper. Unless zypper has a “no-network” mode, which I suspect is a good idea, but that one just occured me right now. :slight_smile:

If anyone can point me where should I look or which script should I check and spare me some hacking, I can certainly examine the problem more closely, otherwise I will have to wait until I can put my system down again without risking punishment from my folks at home :slight_smile:

Regards!

#17

Ok, here comes the log of the loop.

lspci is:

01:00.0 VGA compatible controller: NVIDIA Corporation GF108 [GeForce GT 730] (rev a1)

Nvidia site explicitly says that the 367.44 driver does support GT 730.

In my opinion, there are two problems here:
1 - The system initialization should not loop forever trying to load a module it has already tryed to load and failed
2 - The 367.44 NVIDIA driver should support my hardware.

The NVRM messages repeat forever:


...
Oct 09 09:17:23 linux-af78 kernel: [drm] Initialized drm 1.1.0 20060810
Oct 09 09:17:23 linux-af78 kernel: nvidia_modeset: module license 'NVIDIA' taints kernel.
Oct 09 09:17:23 linux-af78 kernel: Disabling lock debugging due to kernel taint
Oct 09 09:17:23 linux-af78 systemd-journal[120]: Journal started
Oct 09 09:17:23 linux-af78 kernel: nvidia_modeset: Unknown symbol nvidia_register_module (err 0)
Oct 09 09:17:23 linux-af78 kernel: nvidia_modeset: Unknown symbol nvidia_get_rm_ops (err 0)
Oct 09 09:17:23 linux-af78 kernel: nvidia_modeset: Unknown symbol nvidia_unregister_module (err 0)
Oct 09 09:17:23 linux-af78 systemd[1]: Started Journal Service.
...
Oct 09 09:17:31 linux-af78 systemd[1]: Mounting /home...
Oct 09 09:17:31 linux-af78 mtp-probe[579]: checking bus 6, device 2: "/sys/devices/pci0000:00/0000:00:1d.2/usb6/6-1"
Oct 09 09:17:31 linux-af78 mtp-probe[578]: checking bus 5, device 2: "/sys/devices/pci0000:00/0000:00:1d.1/usb5/5-2"
Oct 09 09:17:31 linux-af78 mtp-probe[578]: bus: 5, device: 2 was not an MTP device
Oct 09 09:17:31 linux-af78 mtp-probe[579]: bus: 6, device: 2 was not an MTP device
Oct 09 09:17:32 linux-af78 kernel: vgaarb: device changed decodes: PCI:0000:01:00.0,olddecodes=io+mem,decodes=none:owns=io+mem
Oct 09 09:17:32 linux-af78 kernel: NVRM: The Unknown NVIDIA G8x Series GPU installed in this
                                   NVRM:  system is not supported through the 367.44 NVIDIA
                                   NVRM:  driver. You can try the NVIDIA 340.xx Legacy drivers.
                                   NVRM:  Please visit http://www.nvidia.com/object/unix.html
                                   NVRM:  for download information.  The 367.44 NVIDIA driver
                                   NVRM:  will ignore this GPU.  Continuing probe...
Oct 09 09:17:32 linux-af78 kernel: nvidia: probe of 0000:01:00.0 failed with error -1
Oct 09 09:17:32 linux-af78 kernel: nvidia-nvlink: Nvlink Core is being initialized, major device number 246
Oct 09 09:17:32 linux-af78 kernel: NVRM: The NVIDIA probe routine failed for 1 device(s).
Oct 09 09:17:32 linux-af78 kernel: NVRM: None of the NVIDIA graphics adapters were initialized!
Oct 09 09:17:32 linux-af78 kernel: nvidia-nvlink: Unregistered the Nvlink Core, major device number 246
Oct 09 09:17:32 linux-af78 kernel: NVRM: NVIDIA init module failed!
Oct 09 09:17:32 linux-af78 systemd[1]: Mounted /home.
Oct 09 09:17:32 linux-af78 systemd[1]: Starting Local File Systems.
Oct 09 09:17:32 linux-af78 systemd[1]: Reached target Local File Systems.
Oct 09 09:17:32 linux-af78 systemd[1]: Starting Tell Plymouth To Write Out Runtime Data...
Oct 09 09:17:32 linux-af78 systemd[1]: Starting Trigger Flushing of Journal to Persistent Storage...
Oct 09 09:17:32 linux-af78 kernel: EXT4-fs (sdb1): mounted filesystem with ordered data mode. Opts: (null)
Oct 09 09:17:32 linux-af78 kernel: vgaarb: device changed decodes: PCI:0000:01:00.0,olddecodes=none,decodes=none:owns=io+mem
Oct 09 09:17:32 linux-af78 kernel: NVRM: The Unknown NVIDIA G8x Series GPU installed in this
                                   NVRM:  system is not supported through the 367.44 NVIDIA
                                   NVRM:  driver. You can try the NVIDIA 340.xx Legacy drivers.
                                   NVRM:  Please visit http://www.nvidia.com/object/unix.html
                                   NVRM:  for download information.  The 367.44 NVIDIA driver
                                   NVRM:  will ignore this GPU.  Continuing probe...
Oct 09 09:17:32 linux-af78 kernel: nvidia: probe of 0000:01:00.0 failed with error -1
Oct 09 09:17:32 linux-af78 kernel: nvidia-nvlink: Nvlink Core is being initialized, major device number 246
Oct 09 09:17:32 linux-af78 kernel: NVRM: The NVIDIA probe routine failed for 1 device(s).
Oct 09 09:17:32 linux-af78 kernel: NVRM: None of the NVIDIA graphics adapters were initialized!
Oct 09 09:17:32 linux-af78 kernel: nvidia-nvlink: Unregistered the Nvlink Core, major device number 246
Oct 09 09:17:32 linux-af78 kernel: NVRM: NVIDIA init module failed!
Oct 09 09:17:32 linux-af78 kernel: vgaarb: device changed decodes: PCI:0000:01:00.0,olddecodes=none,decodes=none:owns=io+mem
Oct 09 09:17:32 linux-af78 kernel: NVRM: The Unknown NVIDIA G8x Series GPU installed in this
                                   NVRM:  system is not supported through the 367.44 NVIDIA
                                   NVRM:  driver. You can try the NVIDIA 340.xx Legacy drivers.
                                   NVRM:  Please visit http://www.nvidia.com/object/unix.html
                                   NVRM:  for download information.  The 367.44 NVIDIA driver
                                   NVRM:  will ignore this GPU.  Continuing probe...
Oct 09 09:17:32 linux-af78 kernel: nvidia: probe of 0000:01:00.0 failed with error -1
Oct 09 09:17:32 linux-af78 kernel: nvidia-nvlink: Nvlink Core is being initialized, major device number 246
Oct 09 09:17:32 linux-af78 kernel: NVRM: The NVIDIA probe routine failed for 1 device(s).
Oct 09 09:17:32 linux-af78 kernel: NVRM: None of the NVIDIA graphics adapters were initialized!
Oct 09 09:17:32 linux-af78 kernel: nvidia-nvlink: Unregistered the Nvlink Core, major device number 246
Oct 09 09:17:32 linux-af78 kernel: NVRM: NVIDIA init module failed!
Oct 09 09:17:32 linux-af78 systemd[1]: Started Tell Plymouth To Write Out Runtime Data.
Oct 09 09:17:32 linux-af78 kernel: vgaarb: device changed decodes: PCI:0000:01:00.0,olddecodes=none,decodes=none:owns=io+mem