Blank screen after upgrade to 12.3

I’ve done an upgrade from 12.2 to 12.3 using zypper. On reboot it gets past the boot screen and then stalls for 2 or 3 minutes at the lines:

[Linux-bzImage, setup=0x4200, size=0x4c00381]
initrd /initrd-3.7.10-1.16-desktop

The screen then goes blank. I can then ssh into it fine and it’s functioning fine in its role as webserver. But the attached screen stays blank, whether booting to desktop or cli.

I followed some of the guidance at SDB:Configuring graphics cards - openSUSE as follows:
Tried booting from grub with boot code: nomodeset.
Tried the boot option failsafe.
Tried booting into runlevel 3 by typing 3 when the very first openSUSE menu appears.
Tried this: ‘cp /etc/X11/xorg.conf.install /etc/X11/xorg.conf’

But exactly the same thing happens every time, stalling at that same line and then a blank screen. Can’t even get command line when trying to run level 3.

Would the video driver affect whether I can get the command line?

Anyway, video driver details:
04:00.0 VGA compatible controller [0300]: Matrox Electronics Systems Ltd. MGA G200e [Pilot] ServerEngines (SEP1) [102b:0522] (rev 02)
Subsystem: Fujitsu Technology Solutions Device [1734:108e]
Kernel driver in use: mgag200

Grub details: grub (GNU GRUB 0.97)

Would getting a newer kernel help?

If anything has to be set for X, don’t use xorg.conf, but use /etc/X11/xorg.conf.d and put what’s needed in 50-device.conf, uncommenting the tags.

  • Its an issue with the DRM/KMS driver: mgag200
  • yes, it will prevent you from getting a commandline … the framebuffer console won’t show if the fb driver is borking … when you boot, the fb console is set up, and initially driven by the vesafb (a kernel framebuffer driver), but if you are using a KMS based kernel video driver (which mgag200 now is), a hand-off to the kernel video driver is performed (i.e. vesafb → the fb supplied by the DRM/KMS driver … the kernel video driver has an inseparable fb driver embedded within it). Any messages you see on screen before the freeze would be from prior the point that the DRM driver tries to kick in. And although the video goes kaput from that point, as you’ve discovered, its none fatal and things proceed as expected, and hence why you can ssh into the machine. I’m sure if you look through your dmesg or boot log, you will find DRM related errors.
  • Updating to a newer kernel will likely resolve this (if it is indeed the same problem I believe it to be, as some fixes were commited to the DRM driver)
  • Note: when you resolve that issue, know that you shouldn’t need any xorg configurations … X should automatically load the generic “modesetting” driver for use with the mgag200 kernel driver … (there is no DDX driver available that supports the KMS pathway, hence the use of the generic driver)

Hmm … could also be same as this: https://bugzilla.redhat.com/show_bug.cgi?id=873542

Tried booting from grub with boot code: nomodeset…But exactly the same thing happens
This, however, is not what I would expect to have happened … would need to see more info.

In any regard, check your dmesg … if you want to share it with us (use susepaste) that’s fine too.

I tried installing the latest kernel but run into a whole different problem. Perhaps I should start a new thread? Anyway, I started off like this:
[size=2]sudo zypper ar [/size]sudo zypper ar http: //download.opensuse.org/repositories/Kernel:/stable/standard/ Kernel-Stable[size=2]
sudo zypper refresh
sudo zypper in --from Kernel-Stable kernel-desktop
[/size]

but ran out of space on the boot partition with the following messages:
Initrd image: /boot/initrd-3.11.3-1.gb1e6ec5-desktop
Root device: /dev/md3 (mounted on / as ext4)
Resume device: /dev/md0
Kernel Modules: thermal_sys thermal processor fan scsi_transport_sas mptbase mptscsih mptsas pata_acpi scsi_dh scsi_dh_rdac scsi_dh_hp_sw scsi_dh_alua scsi_dh_emc xhci-hcd hid-logitech-dj raid0 raid1 raid10 async_tx async_memcpy xor async_xor raid6_pq async_pq async_raid6_recov raid456
Features: acpi plymouth block usb md resume.userspace resume.kernel
cp: writing ‘/boot/initrd-3.11.3-1.gb1e6ec5-desktop’: No space left on device
cp: failed to extend ‘/boot/initrd-3.11.3-1.gb1e6ec5-desktop’: No space left on device
Failed to install initrd
There was an error generating the initrd (8)

Here’s the contents of my boot partition.
root@damh:/etc/X11> ls -alh /boot
total 77M
drwxr-xr-x 5 root root 1.0K Oct 7 10:35 .
drwxr-xr-x 25 root root 4.0K Oct 4 10:18 …
-rw------- 1 root root 512 Nov 18 2009 backup_mbr
lrwxrwxrwx 1 root root 1 Oct 3 10:34 boot → .
-rw-r–r-- 1 root root 1.5K Feb 26 2013 boot.readme
-rw-r–r-- 1 root root 138K Oct 3 16:43 config-3.11.3-1.gb1e6ec5-desktop
-rw-r–r-- 1 root root 133K Jun 7 10:12 config-3.7.10-1.16-desktop
-rw-r–r-- 1 root root 0 Oct 7 10:34 do_purge_kernels
drwxr-xr-x 2 root root 1.0K Oct 7 10:35 grub
drwxr-xr-x 7 root root 1.0K Oct 3 10:33 grub2
lrwxrwxrwx 1 root root 5 Oct 3 10:33 grub2-efi → grub2
lrwxrwxrwx 1 root root 32 Oct 7 10:34 initrd → initrd-3.11.3-1.gb1e6ec5-desktop
-rw------- 1 root root 17M Oct 7 10:35 initrd-3.11.3-1.gb1e6ec5-desktop
-rw------- 1 root root 32M Oct 3 10:39 initrd-3.7.10-1.16-desktop
drwx------ 2 root root 12K Nov 18 2009 lost+found
-rw-r–r-- 1 root root 607K Oct 3 10:37 message
-rw-r–r-- 1 root root 256K Oct 3 17:37 symvers-3.11.3-1.gb1e6ec5-desktop.gz
-rw-r–r-- 1 root root 236K Jun 7 11:28 symvers-3.7.10-1.16-desktop.gz
-rw-r–r-- 1 root root 516 Oct 3 17:37 sysctl.conf-3.11.3-1.gb1e6ec5-desktop
-rw-r–r-- 1 root root 516 Jun 7 11:28 sysctl.conf-3.7.10-1.16-desktop
-rw-r–r-- 1 root root 2.6M Oct 3 17:26 System.map-3.11.3-1.gb1e6ec5-desktop
-rw-r–r-- 1 root root 2.5M Jun 7 11:17 System.map-3.7.10-1.16-desktop
-rw-r–r-- 1 root root 5.9M Oct 3 17:37 vmlinux-3.11.3-1.gb1e6ec5-desktop.gz
-rw-r–r-- 1 root root 5.6M Jun 7 11:28 vmlinux-3.7.10-1.16-desktop.gz
lrwxrwxrwx 1 root root 33 Oct 7 10:34 vmlinuz → vmlinuz-3.11.3-1.gb1e6ec5-desktop
-rw-r–r-- 1 root root 5.1M Oct 3 20:33 vmlinuz-3.11.3-1.gb1e6ec5-desktop
-rw-r–r-- 1 root root 4.8M Jun 7 12:48 vmlinuz-3.7.10-1.16-desktop

Is there anything I can safely get rid of? I haven’t had this problem before during 3 or 4 previous upgrades on this machine. Will it be safe to reboot? Or will it try to use the new kernel which hasn’t installed correctly?

The new kernel will not work I guess.
But you should be able to boot the older kernel by choosing “Advanced Options” in the boot menu, maybe try one of the “recovery mode” options.

Your /boot seems to be bit too small, how big is it?

df -h

You could of course try to uninstall the 3.7.10 kernel and reinstall the 3.11 one (which then shouldn’t run out of space), but you won’t have the safety net of booting the older kernel then.

On 2013-10-07 12:36, domhnalldubh wrote:

> but ran out of space on the boot partition with the following
> messages:

> Here’s the contents of my boot partition.

Unreadable unless you use code tags. ‘#’ button.


> root@damh:/etc/X11> ls -alh /boot
> total 77M
> drwxr-xr-x  5 root root 1.0K Oct  7 10:35 .
> drwxr-xr-x 25 root root 4.0K Oct  4 10:18 ..
> -rw-------  1 root root  512 Nov 18  2009 backup_mbr
> lrwxrwxrwx  1 root root    1 Oct  3 10:34 boot -> .
> -rw-r--r--  1 root root 1.5K Feb 26  2013 boot.readme
> -rw-r--r--  1 root root 138K Oct  3 16:43 config-3.11.3-1.gb1e6ec5-desktop
> -rw-r--r--  1 root root 133K Jun  7 10:12 config-3.7.10-1.16-desktop
> -rw-r--r--  1 root root    0 Oct  7 10:34 do_purge_kernels
> drwxr-xr-x  2 root root 1.0K Oct  7 10:35 grub
> drwxr-xr-x  7 root root 1.0K Oct  3 10:33 grub2
> lrwxrwxrwx  1 root root    5 Oct  3 10:33 grub2-efi -> grub2
> lrwxrwxrwx  1 root root   32 Oct  7 10:34 initrd -> initrd-3.11.3-1.gb1e6ec5-desktop
> -rw-------  1 root root  17M Oct  7 10:35 initrd-3.11.3-1.gb1e6ec5-desktop
> -rw-------  1 root root  32M Oct  3 10:39 initrd-3.7.10-1.16-desktop
> drwx------  2 root root  12K Nov 18  2009 lost+found
> -rw-r--r--  1 root root 607K Oct  3 10:37 message
> -rw-r--r--  1 root root 256K Oct  3 17:37 symvers-3.11.3-1.gb1e6ec5-desktop.gz
> -rw-r--r--  1 root root 236K Jun  7 11:28 symvers-3.7.10-1.16-desktop.gz
> -rw-r--r--  1 root root  516 Oct  3 17:37 sysctl.conf-3.11.3-1.gb1e6ec5-desktop
> -rw-r--r--  1 root root  516 Jun  7 11:28 sysctl.conf-3.7.10-1.16-desktop
> -rw-r--r--  1 root root 2.6M Oct  3 17:26 System.map-3.11.3-1.gb1e6ec5-desktop
> -rw-r--r--  1 root root 2.5M Jun  7 11:17 System.map-3.7.10-1.16-desktop
> -rw-r--r--  1 root root 5.9M Oct  3 17:37 vmlinux-3.11.3-1.gb1e6ec5-desktop.gz
> -rw-r--r--  1 root root 5.6M Jun  7 11:28 vmlinux-3.7.10-1.16-desktop.gz
> lrwxrwxrwx  1 root root   33 Oct  7 10:34 vmlinuz -> vmlinuz-3.11.3-1.gb1e6ec5-desktop
> -rw-r--r--  1 root root 5.1M Oct  3 20:33 vmlinuz-3.11.3-1.gb1e6ec5-desktop
> -rw-r--r--  1 root root 4.8M Jun  7 12:48 vmlinuz-3.7.10-1.16-desktop

Is there anything I can safely get rid of? I haven’t had this problem
before during 3 or 4 previous upgrades on this machine. Will it be safe
to reboot? Or will it try to use the new kernel which hasn’t installed
correctly?

No, reboot is not safe at all.

Uninstall plymouth and try again


su -
mkinitrd


Cheers / Saludos,

Carlos E. R.
(from 11.4, with Evergreen, x86_64 “Celadon” (Minas Tirith))