Can't get into the live session (openSUSE 12.3 KDE)

Hello guys, first post here.

I’m trying to install openSUSE 12.3 KDE in my Samsung Notebook. I have one Windows 8 installation there. I think this laptop doesn’t have UEFI, because I bought it with a Windows 7 installation; I only installed Win 8 later. Also, my BIOS has a ugly style (classic blue background), not a beautiful / mouse-supported one.

Okay, so my problem is that I can’t even get into the live session in the notebook. I am trying to boot by a USB Flash Disk, and I know it is working because I installed openSUSE 12.3 in another computer (a desktop one) a few days ago. But when I try to boot from this USB Disk, I get into the screen options (installation; log into the live session; test RAM; etc) and select live cd or installation, whatever. After that, the system begins to load itself, I can see the default openSUSE wallpaper loading up but, after a while, the system freezes there. Sometimes if freezes in the wallpaper, sometimes it freezes with a blank screen.

I’ve already tried several options, like loading the kernel in safe / secure mode (something like that), trying to add a “nomodeset” parameter, trying to change the resolution. But I can’t get into the live session.

I searched so much in google but I can’t find a solution there (neither here in the forums). It’s easy to find several people who can’t boot in the live session, but I think their problems are other ones.

My Processor is AMD, and I think I have a nvidia driver. The openSUSE version is 64 bits (my notebook supports it; my Windows 8 is 64 bits).

A little curiosity: I’ve tested live CDs of other distros, and I can’t boot into them either. I tried CrunchBang, Ubuntu, Xubuntu, Fedora. Neither of them will work.

I think the problem has something to do either with Windows 8 blocking my acess or with the graphical drivers of this laptop.

Does somebody here know anything related to that? Thanks in advance.

There is this issue but it doesn’t seem to be related to yours
Windows 8 Hybrid Boot

On 05/26/2013 06:26 PM, thiagowfx wrote:
> when I try to boot from this USB Disk, I get
> into the screen options (installation; log into the live session; test
> RAM; etc) and select live cd or installation, whatever.

going on the above plus “I’ve tested live CDs of other distros, and I
can’t boot into them either.” and conclude:

if you get to that screen you describe the first thing you should do
is select “Check Installation Media” it should run a program and
return some information to you, like the media is good or bad.

what does it say?

if it says the media is good (or if it says NO errors) then on the
next boot select “Memory Test” and let it run 12 hours at least! just
walk away from it…or if you wanna glance at it every once in a
while you can, because any error is too many!! if your RAM errors
out, let us know there is more help we can give you…

if both the media and the RAM checks out good then give us more info:
model number of the machine…and, go into the BIOS and confirm that
it does or does not have UEFI (some win7 machines did have that)…


dd
openSUSE®, the “German Engineered Automobile” of operating systems!
http://goo.gl/PUjnL
http://tinyurl.com/DD-Caveat
http://tinyurl.com/DD-Hardware
http://tinyurl.com/DD-Software

If it is at all like my experience, it says:

“Insert CD-ROM or DVD and press enter”

And then, if I press enter, it says:

“No CD-ROM or DVD found.”

In short, the “Check Installation Media” does not work when the installation media is a USB.

I’m sending you a video with the issue: openSUSE 12.3 KDE can’t boot into live session - YouTube (recorded right now)

In the BIOS I can find one option named “UEFI Boot Support”, which is currently disabled now. (BTW, when I installed Windows 8, this was disabled too.)
Here is one screenshot: SUSE Paste

I’m not sure what the model of my machine is. Is there an easy way to discover that? Now I can only tell you that is a Samsung laptop with an AMD A4-3305M APU 1.90GHz Processor, and BIOS and MICOM version is 03QU (information extracted from the BIOS SysInfo).

On that opening screen, hit F3. I think one of the options will be “NoKMS”. Select that, and see what happens.

I think you are having video problems. The NoKMS will give you a different video driver of poorer quality. If that works, it will confirm that problem is a video driver freezing.

Usually there is a label on the bottom of the laptop that has the Model #.

Also, if a Windows variant will boot, usually it has a System Information application that will provide in formation on the Graphic hardware details.

So, with NoMKS my system will freeze too. Video:

//youtu.be/ZkV69f5T3ek

But this time I can log into a shell by pressing Ctrl+Alt+F1. Is there something useful I could to do there? I only tried to execute xterm: SUSE Paste , which gave me an error.

In the shell, you could find out precisely what graphics are installed by typing:


/sbin/lspci -nnk | grep VGA -A2

Look to see what hard drives you have by typing:


su -c 'fdisk -l'

and press < enter > for live session password.
(you will need to know the drives later when trying to mount a usb stick). Likely you only have /dev/sda

Then mount a USB stick manually and copy some log files. To mount the USB stick create a mount directory in the /home/linux directory with something like:


mkdir usbstick

and then plug in the USB stick and check to see its label by typing


su -c 'fdisk -l'

and look for the additional drive. Likely /dev/sdb

assuming it is /dev/sdb, then mount it with something like:


mount -t vfat -o rw,users,uid=linux /dev/sdb1 /home/linux/usbstick

… I’m assuming ‘linux’ is the liveCD/DVD username

and everything yousee in /home/linux/usbstick will be the memory stock content. Then change to that directory with:


cd /home/linux/usbstick

and copy some files to it:


dmesg > dmesg-copy.txt
cp /var/log/Xorg.0.log /home/linux/usbstick/Xorg.0.log
su -c 'cp /var/log/messages /home/linux/usbstick/messages.txt

then


cd ..

and umount the usbstick


umount /dev/sdb

and remove the usb stick. Then you have some log information you can post on SUSE Paste and share the urls here, which may explain why the boot problem.

Thanks for the support, oldcpu! So, let’s go:

dmesg:
SUSE Paste

log:
SUSE Paste

Explanation:
sdb is my USB Flash Drive and
sda is my HDD. I have Windows 8 and Crunchbang Linux installed there (but I can’t boot into Crunchbang, it was only a test I made some time ago)

messages:
SUSE Paste

Xorg.0.log:
SUSE Paste

> In short, the “Check Installation Media” does not work when the
> installation media is a USB.

really! how rude is that??


dd

I note the video hardware is


00:01.0 VGA compatible controller [0300]: Advanced Micro Devices [AMD] nee ATI Sumo **Radeon HD 6480G**] [1002:9649]
        Subsystem: Samsung Electronics Co Ltd Device [144d:c624]

My understanding is with ‘nomodeset’ specified, the PC should boot to either the VESA driver or the FBDEV driver … but it appears to be failing with a variant of fbdev ? I note from the Xorg.0.log


    17.648] (WW) Falling back to old probe method for fbdev
    17.648] (II) Loading sub module "fbdevhw"
    17.648] (II) LoadModule: "fbdevhw"
    17.648] (II) Loading /usr/lib64/xorg/modules/libfbdevhw.so
    17.671] (II) Module fbdevhw: vendor="X.Org Foundation"
    17.671]    compiled for 1.13.2, module version = 0.0.2
    17.671]    ABI class: X.Org Video Driver, version 13.1
    17.671] (EE) open /dev/fb0: No such file or directory
    17.671] (WW) Falling back to old probe method for vesa
    17.671] (EE) Screen 0 deleted because of no matching config section.
    17.671] (II) UnloadModule: "radeon"
    17.671] (EE) Screen 0 deleted because of no matching config section.
    17.671] (II) UnloadModule: "radeon"
    17.671] (EE) Screen 0 deleted because of no matching config section.
    17.671] (II) UnloadModule: "radeon"
    17.671] (EE) Screen 0 deleted because of no matching config section.
    17.671] (II) UnloadModule: "radeon"
    17.673] (II) modesetting(0): Creating default Display subsection in Screen section
        "Default Screen Section" for depth/fbbpp 24/32
    17.673] (==) modesetting(0): Depth 24, (--) framebuffer bpp 32
    17.680] (==) modesetting(0): RGB weight 888
    17.680] (==) modesetting(0): Default visual is TrueColor
    17.680] (II) modesetting(0): ShadowFB: preferred NO, enabled NO
    17.680] (EE) modesetting(0): KMS doesn't support dumb interface
    17.680] (EE) modesetting(0): KMS setup failed
    17.680] (II) UnloadModule: "modesetting"
    17.680] (II) UnloadModule: "radeon"
    17.680] (EE) Screen(s) found, but none have a usable configuration.
    17.680]
Fatal server error:
    17.680] no screens found
    17.680] (EE)
Please consult the The X.Org Foundation support
         at http://wiki.x.org
 for help.
    17.680] (EE) Please also check the log file at "/var/log/Xorg.0.log" for additional information.
    17.680] (EE)
    17.686] Server terminated with error (1). Closing log file.

I don’t know why that is happening.

One thing you could try, is instead of nomodeset, try the boot code " radeon.modeset=1 " (that is rather speculative on my part).

Another thing to try, with and without nomodeset, is to remove plymouth interference from the equation, and try the boot code " plymouth.enable=0 " (that is also rather speculative on my part)

There is a lot of information in the dmesg and messages that I have not yet been able to digest nor point a finger at any specific issue.

Instead of the above, I stumbled across a thread for a different distro where they recommended the following two boot codes:


nomodeset acpi=off

I note in the dmesg, a massive number of messages associated with acpi


    0.000000] ACPI BIOS Bug: Warning: Optional FADT field Pm2ControlBlock has zero address or length: 0x0000000000000000/0x1 (20120913/tbfadt-598)
........
    0.211144] [Firmware Bug]: ACPI: No _BQC method, cannot determine initial brightness
    0.211329] [Firmware Bug]: ACPI: No _BQC method, cannot determine initial brightness
.....
    0.246377] PCI: Using host bridge windows from ACPI; if necessary, use "pci=nocrs" and report a bug
    0.246639] ACPI: PCI Root Bridge [PCI0] (domain 0000 [bus 00-ff])
.......
    0.254444]  pci0000:00: ACPI _OSC support notification failed, disabling PCIe ASPM
    0.254447]  pci0000:00: Unable to request _OSC control (_OSC support mask: 0x08)
..........
many entries of 
    0.265706] ACPI: PCI Interrupt Link [LN24] (IRQs *24)
.......
    3.487590] [Firmware Bug]: ACPI: No _BQC method, cannot determine initial brightness 

and the /var/log/messages more of the same:



2013-05-26T16:09:04.207506+00:00 linux kernel:     0.211144] [Firmware Bug]: ACPI: No _BQC method, cannot determine initial brightness
2013-05-26T16:09:04.207506+00:00 linux kernel:     0.211329] [Firmware Bug]: ACPI: No _BQC method, cannot determine initial brightness
.......
2013-05-26T16:09:04.207509+00:00 linux kernel:     0.246377] PCI: Using host bridge windows from ACPI; if necessary, use "pci=nocrs" and report a bug
.......
2013-05-26T16:09:04.207767+00:00 linux kernel:     0.254444]  pci0000:00: ACPI _OSC support notification failed, disabling PCIe ASPM
2013-05-26T16:09:04.207768+00:00 linux kernel:     0.254447]  pci0000:00: Unable to request _OSC control (_OSC support mask: 0x08)
.........
2013-05-26T16:09:04.208851+00:00 linux kernel:     3.487590] [Firmware Bug]: ACPI: No _BQC method, cannot determine initial brightness

Booting a laptop with ‘acpi=off’ is not pleasant and not IMHO I good long term solution…

OK, I’ll try that.

I understand almost nothing about hardware / kernel logs at this time, so I don’t know what exactly is happening behind the scenes.

Neither of these two options worked. I got the same previous problems (freezing blank screen).

Did you try both “nomodeset” and “acpi=off” at the same time ?

Possibly try all three “nomodeset” and “acpi=off” and "plymouth.enable=0 " all at the same time.

The ideas being:

  • acpi=off is to work around the " [Firmware Bug]: ACPI: No _BQC method, cannot determine initial brightness " problem that clearly will stop successful booting. I think there is also a ‘kernel’ selection in the grub boot menu which does the same as the ‘acpi=off’ boot code. Look for it. Apply it.
  • nomodest is to work around a possible kMS compatibility the Radeon HD 6480G hardware. There is also a grub boot menu “graphic” selection of ‘NoKMS’ which does the same. Look for it). The idea here is
  • plymouth.enable=0 is to work around a possible incompatibility with the plymouth splash screen. Alternatively press <escape> shortly after the grub menu disappears (ie when the SuSE splash ‘wall paper’ appears) which should give text scrolling by during the boot and it should do the same as that plymouth.enable=0 boot code [you ‘may’ also see the LAST error message that way, if the system hangs].

Also make certain you followed the advice given wrt checking the md5sum of the downloaded .iso file against the md5sum posted on the website from whence the file was downloaded. USB or not, it needs to be checked. While a Torrent download improves the possibility of a good downloaded, I have seen cases when an .iso downloaded via torrent failed the md5sum check.

Yeah, I tried that. Two times:

  • nomodeset acpi=off
  • radeon.modeset=1 plymouth.enable=0

Tried that, no luck too: SUSE Paste . I’ve pressed ESC, like you said, right after the GRUB menu.
This time I got directly booted into a terminal. But the terminal was not a separate shell (aka tty1, tty2), but it was right after (below) the error messages, prompting me for a login. But the problem was that I couldn’t write in it, it was like the keyboard was disabled.

I think the checksum isn’t the problem because I’m writing here right from openSUSE from my desktop PC, which I installed by the same USB Stick.

This thing is getting hard to solve. Do you have any more ideas?

If you installed OK from it, the checksum should be OK.

This thing is getting hard to solve. Do you have any more ideas?

Don’t give up, it will work.
If you try to start a text based install (as long as you don’t actually start the install, nothing happens to your machine.