Screen frozen after zypper update

Hello,

after a zypper update the windowmanager seems to quit silently. System hangs with “Reached target Graphical Interface”. With Alt-F i can log in and start as root icewm.

#lspci | grep VGA

gives:

00:02.0 VGA compatible controller: Intel Corporation 2nd Generation Core Processor Family Integrated Graphics Controller (rev 09)

What is going wrong, i found some hints in the net, often the solution was to install a nvidia-driver but i have an integrated graphic controller and it seems as if all needed driver are installed.

regards

I have opensuse 13.1 with Linux hname0012 3.18.0-11-desktop.

Yast is not running either, it just doesnt start with no error message or whatever.

You should only install the nvidia driver if you have an nvidia graphics card obviously. Otherwise you will break your system, in particular the intel driver.

Please do a normal boot, when you drop to text mode, reboot and select “Recovery mode” in the boot menu (located in “Advanced Options”).
If graphics mode works then, please post the file /var/log/Xorg.0.log.old .

Apparently you don’t have a problem with the graphics driver though, as you can start X manually.
So probably it’s the displaymanager (login screen) that crashes.
Which one are you using? (kdm, gdm, …)
What does “systemctl status xdm” say? (run it in text mode after a failed boot)

Hello wolfi,

no i do not load the nvidia driver, i only mentioned that as an example.

You are right, it seems, there is a problem with the displaymanger, maybe it doesnt know, where to find it because of new config files or so.

The output of
#systemctl status xdm:

xdm.service - LSB: X Display Manager
Loaded: loaded (/etc/init.d/xdm)
Active: inactive (dead)

Feb 10 14:08:57 as501755 systemd[1]: Dependency failed for LSB: X Display Manager.

regards
and thank you

Yes, I just wanted to make sure that you do not try this… :wink:

You are right, it seems, there is a problem with the displaymanger, maybe it doesnt know, where to find it because of new config files or so.

Well, what displaymanager are you using?
If it’s kdm, the log file is /var/log/kdm.log e.g., this should hopefully contain more information why it is crashing.

Did you try recovery mode? If it works there, it should not be an installation problem.

You can switch to xdm as a test, this should probably still work. Just change the line DISPLAYMANAGER=“xxx” in /etc/sysconfig/displaymanager (with a text editor) to:

DISPLAYMANAGER="xdm"

Do you get to graphical mode then?

As you say that this started after an update, the update probably wasn’t installed completely leaving the system in an inconsistent state, or you have some bad repositories.
So try to run “zypper up” again, and please post your repo list:

zypper lr -d

PS: I noticed now that you are using kernel-3.18.0. Any particular reason for that?
Normally it’s better to just stay with the kernel included in the distribution.
And 3.18.0 is old already anyway, there have been 5 bugfix releases meanwhile.

Well, what displaymanager are you using?

Well, frankly spoken, i actually didnt care much about that since i had to use a standard openSuse Installation and installed it all automaticly. In the /etc/sysconfig/displaymanager
there stood originally Xorg. and so is the Xorg.0.log

If it’s kdm, the log file is /var/log/kdm.log e.g., this should hopefully contain more information why it is crashing.

I do not find something substantiall in the Xorg.0.log

Did you try recovery mode? If it works there, it should not be an installation problem.

Because of your advice, i tried the recovery mode, but that was also with no success.

You can switch to xdm as a test, this should probably still work. Just change the line DISPLAYMANAGER=“xxx” in /etc/sysconfig/displaymanager (with a text editor) to:

DISPLAYMANAGER="xdm"

Do you get to graphical mode then?

i tried it already, but no luck :frowning:

As you say that this started after an update, the update probably wasn’t installed completely leaving the system in an inconsistent state, or you have some bad repositories.
So try to run “zypper up” again, and please post your repo list:

zypper lr -d

| Alias | Name | Enabled | Refresh | Priority | Type | URI | Service

—±--------------------------±-----------------------------------±--------±--------±---------±-------±-----------------------------------------------------------------------------------------------±-------
1 | openSUSE-13.1-1.10 | openSUSE-13.1-1.10 | No | No | 99 | yast2 | hd:///?device=/dev/disk/by-id/usb-Kingston_DataTraveler_SE9_0019E06B07EEBCA137410006-0:0-part2 |
2 | openSUSE:13.2 | openSUSE:13.2 | Yes | Yes | 99 | yast2 | http://download.opensuse.org/distribution/13.2/repo/oss/ |
3 | repo-debug | openSUSE-13.1-Debug | Yes | No | 99 | yast2 | http://download.opensuse.org/debug/distribution/13.1/repo/oss/ |
4 | repo-debug-update | openSUSE-13.1-Update-Debug | Yes | No | 99 | rpm-md | http://download.opensuse.org/debug/update/13.1/ |
5 | repo-debug-update-non-oss | openSUSE-13.1-Update-Debug-Non-Oss | Yes | No | 99 | rpm-md | http://download.opensuse.org/debug/update/13.1-non-oss/ |
6 | repo-non-oss | openSUSE-13.1-Non-Oss | Yes | Yes | 99 | yast2 | http://download.opensuse.org/distribution/13.1/repo/non-oss/ |
7 | repo-oss | openSUSE-13.1-Oss | Yes | Yes | 99 | yast2 | http://download.opensuse.org/distribution/13.1/repo/oss/ |
8 | repo-source | openSUSE-13.1-Source | Yes | Yes | 99 | yast2 | http://download.opensuse.org/source/distribution/13.1/repo/oss/ |
9 | repo-update | openSUSE-13.1-Update | Yes | Yes | 99 | rpm-md | http://download.opensuse.org/update/13.1/ |
10 | repo-update-non-oss | openSUSE-13.1-Update-Non-Oss | Yes | Yes | 99 | rpm-md | http://download.opensuse.org/update/13.1-non-oss/ |
11 | rubyrepo | rubyrepo | Yes | No | 90 | rpm-md | http://download.opensuse.org/repositories/devel:/languages:/ruby/openSUSE_13.1/ |

PS: I noticed now that you are using kernel-3.18.0. Any particular reason for that?

No, i had to change it sometime, but for other reasons, now i could take a newer one if this would help.

Normally it’s better to just stay with the kernel included in the distribution.
And 3.18.0 is old already anyway, there have been 5 bugfix releases meanwhile.

Ok, i try to find a newer one.

(But i do not understand why this happens without some error messages)

Many thanks!

Ok, but what desktop did you install/are you using? KDE, GNOME, …

If you installed from the full DVD or the NETinstall CD, you probably are using KDE and kdm, as this is the default.

In the /etc/sysconfig/displaymanager
there stood originally Xorg. and so is the Xorg.0.log

The DISPLAYMANAGER line definitely should not have “Xorg”. It should be set to “kdm” or “gdm” most likely.

I do not find something substantiall in the Xorg.0.log

Of course. If the displaymanager crashes, this will not be logged to Xorg’s log. As I said, for kdm the log is /var/log/kdm.log, so please post this.

Because of your advice, i tried the recovery mode, but that was also with no success.

So there’s something wrong with the displaymanager in general.

i tried it already, but no luck :frowning:

As you say that this started after an update, the update probably wasn’t installed completely leaving the system in an inconsistent state, or you have some bad repositories.
So try to run “zypper up” again, and please post your repo list:

zypper lr -d

You have a mixture of 13.1 and 13.2 there, this is definitely calling for trouble.

I suppose you added the 13.2 repo by mistake?
So remove it and run “zypper dup”, that should fix your system:

sudo zypper rr 2
sudo zypper dup

If you wanted to upgrade to 13.2, you should remove all 13.1 repos, add the other 13.2 repos (update, non-oss, and non-oss-update at least) and run “zypper dup” to fully upgrade. See also here: SDB:System upgrade - openSUSE Wiki

No, i had to change it sometime, but for other reasons, now i could take a newer one if this would help.

Well, it won’t help for this particular problem, I suppose.

Ok, i try to find a newer one.

(But i do not understand why this happens without some error messages)

Many thanks![/QUOTE]

Hello wolfi,

because of the discussion with you, i checked some system configs and i saw some wrong links in /boot, which i corrected. After that i booted and now the displaymanager seems to start the icewm. I don´t understand this, because in the displaymanager i configured:

DISPLAYMANAGER="Xorg

on the other hand, in the .xinitrc of root, there is :

/usr/bin/icewm
and it seems that linux prefers this.

Maybe it needs some more configuration, but at least, i have now a graphical login.

Thank you, i appreciate your help very much!

Probably “initrd” and “vmlinuz”? Those are not used/necessary normally. But they should be created correctly when you install/remove kernels via RPM packages.

After that i booted and now the displaymanager seems to start the icewm. I don´t understand this, because in the displaymanager i configured:

DISPLAYMANAGER="Xorg

Again, this is INVALID!
It should be DISPLAYMANAGER=“xdm”, DISPLAYMANAGER=“kdm”, or similar.

As I still don’t know whether you use KDE or GNOME (or anything else), I can only guess.
Try to set it to “kdm”, if kdm is not installed (i.e. if you are using GNOME), you should get xdm as fallback anyway.
If kdm does not work at all, there’s still something wrong with your installation. Set it to “xdm” then, and ask back.

on the other hand, in the .xinitrc of root, there is : and it seems that linux prefers this.

If that is in a user’s .xinitrc, the mentioned command is run at login for that user. In this case, icewm is started when you login as root.
You should not do this, better remove that file. You can select the desktop on the login screen anyway.

Maybe it needs some more configuration, but at least, i have now a graphical login.

Yeah, but apparently you haven’t solved your problem yet.
Again, remove the 13.2 repo and run “zypper dup”, set back DISPLAYMANAGER to kdm, and it will work then again hopefully.

Once again:

i made the

zypper dup

but now i am not able to reboot anymore:

#reboot

Failed to open /dev/initctl: No such device or address
Failed to talk to init daemon.

How did you boot in the first place?

Have you removed the 13.2 repo first?
Did “zypper dup” install anything?

Press your “Reset” button, or force a shutdown by pressing the power button.

After a hard reset

#sync
#reboot

it finds “reboot” again.

Is it a secret what desktop you use???

Really it is a simple question and helps us help you.

You appear to be making random changes this seldom works stop take the advice we give you and answer the questions we ask

AS mentioned in my last post, icewm is now running (although the configuration seens to be untransparent).

Gogalthorp,

i never cared about the desktop, when i installed linux, i needed only terminals, when i had to use a whole distribution, i clicked on “KDE”, but i can´t say, which desktop is running now on my system. How can i find out? Since there are many “kdm” processes, maybe kdm? I dont know. Now icewm is running, i have to spend some time to investigate the desktop. But i am looking for something leaner.

regards

If you selected “KDE”, kdm should be installed and set up as displaymanager (login screen).
So /var/log/kdm.log should (hopefully) contain information about the crash.

But what is your /etc/sysconfig/displaymanager saying now? Please post the whole file.
If DISPLAYMANAGER is set to "Xorg (which is invalid as I wrote already) or “xdm”, i.e. xdm is used as displaymanager, there should not be any kdm processes running.
Or is kdm working now?
You should be able to tell by the looks of the login screen. If in doubt, maybe post a picture.

Btw, as I wrote already, you should not start the desired windowmanager (desktop environment) via ~/.xinitrc.
You can set the default one in /etc/sysconfig/windowmanager (DEFAULT_WM). xdm starts that one, other login managers allow you to choose one per user/login anyway.

Now icewm is running, i have to spend some time to investigate the desktop. But i am looking for something leaner.

What do you mean with “something leaner”?
IceWM is as lean as you can get I suppose.

Other light-weight alternatives would be: XFCE, LXDE, and LXQT.
You should find patterns to install them in YaST->Software Management (View->Patterns).
To start any of them, you have to remove your ~/.xinitrc and then either set it in /etc/sysconfig/windowmanager as default or choose it at the login screen. A (lightweight) displaymanager alternative to kdm would be lightdm or lxdm, but this might even be installed by those patterns (I’m not sure).

If you mean something more user-friendly, we might also try to get KDE running (which is independent of kdm btw, you can run it with any displaymanager). Maybe it does work already anyway.
But again, the first step would be to remove ~/.xinitrc.

Hello,

i´ve deleted now ~/.xinitrc and in the /etc/sysconfig/displaymanager i put:

DISPLAYMANAGER=“lxdm”

My company want´s now that i am productive again, so i will keep this solution because i do not have more time playing around, at least i have a window/display/whatever manager now.

Thank you very much for your help!

regards

I hope you actually installed lxdm, otherwise it won’t be different from putting “xdm” there… :wink:

My company want´s now that i am productive again, so i will keep this solution because i do not have more time playing around, at least i have a window/display/whatever manager now.

Ok.
If you want to pursue this further at one point, feel free to ask again if you have questions.