Well, title says it already. I can login in a console. “dmesg” throws no errors. If it’s of any concern, I use the same /home for tumbleweed, leap 42.1 and 13.2. The latter two boot okay. Any ideas what may be the cause?
What video card?? TW from 6 months back would need a bunch of updates I’d guess
The video card is a MSI with a Tobago Pro Radeon R7 360. Functioned well so far.And yes, it were about 1900+ updates. Loaded them down and installed all well, but … no graphical login anymore.
In fact I think it is sddm not starting. How could I check?
Try setting DISPLAYMANAGER=“xdm” in /etc/sysconfig/displaymanager.
And if you are using fglrx, that doesn’t work with the latest Xorg in Tumbleweed, so that would explain your problem.
sddm requires working OpenGL/GLX, if you have fglrx installed (and not working), it cannot start.
You would have the same problem with gdm in that case, btw, and Plasma5 or GNOME would not work either.
So if you are really using fglrx, either uninstall it or downgrade Xorg to the Leap version as recommended by Sebastian Siebert:
https://www.sebastian-siebert.de/2015/12/05/opensuse-proprietaeren-grafik-treiber-amd-catalyst-15-11-als-rpm-installieren/#comment-4921
http://lists.opensuse.org/opensuse-factory/2016-01/msg00366.html
Setting DISPLAYMANAGER=“xdm” in /etc/sysconfig/displaymanager didn’t help. I previously had issued from command line (as root) xdm (as well as kdm) with “command not found” both times as results, so this is no real surprise.I don’t use fglrx but radeon, which seems to be loaded according to “lsmod: radeon 1605632 1”.Running startx as root I get:xauth: file /root/.serverauth.1998 does not existX: error while loading shared libraries: /usr/lib64/libfreetype.so.6 file too shortxinit: giving upxinit: unable to connect to the X-server: connection refusedxinit: failed. /usr/bin/Xorg is not setuid. Maybe that’s the reason?If so, either use a display manager (strongly recommended) or adjust /etc/permissions.local and run chkstat --system --set afterwards.That of any help?
Sorry, I don’t know why it mangled the post this way. And it does not accept editing (Firefox on SUSE Leasp42.1…).
Startx as user has not worked for ages since it was decided it was a security hole. Use init 5 as root to start the dm.
You set the dm in yast etc-config editor - Desktop-display manager - DISPLAYMANAGER
I tried it as root. I can try init 5 in the evening.
Yes, you do that, when you got into the graphical mode, but YAST without X? I did it with vi…
You cannot run “xdm” or “kdm” like that.
Here’s your problem:
X: error while loading shared libraries: /usr/lib64/libfreetype.so.6 file too short
It seems some files on your hard disk are corrupted, in particular /usr/lib64/libfreetype.so.6.
So reinstall libfreetype and hope that there are not more broken files…
sudo zypper in -f libfreetype6
YaST works in text mode too.
This seems to be a dead end – and THE end: zypper verify: error while loading shared libraries /usr/lib64/liblber-2.4.so.2 file too short
I guess, I’d better download a newer version and re-install. I can’t see how to fix that one…
That’s no deadend, but more files seem to be broken. I would have expected that actually.
Reinstall libldap-2_4-2 too, and so on until all files are fixed.
You might run “rpm -Va” to verify all installed files/packages, but this will probably take a while and will also give a lot of false positives.
I guess, I’d better download a newer version and re-install. I can’t see how to fix that one…
Actually you’d just need to reinstall all packages in the worst case. You can do that with YaST in text mode (go to “View”->Repositories, select “@System” and do “All in this list”->"Update Unconditionally), maybe a “zypper dup” to fully update your system would help already as well.
Yeah. Result:
Internal error: Please report a bug with logs.
Details: component cannot import namespace ‘Pkg’
Caller: /usr/lib64/ruby/vendor_ruby/2.2.0/yast/yast.rb:162: in ‘import_pure’
My words, exactly ;-). I’ll re-install now.
So some ruby packages/files seem to be broken as well.
My words, exactly ;-). I’ll re-install now.
Well, that should definitely help, but where’s the sport in that? 
Anyway, this definitely looks like it’s been caused by some filesystem corruption. Maybe just a crash/unclean shutdown after installing updates, but might also indicate a dying harddrive.
Sport is all fine with me – for a time. But I need that box. I do home banking, mail, … on it and there comes a point in time, when playing around simply doesn’t fit into the schedule anymore. I don’t have windows for work and play around with linux in my leisure time. I’m a linux only guy, which means things have to work at some point in time. Thus – re-install.
This was meant as a joke…
If using btrfs (default since 13.2), it might be easier to rollback to a previous snapshot though than doing a fresh install.
Although an “Upgrade” from the DVD (or NET-install CD) should be a good choice too and preserve all settings and installed packages.
Okayokay ;).
There were (probably no usably(?)) snapshots yet. Install - update - crash. But I admit, I didn’t look. New install and updates work by the way … no settings to save on a new install and the old /home stayed untouched anyway. But, as mentioned before, thanks for your help.