GUI "Authentication failed" 11.0

I am on openSUSE 11.0 with kernel 2.6.27.7-9-pae

System has been working well for many months. Today I boot and attempts to login through the GUI as any user (including root) fail, presenting the message:

Authentication failed

However, the machine is running. If I switch to a console using Alt-F1 I can login as a user and su to root. Also, samba is working and a Windows machine can connect to shares as it has for some time.

I have been looking around the net for some way that the GUI might have been configured for a different authentication, such as LDAP, that I am not running. There are many places to find the GUI tools described, but I cannot find one that describes command line configuration.

Do you think this authentication reconfiguration might be the problem? If so, how can I tell and how can I change it?

What has changed most recently is that yesterday I installed the Google Chrome browser. Appeared to install correctly and was working.

Any ideas or suggestions? Thanks.

If you login as user can you then do: startx

?

When I run ‘startx’ there is a message…

Fatal server error:
Server is already active for display 0
    If this server is no longer running, remoe /tmp/.X0-lock
    and start again.

Since the GUI login window is still up, I am assuming that this is the X instance that is running. Is that correct?

That is telling you that you already have a X session logged in
Do you have auto login enabled? If you do:
Reboot to level 3 and start yast and disable auto-login
Boot to Level 3, then Yast and More… - openSUSE Forums

Or reboot to level 3, login as user, then do: init 5
then: startx

Many thanks for the detailed description to get into yast. Reminds me of the 24x80 days.

Auto Login was -NOT- enabled.

Using ‘ps -ef’ shows that /usr/bin/X is running as root on tty7.

Any other ideas or suggestions?

Is there any hope for resolution on this? Any ideas or suggestions?

I’m out of ideas really.
You know 11.0 support ends very soon?

How many user accounts do you have? (You should never login as root BTW)

Is there anyone else in the forum that might have a suggestion?

No, I did not know 11.0 support ends soon.

There are 8-10 user accounts. Probably 20+ application accounts. Yes, I know not to login as root. That’s why I said that I logged in as a user, then did an su to root.

At this point, I would login as anything that worked.

If nothing can be done, I might give Ubuntu a spin, but I would rather be using openSUSE.

If you would rather be using openSUSE - why use Ub*?

Support ends June 30th for 11.0

That’s a heck of a lot of user accounts:O
IMO: Time to backup what you really need

Format it all

Install 11.2

pwatson wrote:
> If nothing can be done, I might give Ubuntu a spin

  1. what happens if you, at the first green screen select something
    other than the default, like maybe “failsafe” or one of the
    non-default kernels?

  2. what happens if you, at the first green screen type 3 (which will
    appear then on the line for kernel parameters) and then press the
    “enter” key? it should dump you at a command line in runlevel 3 (no X
    running anywhere), where you might be able to log in, issue a init 5
    and be presented with a GUI login screen…

or, at the runlevel 3 screen log in as yourself and give startx (what
happens?)

now, we all know that mommy’s vase didn’t jump off of the table by
itself while while little Paula was not touching it…so, tell us
what all happened between (post one’s) “System has been working well
for many months.” and “Today I boot and…”

that is, what config files might have been changed? what updates were
applied? what was installed? what was ‘tuned’, ‘improved’,
‘customized’ and/or etc? who else had access and could be the one who
“didn’t do nothing” it just jumped off the table?

do you know the date of your “Today” in post one of this thread? can
you search /var/log/messages (and others) for evidence that somewhere
between “many months” and “Today” someone did something as root,
somewhere??

and, once know how the error was introduced into the system: is it
possible to back out, and get back to the workig system?

if not, why not just go ahead and restore the system from the backup
nearest the date of “Today”?

or maybe it did jump off the table by itself: that is, maybe you
have some corrupt RAM that wrote a mistake to your hard drive and THAT
is the change that occurred! or, maybe your hard disk is flaky and a
few bits in exactly the wrong place flopped from one to zero, or vice
versa…

in any event, since there are hundreds of thousands of openSUSE users
who have not had your unusual problem, you will get no relief by
switching to Ubuntu…because there you will find the same hardware
problem and the same keyboard operator not doing nothing to cause
problems…


DenverD (Linux Counter 282315)
CAVEAT: http://is.gd/bpoMD
posted via NNTP w/TBird 2.0.0.23 | KDE 3.5.7 | openSUSE 10.3
2.6.22.19-0.4-default SMP i686
AMD Athlon 1 GB RAM | GeForce FX 5500 | ASRock K8Upgrade-760GX |
CMedia 9761 AC’97 Audio

Log in as a user,


su -c 'init 3'
startx

Try logging in, using a different keyboard.

Did you update recently? KDE/Gnome version?

It may be a small number of system accounts, but that is not directly indicative of the significant work done using those accounts.

I already have a backup of the user files.

I am going to run through DenverD’s list, even the redundant ones.

  1. Selecting Failsafe or Xen produces the same failure mode.

  2. Booting to runlevel 3, login as user, su - root, init 5 does present a GUI login screen. However, it also presents the same failure mode.

Booting to runlevel 3 provides a prompt where I am able to login. (Same as I reported before.) Running ‘startx’ produces a gray (dithered) screen with an X cursor, but nothing else. No icons, no right-click menu, nothing.

I mentioned earlier that I installed Google Chrome on the system during operation prior to the failure. No one else who knows the root password has accessed the machine.

I do have a backup of the most significant files on the system. I trust that you are not suggesting that my not having a recent system backup that I would want to restore is the cause of this failure.

The condescending tone and childish treatment presented here are not making me think that openSUSE is where I should be. You can say all day long that is not what was intended, but it is what I got out of it.

Did you try removing chrome??

You can run Yast from the command line in terminal mode. Log as root or become root with su - command

type
yast

Navigate with TAB key and select with space

Thanks for your reply. This does work. I could get into yast and remove the google-chrome-* items. When I did so, it wanted to update three other kernel-related items.

Well, it is good that I have already grabbed the significant files. Now the machine will not boot at all. /dev/sda1 not found

Thanks to everyone for their help. I believe my work here is complete.