After upgrading to Lucid, xorg exits after a couple of minutes

Forgot to mention that I'm using openbox as a window manager, no compiz,
and the session is handled by xfce4-session (exec'd via .xsession).

Please run apport-collect 560899 after X dies and you start it again to attach your logs to this bug automatically so we can help you with your problem.

apport information

apport information

On Mon, Apr 12, 2010 at 12:25:02AM -0000, Robert Hooker wrote:
> Please run apport-collect 560899 after X dies and you start it again to
> attach your logs to this bug automatically so we can help you with your
> problem.

Just done that.

This was the output:

 $ apport-collect 560899
 The authorization page:
    (https://edge.launchpad.net/+authorize-token?oauth_token=REMOVED&allow_permission=WRITE_PRIVATE)
 should be opening in your browser. After you have authorized
 this program to access Launchpad on your behalf you should come
 back here and press <Enter> to finish the authentication process.

 hook /usr/share/apport/package-hooks//source_xserver-xorg-video-intel.py crashed:
 Traceback (most recent call last):
   File "/usr/lib/python2.6/dist-packages/apport/report.py", line 636, in add_hooks_info
     symb['add_info'](self, ui)
   File "/usr/share/apport/package-hooks//source_xserver-xorg-video-intel.py", line 63, in add_info
     report['UnreportableReason'] = _('The running kernel is not an Ubuntu kernel')
 NameError: global name '_' is not defined

(the oauth_token was removed just in case, I don't know if it's supposed to be
public, or the implications of other people seeing that. Let me know if you
need it)

By the way, given that it looks as if it wants to complain my kernel is not
Ubuntu's, I'm running a stock, non-patched 2.6.33.1 built with the same config
I was using in 9.04. It should work, shouldn't it?

If you need the .config or anything else, I'd be happy to provide it.

It just happened again, as I was typing this, three times in a short
succession.

Two of those three, vim was not open and I was typing on irsii over a screen
over ssh.

At least two of those times it happened the moment I pressed enter (or was
quite an incredible coincidence). Now that I think of it, maybe the other
times it also happened on enter, but I can't be sure.

The application never received the enter keypress (as it was running in a
screen, it was easy to verify).

Let me know if you need more information.

Thanks,
  Alberto

On Mon, Apr 12, 2010 at 04:00:11AM -0000, Alberto Bertogli wrote:
> By the way, given that it looks as if it wants to complain my kernel is not
> Ubuntu's, I'm running a stock, non-patched 2.6.33.1 built with the same config
> I was using in 9.04. It should work, shouldn't it?

I tried to run using Ubuntu's kernel, but xdm didn't even start.

Below is the output of /var/log/Xorg.0.log, which I verified was being touched
each time I did /etc/init.d/xdm stop and then start.

I also checked via ps that no X or xdm was running.

I'm also including the /var/log/Xorg.0.log generated when it starts using
Linus' kernel.

I think I forgot to tell you, in case that it matters, that all of this is
happening on a fairly recent (as of September last year, IIRC) Thinkpad T400.

I'm also having apparently unrelated issues with Ubuntu's kernel asking for my
encrypted (dm-crypt) drive keys at boot time, but I've disabled them so they
don't interfere and will file a bug for that after this one is dealt with (it
makes my laptop really difficult to use, and I can work around it).

Thanks again,
  Alberto

PS: Just happened again, also after pressing enter, while typing this email
(vim on screen over ssh over xfce4-terminal, same as before). Also, as before,
the keypress didn't make it to the application.

/var/log/Xorg.0.log, ubuntu kernel (2.6.32-19-generic)
======================================================

X.Org X Server 1.7.6
Release Date: 2010-03-17
X Protocol Version 11, Revision 0
Build Operating System: Linux 2.6.24-23-server x86_64 Ubuntu
Current Operating System: Linux acacia 2.6.32-19-generic #28-Ubuntu SMP Thu Apr 1 10:39:41 UTC 2010 x86_64
Kernel command line: root=UUID=3e0f0fb3-6d42-4537-98aa-a38655ac03dd ro
Build Date: 09 April 2010 01:56:13AM
xorg-server 2:1.7.6-2ubuntu2 (Bryce Harrington <email address hidden>)
Current version of pixman: 0.16.4
 Before reporting problems, check http://wiki.x.org
 to make sure that you have the latest version.
Markers: (--) probed, (**) from config file, (==) default setting,
 (++) from command line, (!!) notice, (II) informational,
 (WW) warning, (EE) error, (NI) not implemented, (??) unknown.
(==) Log file: "/var/log/Xorg.0.log", Time: Mon Apr 12 01:13:35 2010
(==) Using config file: "/etc/X11/xorg.conf"
(==) Using config directory: "/usr/lib/X11/xorg.conf.d"
(==) No Layout section. Using the first Screen section.
(==) No screen section available. Using defaults.
(**) |-->Screen "Default Screen Section" (0)
(**) | |-->Monitor "<default monitor>"
(==) No monitor specified for screen "Default Screen Section".
 Using a default monitor configuration.
(**) Option "DontZap" "False"
(==) Automatically adding devices
(==) Automatically enabling devices
(WW) The directory "/usr/share/fonts/X11/cyrillic" does not exist.
 Entry deleted from font path.
(==) FontPath set to:
 /usr/share/fonts/X11/misc,
 /usr/share/fonts/X11/100dpi/:unscaled,
 /usr/share/fonts/X11/75dpi/:unscaled,
 /usr/share/fonts/X11/Type1,
 /usr/share/fonts/X11/100dpi,
 /usr/share/fonts/X11/75dpi,
 /var/lib/defoma/x-ttcidfont-conf.d/dirs/TrueType,
 built-ins
(==) ModulePath set to "/usr/lib/xorg/extra-modules,/usr/lib/xorg/modules"
(II) The ser...

On Mon, Apr 12, 2010 at 04:00:11AM -0000, Alberto Bertogli wrote:
> It just happened again, as I was typing this, three times in a short
> succession.
>
> Two of those three, vim was not open and I was typing on irsii over a screen
> over ssh.
>
> At least two of those times it happened the moment I pressed enter (or was
> quite an incredible coincidence). Now that I think of it, maybe the other
> times it also happened on enter, but I can't be sure.

It happened at least 4 more times since then (in less than an hour), and all of
them were right after pressing enter. In all cases, the keypress didn't make
it to the application.

Thanks,
  Alberto

On Mon, Apr 12, 2010 at 04:50:25AM -0000, Alberto Bertogli wrote:
> On Mon, Apr 12, 2010 at 04:00:11AM -0000, Alberto Bertogli wrote:
> > It just happened again, as I was typing this, three times in a short
> > succession.
> >
> > Two of those three, vim was not open and I was typing on irsii over a screen
> > over ssh.
> >
> > At least two of those times it happened the moment I pressed enter (or was
> > quite an incredible coincidence). Now that I think of it, maybe the other
> > times it also happened on enter, but I can't be sure.
>
> It happened at least 4 more times since then (in less than an hour), and all of
> them were right after pressing enter. In all cases, the keypress didn't make
> it to the application.

I attached gdb to the currently-running X, and I've not been able to reproduce
it again. It's been running for a couple of hours now, without luck.

I'll keep running it to see what happens. At least it makes an awkward
workaround =)

Thanks,
  Alberto

On Mon, Apr 12, 2010 at 07:10:19AM -0000, Alberto Bertogli wrote:
> On Mon, Apr 12, 2010 at 04:50:25AM -0000, Alberto Bertogli wrote:
> > On Mon, Apr 12, 2010 at 04:00:11AM -0000, Alberto Bertogli wrote:
> > > It just happened again, as I was typing this, three times in a short
> > > succession.
> > >
> > > Two of those three, vim was not open and I was typing on irsii over a screen
> > > over ssh.
> > >
> > > At least two of those times it happened the moment I pressed enter (or was
> > > quite an incredible coincidence). Now that I think of it, maybe the other
> > > times it also happened on enter, but I can't be sure.
> >
> > It happened at least 4 more times since then (in less than an hour), and all of
> > them were right after pressing enter. In all cases, the keypress didn't make
> > it to the application.
>
> I attached gdb to the currently-running X, and I've not been able to reproduce
> it again. It's been running for a couple of hours now, without luck.

Just crashed:

 Program received signal SIGQUIT, Quit.
 0x00007f46fcb33fd3 in __select_nocancel ()
     at ../sysdeps/unix/syscall-template.S:82
 82 in ../sysdeps/unix/syscall-template.S
 (gdb) bt
 #0 0x00007f46fcb33fd3 in __select_nocancel ()
     at ../sysdeps/unix/syscall-template.S:82
 #1 0x000000000045f97a in WaitForSomething ()
 #2 0x0000000000430952 in ?? ()
 #3 0x00000000004261aa in _start ()
 (gdb) c
 Continuing.

 Program exited with code 01.

Sorry about the lack of symbols, but I just noticed that I missed gdb's
complaint about symbols mismatch (there were more, pasting only a few):

 warning: the debug information found in "/usr/lib/debug/usr/bin/Xorg" does not match "/usr/bin/Xorg" (CRC mismatch).
 warning: the debug information found in "/usr/lib/debug/usr/lib/xorg/modules/extensions/libextmod.so" does not match "/usr/lib/xorg/modules/extensions/libextmod.so" (CRC mismatch).
 warning: the debug information found in "/usr/lib/debug/usr/lib/xorg/modules/extensions/libdbe.so" does not match "/usr/lib/xorg/modules/extensions/libdbe.so" (CRC mismatch).
 warning: the debug information found in "/usr/lib/debug/usr/lib/xorg/modules/extensions/libglx.so" does not match "/usr/lib/xorg/modules/extensions/libglx.so" (CRC mismatch).

I got the symbols from the repository yesterday, packages
xserver-xorg-core-dbg, libgl1-mesa-dri-dbg, and xserver-xorg-video-intel-dbg.

Thanks,
  Alberto

On Mon, Apr 12, 2010 at 02:37:09PM -0000, Alberto Bertogli wrote:
> On Mon, Apr 12, 2010 at 07:10:19AM -0000, Alberto Bertogli wrote:
> Sorry about the lack of symbols, but I just noticed that I missed gdb's
> complaint about symbols mismatch (there were more, pasting only a few):

I filed a bug for this, bug 562418.

I also collected apport information with the Ubuntu kernel. Note that it does
not start X, but as it may contain relevant information anyway (and you asked
for it before) I thought it would be appropriate to add it to this bug report.

I'll file a different bug report for that, since they seem unrelated.

Thanks a lot,
  Alberto

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

On Thu, Apr 15, 2010 at 12:42:01AM -0300, Alberto Bertogli wrote:
> I also collected apport information with the Ubuntu kernel. Note that it does
> not start X, but as it may contain relevant information anyway (and you asked
> for it before) I thought it would be appropriate to add it to this bug report.

I did, using

 apport-cli -u 560899 -c apport-save

where apport-save is the file I collected with apport-cli --save when I was
under the Ubuntu kernel. I can't send from there because I have no network
manager, so I have no network (I'm connecting this laptop via wireless).

I hit the "no Ubuntu kernel" message, which seemed odd since I had collected
the information using an Ubuntu kernel (that was the whole point), so I
commented it out.

However, it seems that it sent the _current_ information and not the one in
the file. At least the BootDmesg.txt shows the contents of the current kernel
boot, and not the one apport saved in the file.

So I'm attaching the file apport generated with the Ubuntu kernel. If there's
a better way, please let me know.

Thanks,
  Alberto

You've been running a custom kernel - have you done much other customisation? In particular, have you done anything with plymouth? The SIGQUIT there makes me suspicious.

On Tue, Apr 20, 2010 at 05:15:40AM -0000, Chris Halse Rogers wrote:
> You've been running a custom kernel - have you done much other
> customisation? In particular, have you done anything with plymouth?

I'm running a stock, on-patched 2.6.33.1 kernel from kernel.org, the same one
I was using with Karmic (I did not rebuild it, or rebuild the initramfs, it's
really the same kernel), with a pretty normal configuration.

I don't know what plymouth is, so I don't think so.

This bug reproduced right after upgrading to Lucid. I did not have any odd
configuration conflict or anything to adjust, it was a smooth upgrade, except
for this.

After reading https://wiki.ubuntu.com/MainInclusionReport/Plymouth trying to
find more about plymouth, I guess you should know that I don't see any kind of
graphic boot, probably because I do not have a framebuffer console.

I just see the normal text boot sequence until xdm starts. X also runs just
fine until it exits (i.e. all the normal applications, panel firefox, etc.,
open up and run just fine).

Everything runs ok under gdb, that's how I've been using my system since the
upgrade.

> The SIGQUIT there makes me suspicious.

Sorry I can't provide more details about that, I've filed bug #562418 and I
hope it gets solved soon so I can get more useful backtraces.

Thanks for taking a look. Is there anything you want me to do/test?

  Alberto

On Tue, Apr 20, 2010 at 05:41:23AM -0000, Alberto Bertogli wrote:
> On Tue, Apr 20, 2010 at 05:15:40AM -0000, Chris Halse Rogers wrote:
> > You've been running a custom kernel - have you done much other
> > customisation? In particular, have you done anything with plymouth?
>
> I'm running a stock, on-patched 2.6.33.1 kernel from kernel.org, the same one

That, of course, should be "non-patched 2.6.33.1".

Thanks,
  Alberto

After upgrading to 2.6.34 and doing an aptitude full-upgrade (not sure which
update since it's been a long time since I rebooted, and did it only today to
change the kernel), this went away.

So I'm marking the bug invalid.

Thanks,
    Alberto