ubiquity crashed with SIGILL

Bug #440407 reported by Cedric F.
90
This bug affects 18 people
Affects Status Importance Assigned to Milestone
ubiquity (Ubuntu)
Invalid
High
Unassigned
Karmic
Invalid
High
Unassigned
webkit (Ubuntu)
Fix Released
High
Unassigned
Karmic
Fix Released
High
Unassigned

Bug Description

Binary package hint: ubiquity

I install new Ubuntu 9.10 beta1 with a USB key

ProblemType: Crash
Architecture: i386
Date: Fri Oct 2 09:16:33 2009
DistroRelease: Ubuntu 9.10
ExecutablePath: /usr/lib/ubiquity/bin/ubiquity
InterpreterPath: /usr/bin/python2.6
LiveMediaBuild: Ubuntu 9.10 "Karmic Koala" - Beta i386 (20090929.2)
Package: ubiquity 1.99.28
ProcCmdline: /usr/bin/python /usr/lib/ubiquity/bin/ubiquity --only
ProcEnviron:
 LANGUAGE=
 PATH=(custom, no user)
 LANG=fr_FR.UTF-8
ProcVersionSignature: Ubuntu 2.6.31-11.36-generic
Signal: 4
SourcePackage: ubiquity
StacktraceTop:
 ?? ()
 ?? () from /usr/lib/libwebkit-1.0.so.2
 ?? () from /usr/lib/libwebkit-1.0.so.2
 ?? () from /usr/lib/libwebkit-1.0.so.2
 ?? () from /usr/lib/libwebkit-1.0.so.2
Title: ubiquity crashed with SIGILL
Uname: Linux 2.6.31-11-generic i686
UserGroups:

Revision history for this message
Cedric F. (c-fachinetti) wrote :
Cedric F. (c-fachinetti)
visibility: private → public
Revision history for this message
mmbossoni (mmbossoni-gmail) wrote :

I have the same issue when ubiquity start installation (at 2%) doing partitions

Revision history for this message
mmbossoni (mmbossoni-gmail) wrote :

Workaround to install is to uninstall ubiquity slideshow
sudo apttitude purge ubiquity-slideshow-ubuntu ubiquity-slideshow
The installer works in the same way as in 9.04

Revision history for this message
Martin G Miller (mgmiller) wrote :

Attempting to install from desktop CD also gives this problem. The install window fails and drops you back to the live CD desktop. It also does this if you choose to install the system from the live CD without running the live desktop. It fails out the process and puts you onto the live CD desktop.

The command you give above: sudo apttitude purge ubiquity-slideshow-ubuntu ubiquity-slideshow returns error messages.

I went into synaptic and "completely" uninstalled ubiquity-slideshow-ubuntu.

It now seems to have installed normally and is ready to be restarted on the hard drive for the first time.

Amazing that you can do all this from the live session while it's installing.

Revision history for this message
Cedric F. (c-fachinetti) wrote :

So I installed the beta on several machines. This bug affects 2 to 5 machines. It is a problem ...

Evan (ev)
Changed in webkit (Ubuntu):
importance: Undecided → High
Evan (ev)
Changed in ubiquity (Ubuntu):
importance: Undecided → High
Revision history for this message
Apport retracing service (apport) wrote : Stacktrace.txt (retraced)

StacktraceTop:?? ()
?? () from /usr/lib/libwebkit-1.0.so.2
?? () from /usr/lib/libwebkit-1.0.so.2
?? () from /usr/lib/libwebkit-1.0.so.2
?? () from /usr/lib/libwebkit-1.0.so.2

Revision history for this message
Apport retracing service (apport) wrote : ThreadStacktrace.txt (retraced)
Changed in ubiquity (Ubuntu):
status: New → Invalid
Revision history for this message
Apport retracing service (apport) wrote : Crash report cannot be processed

Thank you for your report!

However, processing it in order to get sufficient information for the
developers failed (it does not generate an useful symbolic stack trace). This
might be caused by some outdated packages which were installed on your system
at the time of the report:

libgcc1: installed version 1:4.4.1-4ubuntu4, latest version: 1:4.4.1-4ubuntu6
gcc-4.4-base: installed version 4.4.1-4ubuntu4, latest version: 4.4.1-4ubuntu6
sysv-rc: installed version 2.87dsf-4ubuntu7, latest version: 2.87dsf-4ubuntu8
libdbus-1-3: installed version 1.2.16-0ubuntu6, latest version: 1.2.16-0ubuntu8
console-setup: installed version 1.34ubuntu2, latest version: 1.34ubuntu4
python: installed version 2.6.2-0ubuntu1, latest version: 2.6.3-0ubuntu1
ca-certificates: installed version 20090701, latest version: 20090814
debconf: installed version 1.5.27ubuntu1, latest version: 1.5.27ubuntu2
sysvinit-utils: installed version 2.87dsf-4ubuntu7, latest version: 2.87dsf-4ubuntu8
libglib2.0-0: installed version 2.22.0-1ubuntu1, latest version: 2.22.1-0ubuntu1
ubiquity-casper: installed version 1.196, latest version: 1.198
python2.6: installed version 2.6.2-3ubuntu1, latest version: 2.6.3-0ubuntu2
python-minimal: installed version 2.6.2-0ubuntu1, latest version: 2.6.3-0ubuntu1
libklibc: installed version 1.5.15-1, latest version: 1.5.15-1ubuntu2
libc-bin: installed version 2.10.1-0ubuntu12, latest version: 2.10.1-0ubuntu13
debconf-i18n: installed version 1.5.27ubuntu1, latest version: 1.5.27ubuntu2
tzdata: installed version 2009m-1, latest version: 2009n-1
upstart: installed version 0.6.3-5, latest version: 0.6.3-7
python2.6-minimal: installed version 2.6.2-3ubuntu1, latest version: 2.6.3-0ubuntu2
initscripts: installed version 2.87dsf-4ubuntu7, latest version: 2.87dsf-4ubuntu8
initramfs-tools: installed version 0.92bubuntu50, latest version: 0.92bubuntu51
libstdc++6: installed version 4.4.1-4ubuntu4, latest version: 4.4.1-4ubuntu6
libc6: installed version 2.10.1-0ubuntu12, latest version: 2.10.1-0ubuntu13
klibc-utils: installed version 1.5.15-1, latest version: 1.5.15-1ubuntu2
perl-base: installed version 5.10.0-24ubuntu3, latest version: 5.10.0-24ubuntu4

Please upgrade your system to the latest package versions. If you still
encounter the crash, please file a new report.

Thank you for your understanding, and sorry for the inconvenience!

tags: removed: need-i386-retrace
Revision history for this message
Rob Hamm (drhamm) wrote :

@Martin and mmbossoni:
Martin, if you copied and pasted the purge command, that might have been the reason the workaround didn't work for you. mmbossoni typo'd an extra "t" in "aptitude." I tried the workaround, and so far, so good.
:-)

Revision history for this message
Alexander Sack (asac) wrote :

is this reproducible? Maybe we can get a better backtrace?

Revision history for this message
mmbossoni (mmbossoni-gmail) wrote :

Hi
I can reproduce the bug 100% of the time in my machine.
Maybe the libwebkit was compiled with an incompatible instruction set for my processor (An Athlon XP 1600+).
Access non existent instructions can cause sigill (we cannot use SSE2 instructions in a processor without them for example, but the processor that compiles the code can have it).
On my Vostro1000 laptop with a Turion64 X2 TL-60 everything works.

Revision history for this message
Cedric F. (c-fachinetti) wrote :

Ok, I have an athlon XP 2500+, and I have the same problem !

Revision history for this message
madbiologist (me-again) wrote :

I'm also getting this when attempting to install from the desktop liveCD of Karmic 9.10 beta.

Uname: Linux 2.6.31-11-generic i686

I'm not sure if it is a cross-compilation problem, but just in case, I have an AthlonXP 2700+

Of course the way to test this would be to get the source code for libwebkit and compile it on an affected system, but I can't do this at the moment, if at all (have never compiled anything before).

Revision history for this message
Colin Watson (cjwatson) wrote :

We don't use any compiler flags to set the architecture to anything other than the default when building webkit, but webkit does include inline assembly which is a classic cause of this kind of thing. Unfortunately the automatic retracer doesn't seem to have done anything useful with the stack trace, so it's a bit hard to pin down ...

Changed in webkit (Ubuntu Karmic):
status: New → Incomplete
status: Incomplete → New
Revision history for this message
Sebastien Bacher (seb128) wrote :

could be similar to bug #452574?

Revision history for this message
Cedric F. (c-fachinetti) wrote :

In the last daily-build of 20, this bug seems to have disappeared...
Can anyone confirm ?

Revision history for this message
Julio Danni (jdanni) wrote :

This seems fixes in Releasae Candidate

Revision history for this message
madbiologist (me-again) wrote :

I am no longer getting this with the desktop liveCD of Karmic 9.10 Release Candidate.

Uname: Linux 2.6.31-14-generic i686

The install no longer fails with an error near the end of the install process, followed by returning me to the liveCD desktop. The install now completes successfully and prompts for a restart of the system.

I am also seeing the install slideshow for the first time - it did not appear during the attempt to install from the beta liveCD (and appears to have been the cause of the problem - purging it from the beta as suggested above worked for me too).

Changed in webkit (Ubuntu Karmic):
status: New → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.