firefox crashed with SIGSEGV in __kernel_vsyscall() when I was printing

Bug #217032 reported by berg
6
Affects Status Importance Assigned to Milestone
Mozilla Firefox
Fix Released
Critical
firefox (Ubuntu)
Won't Fix
Undecided
Unassigned
firefox-3.0 (Ubuntu)
Fix Released
Undecided
Alexander Sack

Bug Description

Binary package hint: firefox-3.0

I Browser this page
http://riplinton.blogspot.com/2008/04/using-gnulinux-tools-to-recover-dvd.html

Then I Click to Print (file-> print)
Choose "Print to file"
choose PDF for output format
and click ok
A separate windows popup displaying the progress.
at some 30% stopped and after I played with "alt-tab" , firefox crashed.

seems equal many bug reports reporting crashing ( almost linking with flash : almost site used flash :) , but all need infos because apport cant get them and almost was marked as invalid :(

https://bugs.launchpad.net/ubuntu/+source/firefox-3.0/+bug/184957
https://bugs.launchpad.net/ubuntu/+source/firefox-3.0/+bug/188513
https://bugs.launchpad.net/ubuntu/+source/firefox-3.0/+bug/192070
https://bugs.launchpad.net/ubuntu/+source/firefox-3.0/+bug/194853
https://bugs.launchpad.net/ubuntu/+source/firefox-3.0/+bug/195506
https://bugs.launchpad.net/ubuntu/+source/firefox-3.0/+bug/195596
https://bugs.launchpad.net/ubuntu/+source/firefox-3.0/+bug/196190
https://bugs.launchpad.net/ubuntu/+source/firefox-3.0/+bug/196327
https://bugs.launchpad.net/ubuntu/+source/firefox-3.0/+bug/192070
https://bugs.launchpad.net/ubuntu/+source/firefox-3.0/+bug/196248

https://bugs.launchpad.net/ubuntu/+source/firefox-3.0/+bug/213948

My bug report, I think could be different because I was printing. The page "printed" seems not have flash in it

ProblemType: Crash
Architecture: i386
Date: Sun Apr 13 19:43:34 2008
DistroRelease: Ubuntu 8.04
ExecutablePath: /usr/lib/firefox-3.0b5/firefox
NonfreeKernelModules: nvidia
Package: firefox-3.0 3.0~b5+nobinonly-0ubuntu1
PackageArchitecture: i386
ProcCmdline: /usr/lib/firefox-3.0b5/firefox
ProcEnviron:
 PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games
 LANG=pt_BR.UTF-8
 SHELL=/bin/bash
Signal: 11
SourcePackage: firefox-3.0
StacktraceTop:
 __kernel_vsyscall ()
 raise () from /lib/tls/i686/cmov/libpthread.so.0
 ?? () from /usr/lib/xulrunner-1.9b5/libxul.so
 ?? ()
 ?? ()
Title: firefox crashed with SIGSEGV in __kernel_vsyscall()
Uname: Linux 2.6.24-16-generic i686
UserGroups: adm admin audio cdrom dialout dip floppy lpadmin netdev plugdev powerdev scanner video

Revision history for this message
berg (berg-foss) wrote :
Revision history for this message
Apport retracing service (apport) wrote : Symbolic stack trace

StacktraceTop:_cairo_surface_set_clip (surface=0x93956c0, clip=0xbfe1d06c)
_cairo_pdf_surface_emit_pattern (surface=0x93956c0, pdf_pattern=0xbfe1d350)
_cairo_pdf_surface_show_page (abstract_surface=0x93956c0)
*INT_cairo_surface_show_page (surface=0x93956c0)
_cairo_paginated_surface_show_page (abstract_surface=0x8c9dd90)

Revision history for this message
Apport retracing service (apport) wrote : Symbolic threaded stack trace
Revision history for this message
Apport retracing service (apport) wrote : Stack trace with source code
Revision history for this message
berg (berg-foss) wrote :

there is many reports about crashes and hardy is a LTS version. and there is more importance to find the roots.

Changed in firefox-3.0:
assignee: nobody → asac
Revision history for this message
berg (berg-foss) wrote :

I use the commands copied from some page from ubuntu and adapt and generate this attached gdb.log

mkdir t
apport-unpack /var/crash/_usr_lib_firefox-3.0b5_firefox.1000.crash ./t
cd t
 gdb -nx /usr/lib/firefox-3.0b5/firefox -core CoreDump -ex 'backtrace full' -ex 'thread apply all backtrace full' -ex 'quit' 2>&1 | tee gdb.log

Revision history for this message
berg (berg-foss) wrote :

running this command: strace -r -o trace_firefox1.txt firefox-3.0

I attached the output (big output)

Revision history for this message
In , berg (berg-foss) wrote :
Download full text (4.8 KiB)

User-Agent: Mozilla/5.0 (compatible; Konqueror/4.0; Linux) KHTML/4.0.3 (like Gecko)
Build Identifier: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9b5) Gecko/2008040514 Firefox/3.0b5

From start page ( local page) I Can print without problems
this happen always when :
I Browser this page
 http://riplinton.blogspot.com/2008/04/using-gnulinux-tools-to-recover-dvd.html
Then I Click to Print (file-> print)
 Choose "Print to file"
 choose PDF for output format
 and click ok
 A separate windows popup displaying the progress.
 at some 33% stopped and after I played with "alt-tab" , but firefox crashed.
seems equal many bug reports reporting crashing ( almost linking with flash : almost site used flash :) , but all need infos because apport cant get them and almost was marked as invalid :(
https://bugs.launchpad.net/ubuntu/+source/firefox-3.0/+bug/184957
 https://bugs.launchpad.net/ubuntu/+source/firefox-3.0/+bug/188513
 https://bugs.launchpad.net/ubuntu/+source/firefox-3.0/+bug/192070
 https://bugs.launchpad.net/ubuntu/+source/firefox-3.0/+bug/194853
 https://bugs.launchpad.net/ubuntu/+source/firefox-3.0/+bug/195506
 https://bugs.launchpad.net/ubuntu/+source/firefox-3.0/+bug/195596
 https://bugs.launchpad.net/ubuntu/+source/firefox-3.0/+bug/196190
 https://bugs.launchpad.net/ubuntu/+source/firefox-3.0/+bug/196327
 https://bugs.launchpad.net/ubuntu/+source/firefox-3.0/+bug/192070
 https://bugs.launchpad.net/ubuntu/+source/firefox-3.0/+bug/196248
https://bugs.launchpad.net/ubuntu/+source/firefox-3.0/+bug/213948
My bug report, I think could be different because I was printing. The page "printed" seems not have flash in it
ProblemType: Crash
 Architecture: i386
 Date: Sun Apr 13 19:43:34 2008
 DistroRelease: Ubuntu 8.04
 ExecutablePath: /usr/lib/firefox-3.0b5/firefox
 NonfreeKernelModules: nvidia
 Package: firefox-3.0 3.0~b5+nobinonly-0ubuntu1
 PackageArchitecture: i386
 ProcCmdline: /usr/lib/firefox-3.0b5/firefox
 ProcEnviron:
  PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games
  LANG=pt_BR.UTF-8
  SHELL=/bin/bash
 Signal: 11
 SourcePackage: firefox-3.0
 StacktraceTop:
  __kernel_vsyscall ()
  raise () from /lib/tls/i686/cmov/libpthread.so.0
  ?? () from /usr/lib/xulrunner-1.9b5/libxul.so
  ?? ()
  ?? ()
 Title: firefox crashed with SIGSEGV in __kernel_vsyscall()
 Uname: Linux 2.6.24-16-generic i686
 UserGroups: adm admin audio cdrom dialout dip floppy lpadmin netdev plugdev powerdev scanner video

I entered a bug at lauchpad( ubuntu) https://bugs.launchpad.net/ubuntu/+source/firefox-3.0/+bug/217032

Reproducible: Always

Steps to Reproduce:
1.I Browser this page
 http://riplinton.blogspot.com/2008/04/using-gnulinux-tools-to-recover-dvd.html
2.Then I Click to Print (file-> print)
 Choose "Print to file"
3.choose PDF for output format
 and click ok
Actual Results:
 separate windows popup displaying the progress.
 at some 33% stopped and after I played with "alt-tab" , but firefox crashed.
seems equal many bug reports reporting crashing ( almost linking with flash : almost site used flash :) , but all need infos because apport cant get them and almost was marked as invalid :(
https://bugs.launchpad.net/ubuntu/+source/...

Read more...

Revision history for this message
In , berg (berg-foss) wrote :

Created attachment 315698
I created a backtrace form cashdump

Revision history for this message
In , berg (berg-foss) wrote :

strace -r when I tested a simple local page and works fine.
after I tested with a page from web and crashed

http://launchpadlibrarian.net/13440839/trace_firefox1.txt.tar.gz

Changed in firefox:
status: Unknown → New
Revision history for this message
In , Timeless-bemail (timeless-bemail) wrote :

you're kidding, right?

the first bug you linked to is an invalid mess. and you managed to link to the same report twice. hint: don't link to more than one external report, and if the report is invalid, it's probably not something you want qa to read, so don't bother linking at all.

#3 <signal handler called>
^ this is when you crashed. the stuff past this point is just cleanup so that you can run firefox again.

fwiw, your attachment is almost useful, except it's missing the critical bits:
No symbol table info available.
#4 0xb6dde63b in ?? () from /usr/lib/libcairo.so.2
No symbol table info available.
#5 0xb7c1d50a in ?? () from /usr/lib/libnspr4.so.0d

please install the -dbg packages for these libraries. then get a new stack trace.

fwiw, we don't generally have a use for an strace. we need a complete and valid stack trace. these come from things like debuggers such as gdb. however, in order for them to be able to give you a good stack trace, they need debugging symbols. information on installing those symbols (from -dbg packages on your platform) can be obtained from your distro vendor.

as it happens, your vendor even has a wiki page specifically for this product, and it's in one of the bugs you linked to. in general it's a good idea to read things you want others to read....

https://wiki.ubuntu.com/MozillaTeam/Bugs

note that had you read that wiki you'd at least have symbols for nspr. sadly for unknown reasons they don't tell you to install symbols for cairo. please feel free to kindly suggest to them that they add it to the list. (you should really have symbols for: x, glib, gdk, and cairo in addition to the ones they list.)

Revision history for this message
In , berg (berg-foss) wrote :

First. I am not wasting time and efforts for kiding.

I am trying link diverse reports of bugs ( efforts of many many people trying help , inside of your limit of knowledge but are attempting. he may give a hand & no a punch ).
this is a recommendation wherefore bugs alike ( or but also repeated ) they may have faster solution. more bugs reports more people to help (I suppose)

second : in relation to the fact of the first bug is invalid (I showed others), look the review 19, the bug is invalid exactly because do not have a backtrace useful!!!! ( https://bugs.launchpad.net/ubuntu/+source/firefox-3.0/+bug/184957/comments/19 )
 Look others commentaries and others bugs report , the number is frightful :(

https://bugs.launchpad.net/ubuntu/+source/firefox-3.0?field.searchtext=&orderby=-datecreated&search=Search&field.status%3Alist=NEW&field.status%3Alist=INCOMPLETE_WITH_RESPONSE&field.status%3Alist=CONFIRMED&field.status%3Alist=TRIAGED&field.status%3Alist=INPROGRESS&field.status%3Alist=FIXCOMMITTED&field.assignee=&field.bug_reporter=&field.omit_dupes=on&field.has_patch=&field.has_no_package=

My bug reports ( and not yet is invalid :)

https://bugs.launchpad.net/ubuntu/+source/firefox-3.0/+bug/217032

and I used the instrunctions from https://wiki.ubuntu.com/MozillaTeam/Bugs

I will try install all dbg and dbgsym packages ( https://wiki.ubuntu.com/DebuggingProgramCrash )
using this script
#!/bin/bash
echo on
dpkg -l | cut -c5-40 > /tmp/loop.txt

for arquivo in $(cat /tmp/loop.txt); do
        echo trying DBG "$arquivo-dbg"
        apt-get install -y --force-yes "$arquivo-dbg"
        echo trying DBGSYM "$arquivo-dbgsym"
        apt-get install -y --force-yes "$arquivo-dbgsym"

done

the libcairo2-dbg package is reallly missing from recommendation at mozillateam.

root@unstable-desktop:/home/unstable# dpkg -S libnspr4.so.0d
libnspr4-0d-dbgsym: /usr/lib/debug/usr/lib/libnspr4.so.0d
libnspr4-0d: /usr/lib/libnspr4.so.0d
root@unstable-desktop:/home/unstable# dpkg -S libcairo.so.2
libcairo2-dbg: /usr/lib/debug/usr/lib/libcairo.so.2.17.3
libcairo2: /usr/lib/libcairo.so.2.17.3
libcairo2: /usr/lib/libcairo.so.2
root@unstable-desktop:/home/unstable#

thank you for his considerations, links and by attention.

Revision history for this message
In , berg (berg-foss) wrote :

I tested with the version at http://ftp.mozilla.org/pub/mozilla.org/firefox/nightly/latest-trunk/

Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9pre) Gecko/2008041504 Minefield/3.0pre

And crashed too

I observed that at java.sun.com main page print to file ( pdf option) always crash
Important: ps option don't crash ( yet :)

the firefox had detect by itself and send the following crash report:

Add-ons: <email address hidden>:0.5~rc1,{972ce4c6-7e08-4474-a285-3208198ce6fd}:2.0
BuildID: 2008041504
Comments: I was make a test with the lastest nightly trunk to test the bug report https://bugzilla.mozilla.org/show_bug.cgi?id=429071 and continues to crash when I print to file ( pdf option) on site like java.sun.com
CrashTime: 1208289242
Email: <email address hidden>
InstallTime: 1208289025
ProductName: Firefox
StartupTime: 1208289025
Theme: classic/1.0
URL: http://java.sun.com/
UserID: 1ea4f17d-b9c0-c3ec-0c09-d23d534de442
Vendor: Mozilla
Version: 3.0pre

This report also contains technical information about the state of the application when it crashed.

Revision history for this message
In , Timeless-bemail (timeless-bemail) wrote :

Unless we have symbols for your cairo, the report won't actually be any more useful than what you've already pasted.

sadly, the output you pasted from crash reporter is absolutely useless. please load about:crashes and copy the bp- id.

we want to get the crash reporter ui improved, but we ran afoul of a localization freeze, so there's probably not much we can do about it for 3.0 :(.

Revision history for this message
In , berg (berg-foss) wrote :

Created attachment 315839
second output with more debug symbol installed ( fixing missing important files)

Revision history for this message
In , berg (berg-foss) wrote :

Created attachment 315840
list of all packages installed at my system( I installed all possible dbg and dbgsym packages)

Other information about command-line options used in the latest backtrace
unstable@unstable-desktop:~$ firefox-3.0 -safe-mode
** Message: GetValue variable 1 (1)
** Message: GetValue variable 2 (2)
** Message: GetValue variable 1 (1)
** Message: GetValue variable 2 (2)
** Message: GetValue variable 1 (1)
** Message: GetValue variable 2 (2)
** Message: GetValue variable 1 (1)
** Message: GetValue variable 2 (2)
** Message: GetValue variable 1 (1)
** Message: GetValue variable 2 (2)
** Message: GetValue variable 1 (1)
** Message: GetValue variable 2 (2)
612,000000 792,000000
48960 63360
Falha de segmentação (core dumped)
unstable@unstable-desktop:~$

Revision history for this message
berg (berg-foss) wrote :
Revision history for this message
In , berg (berg-foss) wrote :

>>load about:crashes and copy the bp- id.
        Report ID Date Submitted

    8902a961-0b26-11dd-82ac-001321b13766 15-04-2008 16:00

http://crash-stats.mozilla.com/report/index/8902a961-0b26-11dd-82ac-001321b13766

Sorry for inconveniences, I am learning and each bug report I learn more.
thanks .

Revision history for this message
berg (berg-foss) wrote :
Revision history for this message
In , Timeless-bemail (timeless-bemail) wrote :

oh, heh, i was wrong about cairo. sorry. in our builds, we build our own cairo. in distro builds, you get system cairo.

thanks, both stacks are useful. although sadly they aren't identical.

Revision history for this message
berg (berg-foss) wrote :
Changed in firefox-3.0:
status: New → Confirmed
Revision history for this message
In , Martijn-martijn (martijn-martijn) wrote :

I just filed bug 429678, which might be related to this bug?

Revision history for this message
In , Mats Palmgren (matspal) wrote :

I'm pretty sure it's the same problem as bug 429678. Applying the patch
there fixes it for me (on Linux).

Revision history for this message
In , berg (berg-foss) wrote :

Today I tested with the version below and Woks Fine !! Thanks :)

Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9pre) Gecko/2008050104 Minefield/3.0pre

Changed in firefox:
status: New → Fix Released
Revision history for this message
Alexander Sack (asac) wrote : Re: [Bug 217032] Re: firefox crashed with SIGSEGV in __kernel_vsyscall() when I was printing

On Fri, May 02, 2008 at 03:43:36AM -0000, Bug Watch Updater wrote:
> ** Changed in: firefox
> Status: New => Fix Released
>

 affects ubuntu/firefox
 status wontfix

we don't have to fix it in ubuntu as it will automatically land here
once we update to latest upstream.

 - Alexander

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

fix released in 3.0 final

Changed in firefox-3.0:
status: Confirmed → Fix Released
Changed in firefox:
importance: Unknown → Critical
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.