Bug 146339

Summary: Unable to run firefox after installing FC3 problem with JS_dtostr
Product: [Fedora] Fedora Reporter: Richard Hubbell <richard.hubbell>
Component: firefoxAssignee: Christopher Aillon <caillon>
Status: CLOSED NOTABUG QA Contact:
Severity: medium Docs Contact:
Priority: medium    
Version: 3CC: barryn, mattdm, wtogami
Target Milestone: ---   
Target Release: ---   
Hardware: athlon   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2005-02-24 22:02:55 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:

Description Richard Hubbell 2005-01-27 07:17:45 UTC
Description of problem:
firefox seg faults and drops core on startup

Version-Release number of selected component (if applicable):
firefox 1.0 and fedora core 3 with no updates

How reproducible:
1. install an abbreviated fc3
   o  xfce
   o  xdm
   o  no kde, no gnome, no graphics, just enough to
      have a working system

  
Actual results:

trace from core file reveals we get to nsProfileLock's
FatalSignalHandler
Looks like JS_dtostr in libmozjs.so is the culprit
also js_NumberToString  (looks the same)



Expected results:
firefox starts

Additional info:
I did a lot of searching and subscribed to fedora-devel.
There seems to be a problem with gcc  search google for prdtoa.c
and you'll see what I mean.

I even tried to build firefox (AVIARY branch) but ran into the
same problem but it cropped up in shlibsign.

I installed FC3 twice on a fresh disk same result both times

Comment 1 Sitsofe Wheeler 2005-01-27 23:36:51 UTC
What does "rpm -q firefox" say?

Comment 2 Richard Hubbell 2005-01-28 04:00:21 UTC
I should have said that comes with FC3, I thought it was 1 but it's
firefox-0.10.1-1.0PR1.20


Also included the anaconda kickstart file:
# Kickstart file automatically generated by anaconda.

install
cdrom
lang en_US.UTF-8
langsupport --default=en_US.UTF-8 en_US.UTF-8
keyboard us
xconfig --card "ATI Rage 128" --videoram 32768 --hsync 30-96 --vsync 50-140
--resolution 800x600 --depth 16
network --device eth0 --bootproto dhcp
rootpw --iscrypted 
firewall --enabled
selinux --disabled
authconfig --enableshadow --enablemd5
timezone America/Tijuana
bootloader --location=mbr
# The following is the partition information you requested
# Note that any partitions you deleted are not expressed
# here so unless you clear all partitions first, this is
# not guaranteed to work
#clearpart --linux
#part / --fstype "ext3" --onpart hda1
#part swap --noformat --onpart hda2

%packages
@ editors
@ xemacs
@ system-tools
@ dialup
@ authoring-and-publishing
@ xfce-desktop
@ base-x
@ printing
@ development-tools
-python-ldap
-valgrind
-psgml
-perl-XML-LibXML
-vim-enhanced
-gdm
-gcc-java
-authconfig-gtk
-xorg-x11-twm
-ltrace
-rhn-applet
-up2date-gnome
-zsh
-cscope
-emacs-leim
-elfutils
-usermode-gtk
-switchdesk
-oprofile
-diffstat
-freeglut
-system-config-network
-system-config-printer-gui
-doxygen
-xterm
-system-config-services
-bluez-pin
kernel
-system-config-display
-gcc-g77
-indent
-firstboot
-perl-XML-Dumper
-memprof
sysstat
-openssh-askpass-gnome
-perl-LDAP
-ctags
-cvs
-perl-XML-Parser
-rcs
-screen
-perl-XML-NamespaceSupport
nmap-frontend
-xdelta
-emacspeak
tsclient
-rhgb
-splint
-perl-Crypt-SSLeay
-perl-XML-Grove
e2fsprogs
-perl-XML-LibXML-Common
vnc
-system-config-date
-perl-XML-SAX
-perl-XML-Twig
grub
-valgrind-callgrind
-open
mrtg
-openldap-clients
-linuxwacom
-system-config-packages
-cdecl
-system-config-users
-system-logviewer
-emacs
-synaptics
-system-config-soundcard
-perl-XML-Encoding

%post

Comment 3 Richard Hubbell 2005-02-02 16:25:09 UTC
To 'ye who reads this here message.....shall I be givin' up all hope on this? 
No doubt there are more important bugs.  I have abandoned all hope.  Installing
new firefox from mozilla.org gives me same problem.

Comment 4 Christopher Aillon 2005-02-02 16:41:40 UTC
I think I've heard reports that the i386 version works.  Have you tried that? 
Unfortunately, I won't be able to get to look at this for a while, but it is on
the list of bugs I have flagged to fix for FC4.

Comment 5 Richard Hubbell 2005-02-02 17:03:31 UTC
Dunno where to find i386 version??  I only find this
http://ftp.mozilla.org/pub/mozilla.org/firefox/releases/1.0/linux-i686/en-US/



Comment 6 Barry K. Nathan 2005-02-06 05:25:58 UTC
Between comment #2 and the fact that this bug is filed against athlon not
x86_64, I think Richard Hubbell already tried the i386 version and still hit
this problem. (I haven't seen this problem however.)

Comment 7 Richard Hubbell 2005-02-09 16:23:19 UTC
To Barry K. Nathan, how did you try to repro?  Or are you just saying
"I don't have this problem, haven't ever seen this problem from anyone" ?

Comment 8 Barry K. Nathan 2005-02-09 23:24:20 UTC
I'm saying whenver I've installed FC3, firefox has Just Worked(tm). I've never
seen this crash.

I don't think I've tried installing X and Firefox on a minimal install and
testing that setup, however. (It's possible that I did that once and that it
also worked for me, but I don't remember for sure, so I need to try again at
some point. "some point" definitely won't be today or tomorrow, though.)

Comment 9 Richard Hubbell 2005-02-17 00:20:40 UTC
I suppose I've had stranger hardware problems, but this ranks right up there. 
Turns out my cpu was bad.  Don't understand anything more about it. I could do
most everything on the computer.  Exceptions were running any browser.  firefox,
mozilla, konqueor, opera.  All a no go. The GTK color picker would crash.  I had
problems with gpg, but only in some versions.  I finally realized it was a cpu
problem after running seti.  It behaved very strangely and I got OVERFLOW errors.
I could compile linux kernels with no apparent problems either.  All is a go
after replacing the athlon with a duron.  I can't figure out how to retire this
so I'll leave that up to you.

Comment 10 Barry K. Nathan 2005-02-17 03:06:30 UTC
It should be closed as "NOTABUG". (I'm not the bug reporter, nor am I a Red Hat
employee, so I can't do that myself.)

Comment 11 Matthew Miller 2005-02-24 22:02:55 UTC
closing NOTABUG as per comment #10.