Re: Fwd: Firefox 3.6.3 / xulrunner 1.9.2.3 segmentation fault

[Victor Dorneanu <victor%dornea.nu@localhost>]

Sorry for the late answer. I was able to solve the problem in some
unconventional way. I had a look at the firefox package and found the
binary "firefox" in /usr/pkg/lib/firefox. I was able to run it but I
couldn't access any https sites. Then I took a look at the shell script
"firefox" in /usr/pkg/bin/:

$ cat /usr/pkg/bin/firefox

#! /bin/sh
#
# $NetBSD: firefox.sh.in,v 1.1 2009/09/16 19:06:18 tnn Exp $
#
# We currently overwrite the nsXULStub launcher with this wrapper script.
# The stub causes problems like HTTPS protocol not working.
# XXX to be investigated

exec /usr/pkg/lib/xulrunner/xulrunner-bin
/usr/pkg/lib/firefox/application.ini "$@"

Then I've changed to:

exec /usr/pkg/lib/xulrunner/xulrunner
/usr/pkg/lib/firefox/application.ini "$@"


... and the segfault error disapperead! That's really strange. At the
moment I don't really have the time to investigate that. I'm glad I can
use firefox again.




On 07/13/10 14:11, Tobias Nygren wrote:
> On Mon, 12 Jul 2010 10:26:50 +0000
> Victor Dorneanu <victor%dornea.nu@localhost> wrote:
>
>   
>> Meanwhile I was able to backtrace the problem:
>>     
>   
>> 0x00007f7ffdc08492 in pthread_mutex_lock () from /usr/lib/libpthread.so.0
>>     
> This thread is waiting on a lock and is probably not the one that
> caused the fault. You'll need to make sure to trace the faulting
> thread. Use gdb's "info threads" and "thread <n>" commands.
> Look for any threads that are not in a wait state.
> Also, compiling devel/xulrunner with the "debug" option will
> make your backtraces alot more useful.
>
> Kind regards,
> -Tobias
>
>
>
>   


-- 
Victor Dorneanu

Contact
- Web/Blog: http://dornea.nu
 
GnuPG information
- KeyID = 0xD20870F4 (pgp.mit.edu)
- Key fingerprint = DD6B 5E09 242F 7410 3F90 492A 4CBA FD13 D208 70F4