Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 21 Apr 2005 13:59:02 -0700
From:      "Kevin Oberman" <oberman@es.net>
To:        Steve Kargl <sgk@troutmask.apl.washington.edu>
Cc:        freebsd-current@freebsd.org
Subject:   Re: firefox stuck in libthr or kserel! 
Message-ID:  <20050421205902.6F3E75D07@ptavv.es.net>
In-Reply-To: Your message of "Thu, 21 Apr 2005 13:20:42 PDT." <20050421202042.GA82753@troutmask.apl.washington.edu> 

next in thread | previous in thread | raw e-mail | index | archive | help
> Date: Thu, 21 Apr 2005 13:20:42 -0700
> From: Steve Kargl <sgk@troutmask.apl.washington.edu>
> Sender: owner-freebsd-current@freebsd.org
> 
> How to get firefox stuck in an unresponsive state.
> 
> Build these ports
> firefox-1.0.2,1
> linux-flashplugin-6.0r79_2
> linuxpluginwrapper-20050320
> 
> Go to http://espn.com/
> 
> Watch firefox become unresponsive.
> 
> In top we see
> 29645 kargl     96    0 53452K 44028K lthr     0:09  0.00%  0.00% firefox-bin
> 
> and there she sits.  If one iconifies firefox, then de-iconifies
> it, the window is never redrawn.
> 
> If we remove the mapping of libpthread to libthr in /etc/libmap.conf,
> then we see firefox get stuck in a kserel state.
> 
> 33098 kargl     20    0 53772K 43924K kserel   0:09  0.00%  0.00% firefox-bin

This is a problem that has been brought up on gnome@, ports@ and
current@ from time to time. There are many, many sites, all heavily
flash based, that exhibit this problem with any native browser to use
the linux flash6 plugin. As the flash code is the linux binary, it is
not likely to be there. (These sites don't hang on linux systems.)

Between this problem and the lack of Flash 7 support for these browsers,
an every increasing number of major sites are no longer available. Since
it happens with either threading library, it is probably not in the
threading libs, either. That makes the linuxplugingwrapper the prime
suspect in my mind, but it is also possible that there is some sort of
resource exhaustion biting us, possibly due to some odd interaction of
the various pieces.
-- 
R. Kevin Oberman, Network Engineer
Energy Sciences Network (ESnet)
Ernest O. Lawrence Berkeley National Laboratory (Berkeley Lab)
E-mail: oberman@es.net			Phone: +1 510 486-8634



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20050421205902.6F3E75D07>