Re: [Nautilus-list] Problem with Nautilus Mozilla binding

[Mike Fleming <mfleming eazel com>]

Brian--

I can't reproduce this on my RH6.2 box, but I've seen similar
deadlocks.  I thought that I'd squeezed them all out, but looking at the
stack trace, this looks like a genuinely new case.

I've filed bugzilla.eazel.com 6574 on this.

Does it repro consistantly or intermittently?  Does it happen for you on
other machines?

Mike

Brian Cameron wrote:
> 
> Nautilus Gurus:
> 
> I am running a build of Nautilus 0.8.1 built with Mozilla milestone 18
> and running on Solaris 8.
> 
> I am launching the Nautilus Mozilla integration by following these steps:
> 
>  + When Nautilus comes up I am clicking on the "Help" tab in the
>    left-hand area of the screen.
>  + Then I click on "Applications->bug-buddy", although clicking on any
>    help entry seems to have the same behavior.
> 
> Then the output listed in the attachment "nautilus.errors" is displayed
> on the screen where nautilus was launched, and as this is happening the
> main section of the Nautilus screen greys out.   Finally the program
> hangs.  The throbber and the list of "Help" sections remain active, but
> the rest of Nautilus stops responding to any activity.
> 
> I have also attached a stack trace of the "nautilus" process generated
> by the "ptrace" program.  It seems to indicate that thread #1 Nautilus
> is trying to receive a reply.  Lastly, I have attached a stack trace
> of the "nautilus-mozilla-content-view" process which seems to be
> likewise hung.  I believe this is the process that is trying to
> talk to the "nautilus" process.  It seems to be hung in a method
> nsAppShell::DispatchNativeEvent, called by GtkMozEmbedPrivate::LoadChrome().
> 
> I am working with debug compiles of both Mozilla and Nautilus, and been
> using gdb to trace through the code (with help from the "Debugging Mozilla
> on Linux FAQ"):
> 
>    http://www.mozilla.org/unix/debugging-faq.html
> 
> However, I am not quite familiar enough with the code to know where to
> look from here, so any tips, workarounds, or information regarding
> how to further diagnose or debug this situation would be much
> appreciated.
> 
> Thanks!
> 
> Brian Cameron
>