Re: another gvfs freeze break

[Alexander Larsson <alexl redhat com>]

On Tue, 2009-03-10 at 12:06 -0400, Matthias Clasen wrote:
> On Tue, Mar 10, 2009 at 12:00 PM, Alexander Larsson <alexl redhat com> wrote:
> > Here is another one.
> >
> > We're leaking gvfs daemon side monitor objects because the client code
> > sending dbus messages to unsubscribe to the monitor didn't pass in all
> > required arguments.
> >
> > Bug: http://bugzilla.gnome.org/show_bug.cgi?id=563418
> > Patch: http://bugzilla.gnome.org/attachment.cgi?id=130409&action=view
> >
> > Can i commit this?
> >
> 
> Looks good to me. +1

Turns out that actually unsubscribing things triggers a crash in the
previously not executing code. Basically we're accessing things after
freeing them in the unsubscribe codepath.

Same bug as above, new patch at:
http://bugzilla.gnome.org/attachment.cgi?id=130575

Can I commit this?