[Nautilus-list] GMC broke Nautilus (still broken)

["David Fuchs" <dfuchs uniserve com>]

Hello again,

    I'm still battling the same problem with Nautilus crashing when I
close one of it's windows.  Is this list for developers of Nautilus?  I
ask because if not, I may be asking the wrong people.  I would think
someone must have enough experience with GNOME applications to tell me
how gmc would have affected the workings of Nautilus.  Again, here is
what is wrong:

Using:
    Linux Debian (Potato) v2.2 with kernel 2.4.9.
    XFree86 4.1.0 running Ximian GNOME
    Gnome Nautilus 1.0.4 ('nautilus_1.0.4-ximian.4_i386.deb')

Problem:
    When nautilus starts, everything is fine.  I can click desktop icons
to bring up a window containing the appropriate content of the clicked
item.  I can open as many windows as i want, actually.  As soon as I
close a Nautilus window, Nautilus crashes.  The desktop icons stop
redrawing themselves, and I'm left with nothing.  Nautilus is still
running however, as I'm forced to do a CTRL-ALT-F1, log in as root, and
kill all the Nautilus processes before I can log out (it hangs on logout
until I do so).  This all started after I accidentally ran a copy of
`gmc', the standard gnome file manager.  I killed `gmc', and thought
everything was fine, but this is why Nautilus stopped working properly.

    There MUST be somebody who knows what `gmc' did.  It had to have
tampered with a configuration file or something somewhere on the system.
This problem cannot be coincidental (I've done it on two machines now -
one at work, and I did the same think on my home machine about two hours
ago).  I've purged both gmc and Nautilus and reinstalled only Nautilus -
doesn't fix anything.  I've reinstalled oafd and gconfd (as they seemed
related to gnome's session manager), but that didn't help ether.  I even
tried rebooting my machine with an older kernel (2.2.19pre17), but that
didn't help, of course.  Running an strace provided these two lines
right before the Nautilus crash/freeze, if they help:

rt_sigprocmask(SIG_SETMASK, NULL, [RT_0], 8) = 0
rt_sigsuspend([]

Conclusion:
    Any ideas out there?  Help in regards to this would be greatly
appreciated.

Thanks
-David Fuchs