From Bugzilla Helper:
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20030703
Description of problem:
Some times, but not always, after I log in (which starts two gnome-terminal
windows) and start a full-height GNU Emacs next to the right border on a
1920x1440 screen (I have an icon in my gnome panel that starts it), if I use
Ctrl-Alt-Rightarrow to switch to the second virtual desktop (where there's
nothing yet), the window manager crashes and automatically starts a few times.
It has also happened some times on a 1600x1200 screen on my laptop. Every time
it crashes, the GNU Emacs window moves down a bit, so I know the number of
crashes varies, and I easily notice when it happens. Unfortunately, it's not
100% reproducible, but it has happened very often in the past few days. Is
there any additional information I could provide to make it easier to track this
Version-Release number of selected component (if applicable):
Steps to Reproduce:
1.Log in such that a gnome-terminal is started on desktops 1 and 4
2.Click on an icon that runs `emacs -geometry 80x106-0+0' (or 80x83 for 1600x1200)
Actual Results: The window manager sometimes crashes, sometimes more than once
Expected Results: It shouldn't
Non-default applets I have in my panel (that might be causing such problems?)
are the keyboard layout switcher and the weather applet. I also have panel
One of the times I noticed the crash, I also noticed a run-away rhn applet
eating all CPU. Since I'd logged in right after a reboot, it wasn't there
before, so it might be related, but it might be an entirely unrelated problem,
or even a consequence of the WM crash.
This is fixed in GNOME CVS, so basically we just have to make a new tarball
release and get it in the tree.
Michael I believe this is the same crash you were seeing.
Should be gone in 2.6.2
Confirmed, I couldn't get this with 2.6.2-1 after several attempts, and it would
occur pretty much every time for me. However, I've had WM crashes when closing
apps lately. I still haven't been able to come up with a sequence of steps that
will trigger the bug, though. In fact, I coudln't even come up with a sequence
of steps that will sometimes trigger it :-) I'll file a separate bug.