GNOME Bugzilla – Bug 140289
gnome-terminal repeatedly eats cpu time, refresh stops for long time
Last modified: 2004-12-22 21:47:04 UTC
open more than one gnome terminal with heavy activities (i.e. compile kernel or one port) and large window size. Sometimes gnome-terminal eats all available cpu time (apart form the processes running in the terminals) and screen update stalls for more a network timeout period. This is an old gnome-terminal problem (as far as i use it, at least under all gnome2-versions).
This happens to me too. Here is a more detailed description of the conditions under which this happens: Run irrsi 0.8.9 inside of screen 3.9.15 (other versions will probably work) and log into an active discussion and stay there for awhile. gnome-terminal will eventually end up with a corrupt screen. The bug seems to be tied to updates of the blue status bar between the discussion and the text you type. So, logging into several active channels and moving between them so the channel activity indicators frequently change and change color may help cause the bug to happen faster. Also see these two mailing list messages, others are having the problem too: http://www.redhat.com/archives/fedora-list/2004-January/msg00343.html http://www.redhat.com/archives/fedora-list/2004-January/msg00450.html This is a recent thing, and seems to be tied to gnome-terminal 2.5.x or 2.6.x
the same thing happens to me, I use Mutt on screen and every few minutes gnome terminal hangs with corrupted screen...unfortunately ALL other Gnome terminals hang too :-( it makes my desktop very unusable (it takes 100% of the CPU). I use Debian and 2.6.1 version of the Gnome Terminal. Here is backtrace of the GT process: (gdb) bt full
+ Trace 46474
PLEASE do something with it! :-(
Someone appears to be doing something with this. This is a bug in the 'vte' module. Apparently gnome-terminal hands of most of the terminal emulation duties to a pluggable module. Anyway, here's a link to the vte bug that's actually a duplicate of this one: http://bugzilla.gnome.org/show_bug.cgi?id=117945
I can't see with such clarity why the two bugs are related. In fact, I don't understand this bug at all... It seems to be a case of multiple personality: the initial comment does -not- mention a crash, nor does the second one. The third one, otoh, does. Karol, could you get a stack trace with symbols? And post it on 138866. In any case, the stack trace you provide in comment #2 is broken, because the function vte_terminal_fork_command does not call vte_terminal_set_default_colors. This happens in 138866 too... Please make sure that the package was compiled with debugging symbols and see http://bugzilla.gnome.org/getting-traces.cgi for more information about useful stack traces. --- I'll mark this as a dup of 138866, which has exactly the same stack trace. *** This bug has been marked as a duplicate of 138866 ***