armagetronad.real keeps running and consumes 100% of one core after game exits

[Diego Liedo]

Binary package hint: armagetronad

After playing one round in armagetronad (fullscreen) and exiting via the game menu, the process is still running and consuming an whole core of cpu. Every time I've reproduced it gmpc-mpd were also active (playing) and compiz enabled.
Steps: 1. Turn on computer, 2. Start armagetronad, 3. Exit after playing one round.
After that, top shows armagetronad.re using 100% cpu and must be killed manually.
Thanks.

ProblemType: Bug
Architecture: amd64
Date: Sun Nov 15 19:37:11 2009
DistroRelease: Ubuntu 9.10
ExecutablePath: /usr/games/armagetronad.real
InstallationMedia: Ubuntu 9.10 "Karmic Koala" - Release amd64 (20091027)
Package: armagetronad 0.2.8.2.1-10ubuntu1
ProcEnviron:
 LANG=es_MX.UTF-8
 SHELL=/bin/bash
ProcVersionSignature: Ubuntu 2.6.31-14.48-generic
SourcePackage: armagetronad
Uname: Linux 2.6.31-14-generic x86_64
XsessionErrors:
 (gnome-settings-daemon:1675): GLib-CRITICAL **: g_propagate_error: assertion `src != NULL' failed
 (gnome-settings-daemon:1675): GLib-CRITICAL **: g_propagate_error: assertion `src != NULL' failed
 (polkit-gnome-authentication-agent-1:1805): GLib-CRITICAL **: g_once_init_leave: assertion `initialization_value != 0' failed
 (nautilus:1777): Eel-CRITICAL **: eel_preferences_get_boolean: assertion `preferences_is_initialized ()' failed
 (awn-applet-activation:1862): Vte-CRITICAL **: vte_terminal_set_background_image_file: assertion `VTE_IS_TERMINAL(terminal)' failed

[Christopher Kyle Horton (christhehorton)]

I get something similar, except the offending file is called armagetronad.real.

[Diego Liedo (diegoliedo)]

I did the last ubuntu update which included the 2.6.31-15 kernel and the problem was solved, Armagetronad also gained sound which was missing before. I think this bug should be closed. Was it the same for you WarriorIng64?

[Diego Liedo (diegoliedo)]

I'm really sorry and a bit ashamed about the last comment on this bug (it wasn't solved), but I found out a couple of things:
If the system has just been rebooted armagetronad runs as it should (with sound and not using a complete CPU core).
The problem seems to have a relation with flash: whenever gnash, swfdec or flashplugin are, or have been used during reboot (not session), armagetronad looses sound and hogs the cpu.
I think this is a more general bug but armagetronad is the only 3d demanding program I use apart from compiz.

[Diego Liedo (diegoliedo)]

The relation between armagetronad and the flash plugin didn't exist, I'm clueless again.

[owlstead (maarten-bodewes)]

I've got this problem on both an AMD and an Intel system using 64 bit Ubuntu.

2 different sound cards, 2 different video cards (AMD & NVidia)

I've had quite a few problems with armagetron and sound, and I would not be terribly surprised if it has something to do with that.

[owlstead (maarten-bodewes)]

Update: after disabling sound all issues are gone. Two processes seem to be launched, one for the game and one for sound, probably using inter-process communication. It seems to me that the game is ill-prepared for the latest kind of processors that introduce more dynamic timing. Sound is extremely jittery after which the sound is gone at all, then the sound process becomes a zombie eating away at one of my 4 cores. The other process becomes jittery as well as it tries in vain to communicate with the unstable process. The only other problem I have is that there is sometimes a noticeable lag when the game is trying to do too much within a few single frames - introducing lag. This is also much less pronounced without sound. At least, that is my take on the issue. I'm a software engineer, if anyone is interesting in more details (logs etc.) please replay to this thread.

[Christopher Kyle Horton (christhehorton)]

I'm now using Armagetron Advanced on Ubuntu 10.04 64-bit, and I am no longer getting this problem.