Repeated TV: Attempting to change from None to None rapidly fills log

[Marc Randolph <mrand@…>]

Wife unit tried to switch to LiveTV to watch a show that started at 19:00. All she received was a blank screen. She left it there until I got home at 20:02. I'm sorry to say that I didn't have the presence of mind to do an strace or gdb before hitting escape.

I later found the frontend log to be filled with 29000 occurrences (eight a second) of:

Repeated TV: Attempting to change from None to None

1. Any idea what caused a lockup that resulted in a change from None to None, and therefore a blank screen?

2. Is there any way that the logging function, upon receiving the same message as the previous message, could wait at least a second, if not more, to log an identical message to prevent hyperactive messages like this one from filling up logs in record time?

$ mythfrontend --version
Please include all output in bug reports.
MythTV Version   : 23789
MythTV Branch    : branches/release-0-23-fixes
Network Protocol : 56
Library API      : 0.23.20100314-1
QT Version       : 4.5.2
Options compiled in:
 linux profile using_oss using_alsa using_pulse using_jack using_pulseoutput using_backend using_dvb using_firewire using_frontend using_glx_proc_addr_arb using_hdhomerun using_hdpvr using_iptv using_ivtv using_joystick_menu using_lirc using_mheg using_opengl_video using_opengl_vsync using_qtdbus using_qtwebkit using_v4l using_x11 using_xrandr using_xv using_xvmc using_xvmc_vld using_xvmcw using_bindings_perl using_bindings_python using_opengl using_vdpau using_ffmpeg_threads using_libavc_5_3 using_live using_mheg

[Marc Randolph <mrand@…>]

Stack trace from different user with same repeated message in front-end log

[Marc Randolph <mrand@…>]

Threadtrace to go with stacktrace

[Marc Randolph <mrand@…>]

A user actually had his front-end crash (rather than hang like mine) when this occurred. Can't guarantee it is the same thing, but at first glance it appears to be, so attaching (even if they are much less than ideal traces - here's to hoping they help in some small way).

[Marc Randolph <mrand@…>]

bt by attaching gdb to hung frontend

[Marc Randolph <mrand@…>]

strace of hung frontend

[Marc Randolph <mrand@…>]

Another occurrence

[robertm]

Should no longer be possible in trunk.