On Fri, 2009-05-08 at 09:22 -0400, John Drescher wrote: > Since about 1 month ago I have had a serious problem that is reducing > the WAF (wife acceptance factor)
A critical metric that I must monitor at my home as well. :) > since it seems to manifest only on > the her shows. I mean about 1 day a week I get blank recordings from > my hauppage pvr 500. What I mean is the screen is black with no volume > for the entirety of the recording. Before 1 month ago this had never > happened. There are no error messages in dmesg or mythtv at all. And nothing in /var/log/messages either? When this happens maybe you could turn on lots of ivtv debugging: # echo 2047 > /sys/modules/ivtv/parameters/debug to turn on very verbose debugging. Then try taking some action to start or stop a capture. (modinfo ivtv will show you what the debug flags mean: 2047 is a bit excessive) Also $ v4l2-ctl -d /dev/video1 --log-status mmight show something interesting when you're hung up - I don't know > When > the tuner gets into this blank screen mode if I > > mplayer /dev/video1 > > I do get a blank screen with no sound as well. > > This seems to be more frequent on a reboot or restarting of mythtv. > > Since I updated to the latest myth 0.21 fixes svn, updated to a > 2.6.27.21 kernel and to the bleeding edge ivtv all at the same time I > can not rule out any of these so this makes it difficult to debug. > > Also when it did this yesterday (seemed to be triggered by restarting > mythbackend) after stoping mythbackend I could not modprobe -r ivtv > because it said that the device was in use. That happens for two reasons that I can think of: 1) the mplayer child process is not dead and still has the device open (if you used mplayer). Kill the offending mpayer process. You could also do an fuser on /dev/video* /dev/vbi* /dev/radio* to see what process has what device node open. 2) The kernel oops'ed or bug'ed in the ivtv driver. When this happens, you won't be able to unload the module, but dmesg or /var/log/messages should have something in it. Regards, Andy > However I do not believe > this was the case. There were no myth processes running and I did not > have any other captures running either. > > > Has anyone had this problem? Ideas how to debug this? > _______________________________________________ ivtv-users mailing list [email protected] http://ivtvdriver.org/mailman/listinfo/ivtv-users
