I forced a core dump and did a backtrace on VDR when the problem occurred:
#0 0xb77ec424 in __kernel_vsyscall ()
#1 0xb77a30b9 in __lll_lock_wait () from /lib/i386-linux-gnu/libpthread.so.0
#2 0xb779e54b in _L_lock_791 () from /lib/i386-linux-gnu/libpthread.so.0
#3 0xb779e371 in pthread_mutex_lock () from /lib/i386-linux-gnu/libpthread.so.0
#4 0x0811da32 in cMutex::Lock() ()
#5 0xb5df6928 in Lock (this=0xb0d7ffc8) at ../../../include/vdr/thread.h:91
#6 cLogger::Down (this=0xb0d7ffc8) at cam.c:407
#7 0xb5dfaf39 in cCam::Stop (this=0xb0d456a8) at cam.c:2315
#8 0xb5dfb04e in cCam::Tune (this=0xb0d456a8, channel=0xb0d5229c) at cam.c:2284
#9 0xb5e00e47 in cScDvbDevice::SetChannelDevice (this=0xb42a9da0,
Channel=0xb0d5229c, LiveView=false) at device-tmpl.c:184
#10 0x080ac537 in cDevice::SetChannel(cChannel const*, bool) ()
#11 0x080ac881 in cDevice::SwitchChannel(cChannel const*, bool) ()
#12 0x080beae0 in cEITScanner::Process() ()
#13 0x0812a3c8 in main ()
Is the problem CAM related?
Den 25.11.2011 12:23, skrev Vidar Tyldum:
> After upgrading to 1.7.21 I have started having these issues where the
> picture freezes on the frontend from time to time. It can last from 30
> seconds to several minutes. Looking at the log I see these errors repeating
> vdr-sxfe:  [demux_vdr] ts2es: payload not PES ?
> vdr-sxfe:  [demux_vdr] ts2es: dropping broken PES packet
> VDR itself is not logging anything particular when this happens, despite
> being at loglevel 3.
> I tried just streaming to VLC and the same thing happends, which means it's
> nota frontend issue. I cannot open the OSD when the problem is 'active'.
> If the watchdog is enabled it will cause a restart of VDR.
> If the picture comes back before the watchdog kicks inn I get:
> vdr-sxfe:  [demux_vdr] New PTS: 7275023756 (VIDEO)
> Any ideas on how to proceed to debug this?
> Source is DVB-C, 1.7.17 was my previous version where I did not have this
> issue. Using precompiled packages from
vi...@tyldum.com PGP: 0x3110AA98
vdr mailing list