Satellite Stream Stops working on HD (1 Viewer)

g00se

Portal Member
April 28, 2014
19
1
52
Home Country
United Kingdom United Kingdom
I'm pretty much having the same problems with the 6922SE, MP will record 5-10 minutes of video then the video is wiped by artifacts (green) and XBMC reports no signal. At that point if I stop the recording and start it again it'll be fine. This only seems to happen during scheduled recordings. I have unfortunately wiped my MP installation (and logs) which was getting long in the tooth, up until a month ago I was using a SkyStar USB2HD which never missed a beat for 7 years until its power supply blew.
I tried using Windows Media Center and that was fine so I'm pretty sure this is an MP problem. Will post logs the next time it happens but they look very similar to OP's.
 

g00se

Portal Member
April 28, 2014
19
1
52
Home Country
United Kingdom United Kingdom
Log from today, the stream starts recording properly again when I went to watch the channel on XMBC at 7:26
 

mm1352000

Retired Team Member
  • Premium Supporter
  • September 1, 2008
    21,577
    8,224
    Home Country
    New Zealand New Zealand
    g00se, please provide all the log files.
     

    g00se

    Portal Member
    April 28, 2014
    19
    1
    52
    Home Country
    United Kingdom United Kingdom
    This includes today's error 20 minutes in.
     

    g00se

    Portal Member
    April 28, 2014
    19
    1
    52
    Home Country
    United Kingdom United Kingdom
    Done, hopefully tonights scheduled record will produce something you can work with.

    Thanks for responding by the way :).
     

    g00se

    Portal Member
    April 28, 2014
    19
    1
    52
    Home Country
    United Kingdom United Kingdom
    Logs with debug on attached.
     

    Attachments

    • tvserver-debug.on-g00se.zip
      120.2 KB

    mm1352000

    Retired Team Member
  • Premium Supporter
  • September 1, 2008
    21,577
    8,224
    Home Country
    New Zealand New Zealand
    Thanks g00se :)

    I feel like I still don't have enough information to go on here.

    I'm guessing that these latest logs are intended to show a problem with the recording of "Match Of The Day"?

    From your earlier description my working theory is that you're seeing corruption - "video is wiped by artifacts (green)" - when you try to play the recording in XBMC. If the problem is indeed stream corruption, the solution will involve finding and eliminating the source of that corruption using a methodical approach and process of elimination.

    So where do we stand?
    The logs you've provided don't show any direct evidence of corruption. Having said that, I'm a little suspicious of the signal strength/quality reading at the start of the recording (96% and 81%) and I also don't have the TsWriter log for the first 1.5 hours (TsWriter.bak not provided).

    What would help?
    More information:
    • the TsWriter.bak log
    • confirmation that you were expecting me to investigate "Match of the Day"
    • hardware info - particularly GPU (ie. what GPU do you have) and storage (do you timeshift/record with an HDD or SSD or NAS, and how much free space does that drive have) info
    • software info - eg. Have you tried to play the recordings in other software like MediaPortal (if not, why not), and what was the result? Have you enabled DXVA (hardware accelerated video decoding)? Do you have any security software installed that might be scanning XBMC and/or MediaPortal => load on CPU and/or HDD (ie. have you checked CPU and HDD load in resource monitor during playback) => problems?
    • detail about the corruption - eg. Do the problems occur right from the start of the recording or begin at a certain point? Are the problems more severe at certain times in the recording, or with certain recordings, or at certain times of day, or with certain weather?
    Ultimately I guess what I'm trying to say is that right now there appears to be no evidence that the problem has anything to do with TV Server. For all I know the problem could be XBMC. We need more information to start to narrow down the source.

    mm
     

    g00se

    Portal Member
    April 28, 2014
    19
    1
    52
    Home Country
    United Kingdom United Kingdom
    Hmm...well here's the tswriter.bak, yes this time the usual corruption started later at 40 minutes in as opposed to the usual 20 minutes.

    Yes it was match of the day.
    GPU is an AMD 7750.
    The corruption happens at any time after 20 minutes usually, sometimes as low as 10 and sometimes later. Most of the time it is 20 minutes in though.
    I've played it through XBMC and VLC, the file size indicates that the files aren't the expected size also
    Signal strength has never been a problem to me, I'm usuing a 1M dish in a 45CM area.This never used to happen with the Skystar USB box either, although sensitivity matters.
    I'm running Windows 8.1 Pro 64 bit I've tried this with and without Windows Defender running and also turning off PCiE power saving.With no changes.No other security software.
    Hard Drives are 30 gig free on the system SSD and 600gig free on D:.
     

    Attachments

    • TsWriter.bak
      1.4 MB
    Last edited:

    mm1352000

    Retired Team Member
  • Premium Supporter
  • September 1, 2008
    21,577
    8,224
    Home Country
    New Zealand New Zealand
    Okay, so now we have evidence of corruption.
    The TsWriter.bak log shows continuity errors starting at 40 minutes into the recording:
    [2014-05-03 22:28:06,399] [5be7340] [153c] - Recorder: RECORD Start 'D:\Media\Recordings\BBC One HD\Match of the Day\Match of the Day - 2014-05-03.ts'
    [2014-05-03 22:28:06,399] [5be7340] [18f4] - Recorder: RECORD start of audio detected
    [2014-05-03 22:28:06,400] [5be7340] [18f4] - Recorder: RECORD start of video detected
    [2014-05-03 22:28:06,400] [5be7340] [18f4] - Recorder: RECORD clear TS packet queue
    [2014-05-03 22:28:06,400] [5be7340] [18f4] - CDiskRecorder::WriteToRecording() - Reset write buffer throttle
    [2014-05-03 22:28:06,400] [5be7340] [18f4] - Recorder: RECORD Info : Next broadcaster program clock reference rollover : 0 days 16:18:37 0
    [2014-05-03 22:28:06,400] [5be7340] [18f4] - Recorder: RECORD start of audio detected
    [2014-05-03 23:11:50,241] [5be7340] [18f4] - Recorder:pid 1518 Continuity error... 3 ( prev d ) - bad signal?
    [2014-05-03 23:12:06,097] [5be7340] [18f4] - Recorder:pid 1518 Continuity error... 7 ( prev f ) - bad signal?
    [2014-05-03 23:12:24,406] [5be7340] [18f4] - Recorder:pid 1518 Continuity error... f ( prev 9 ) - bad signal?
    [2014-05-03 23:12:54,676] [5be7340] [18f4] - Recorder:pid 1518 Continuity error... 2 ( prev 8 ) - bad signal?
    [2014-05-03 23:13:20,470] [5be7340] [18f4] - Recorder:pid 1518 Continuity error... 7 ( prev 1 ) - bad signal?
    [2014-05-03 23:13:30,409] [5be7340] [18f4] - Recorder:pid 1518 Continuity error... e ( prev 8 ) - bad signal?
    [2014-05-03 23:13:30,409] [5be7340] [18f4] - Recorder:pid 1519 Continuity error... e ( prev c ) - bad signal?
    [2014-05-03 23:13:46,821] [5be7340] [18f4] - Recorder:pid 1518 Continuity error... f ( prev c ) - bad signal?
    [2014-05-03 23:13:46,821] [5be7340] [18f4] - Recorder:pid 1518 Continuity error... 2 ( prev f ) - bad signal?
    [2014-05-03 23:13:50,220] [5be7340] [18f4] - Recorder:pid 1518 Continuity error... 6 ( prev 2 ) - bad signal?

    In other words, this problem is baked into the recording file. It won't matter which software you try to play it with - you'll have problems.

    A steady stream of continuity errors continue for the next ~10 minutes... and then they stop:
    [2014-05-03 23:24:38,854] [5be7340] [18f4] - Recorder:pid 1518 Continuity error... 5 ( prev e ) - bad signal?
    [2014-05-03 23:24:38,889] [5be7340] [18f4] - Recorder:pid 1518 Continuity error... 9 ( prev 9 ) - bad signal?
    [2014-05-03 23:24:38,896] [5be7340] [18f4] - Recorder:pid 1518 Continuity error... 5 ( prev e ) - bad signal?

    To me this suggests either:
    1. Your PC had a sustained period of heavy HDD load (like a virus scan) at that time.
    2. Bad signal quality/strength at that time.
    I think the first option is possibly more likely.

    Thoughts?
     

    Users who are viewing this thread

    Top Bottom