<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
Before digging too deep into the source, I figured I should be sure
that what changed was the upgrade from 0.24 to 0.25. It was not.
For anyone trying to diagnose similar issues (and my own future
reference), here are the steps I followed to find the problem:<br>
<br>
While Myth was recording a show on my problem channel (the only 720p
channel in my lineup and only channel with a problem) using one of
the dual tuners on my Dvico card, I tuned the other tuner to the
same show using azap (from dvb-apps) and captured another copy of
the same show with "cat /dev/dvb/adapter0/dvr0 >> f.mpg".
When I played back the recordings, I found the same type of
corruption on both recordings, but in different locations. The
corruptions were short, intermittent "holes" where the video gets
blocky and/or grey blocks in the image, and the sound drops out for
a second or two. These would happen anywhere from 10 seconds to a
few minutes apart. I thought they were corresponding to the "pes
offset>181" error from the backend log, but not always. For
every error logged, there was corruption, but there were many more
corrupt spots than errors logged. Since both recordings showed the
same type of corruption, I knew it was not a problem with Myth.<br>
<br>
I still have my old Fedora 13 installation on another partition, so
I rebooted under F13 and found the same corruption. I had done this
before, but the corruption was intermittent enough that I didn't see
it. This time I did.<br>
<br>
This led me to the conclusion that my hardware was, indeed, the
culprit. I cracked open the box and cleaned out the dust, moved the
card to another slot and (what I believe actually fixed it) added an
RF amplifier that I hadn't used since I pulled out the analog tuner
when we went all digital.<br>
<br>
My conclusion is that this error:<br>
"DVBRead mpeg/mpegstreamdata.cpp:362 (AssemblePSIP) - Error:
offset>181, pes length & current cannot be queried"<br>
<br>
is actually caused whenever Myth can not put together a valid
packet, whether it be caused by missing EIT data, a corrupted stream
or ???.<br>
<br>
Now my wife can record Mystery again without missing those
all-important clues, and I can record Joe Bonamassa without missing
those all important crescendos! Yea!<br>
<br>
<br>
<div class="moz-cite-prefix">On 11/08/2012 11:42 PM, DaveD wrote:<br>
</div>
<blockquote cite="mid:509CB3DC.80803@guiplot.com" type="cite">
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
<br>
<div class="moz-cite-prefix">On 11/08/2012 10:01 AM, Martin Moores
wrote:<br>
</div>
<blockquote
cite="mid:CAEBYFR+tN9tL2+O+Rro-WeU-Dim1Fq_8QJWsGU1VPOvzSWMP+Q@mail.gmail.com"
type="cite">
<div class="gmail_extra">
<div class="gmail_quote">On 6 November 2012 03:33, DaveD <span
dir="ltr"><<a moz-do-not-send="true"
href="mailto:mythtv@guiplot.com" target="_blank">mythtv@guiplot.com</a>></span>
wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0
.8ex;border-left:1px #ccc solid;padding-left:1ex">
<div bgcolor="#FFFFFF" text="#000000">
<div><br>
</div>
Nope. No such luck. useonairguide is zero on all
channels. dvb_eitscan is set to zero on all cards in my
capturecard table. My eit_cache table contains zero
records. useeit is set to zero for all sources in my
videosource table. (I found these by searching for
"eit" in my backup dump of my database.). System
restarted. Last night's recording still logged 25
errors similar to<br>
<br>
2012-11-05 04:28:45.637491 E [1370/9180] DVBRead
mpeg/mpegstreamdata.cpp:362 (AssemblePSIP) - Error:
offset>181, pes length & current cannot be
queried<br>
<br>
I don't think this has anything to do with EIT scan.
Don't know why the user in the thread you referenced had
success by setting useonairguide to zero, but no luck
for me.<br>
<br>
Can you tell me exactly what this error means? Might be
time to download the source and take a look.<br>
<br>
Dave D.<br>
</div>
</blockquote>
<div><br>
Has the backend had a reboot since doing the changes?<br>
</div>
</div>
</div>
</blockquote>
Yep. That's what I meant by "System restarted". No EIT or
useonairguide settings anywhere in my database and still lots of
"offset>181" errors.<br>
<br>
I've been looking at the source and there's some issues. I need
to figure out where to find 0.24 so I can do a diff and then
download 0.26 to see if it looks like it will help.<br>
<br>
</blockquote>
<br>
</body>
</html>