[mythtv-users] How do I get coverart/etc in watch recordings?

[f-myth-users at media.mit.edu]

    > Date: Thu, 17 Sep 2009 10:10:42 +0800
    > From: William Kenworthy <billk at iinet.net.au>

    > On Wed, 2009-09-16 at 21:47 -0400, f-myth-users at media.mit.edu wrote:
    > > > Date: Thu, 17 Sep 2009 08:45:57 +1000
    > >     > From: Jean-Yves Avenard <jyavenard at gmail.com>
    > > 
    > >     > The NFS server was 4 minutes ahead of time... even with ntpd running ;
    > > 
    > > ntpd will punt if the difference gets too large and will just let your
    > ...

    > Only if you let it! Read the ntpd man pages, there is a setting that
    > allows it to try and keep tracking no matter what the difference.  There
    > are also settings to "loosen" the default parameters to enable it to
    > track more wildly varying sources (either by stepping, or slewing - your
    > choice), and also a "huffpuff" value that can help deal with overloaded,
    > low bandwidth connections to the reference.  The burst and iburst
    > parameters can also help in some cases.

Thanks!  I should have mentioned that.  And mentioning all the
debugging tools as you did will hopefully also help somebody out.

I think I tried several of those without much luck; 5% fast is so
enormous that it might as well be classified as hardware failure.
(And since it was rare, I didn't really feel like screwing around
with it when it -did- crop up, nor was testing easy, etc, so just
smacking the clock with ntpdate when I detected the pathological
hardware configuration was the fastest way out.)

Nonetheless, I may try again at some point.  But my current solution
works well and has been proven to work several times; there's sure
something to be said for the laziness of letting an already-tested
solution continue to work. :)

    > That being said, there seems to be something badly wrong with your
    > network/systems for this to happen with the default parameters.

I'm not sure here if you're talking to me (who -does- have something
badly wrong with the hardware, every n boots) or JYA (who had a clock
that was 4 minutes fast for unknown reasons), or just in general.
I certainly agree, though.