[mythtv-users] Category M "Missed" recording

David Engel david at istwok.net
Tue Jan 14 17:38:28 UTC 2014


On Tue, Jan 14, 2014 at 11:00:10AM -0500, Eric Sharkey wrote:
> On Tue, Jan 14, 2014 at 10:11 AM, David Engel <david at istwok.net> wrote:
> > How long are your scheduler runs?
> 
> This was the worst case run I could find in the last two days of logs:
> 
> Jan 12 22:12:58 mythtered mythlogserver: mythbackend[5033]: I
> Scheduler scheduler.cpp:2207 (HandleReschedule) Scheduled 1758 items
> in 21.3 = 1.26 match + 16.94 check + 3.15 place
> 
> In all but three cases they completed in under 10 seconds, with an
> average of around 5 seconds.
> 
> The machine should be well over-spec'ed for the job at hand.  It's an
> AMD FX 8150 8-core 3.6GHz CPU, 8GB of RAM, and the database is on an
> SSD.
> 
> >  And do you have any slave backends
> > that could have gotten stuck or been restarted?
> 
> No, I do not have any slave backends.  Just one frontend/backend
> combined system and two frontend only systems.

I can't tell you what happened then.  There should only be two ways to
get a recording marked as missed.  The first requires the master
backend to have been restarted.  You said that didn't happen.  The
second requires the scheduler to get stuck or be busy for the entirety
of a recording and then mysteriously get unstuck.  To the best of my
knowledge, that has never happened to anyone.  When things get stuck,
they tend to stay that way until they are restarted.

David
-- 
David Engel
david at istwok.net


More information about the mythtv-users mailing list