[mythtv] Very long recordings

[David Hampton]

On Mon, 2021-02-08 at 19:40 +0000, John Pilkington wrote:
> On 08/02/2021 19:21, Klaas de Waal wrote:
> > 
> > 
> > On Mon, 8 Feb 2021 at 19:27, John P Poet <jppoet at gmail.com 
> > <mailto:jppoet at gmail.com>> wrote:
> > 
> >     On Mon, Feb 8, 2021 at 11:09 AM Klaas de Waal
> >     <klaas.de.waal at gmail.com <mailto:klaas.de.waal at gmail.com>>
> > wrote:
> > 
> >         Hi,
> > 
> >         Yesterday I updated my backend to the latest master and
> > since
> >         then some recordings are extended way beyond their
> > scheduled end
> >         time, one of them extended to 21 hours (115GB). Guide data
> > looks
> >         OK so I guess it is the recording. Could this be another
> >         "std:chrono" induced bug?
> > 
> > 
> >     I recently saw this behavior as well. In my case it seemed to
> > be
> >     doubling the length of the recording. Since this was my
> > production
> >     machine, I reverted to 8e14fe7ceba96f and everything seems okay
> >     again. My development machine is not really usable right now.
> > 
> > 
> > Looks here like recordings now extend until the start of the next 
> > recording on the same tuner. This is with master of Sunday on
> > Fedora 31.
> > 
> >   Klaas.
> 
> Both my boxes are on e3cafbc of Friday 5th and are as normal.  F32
> and el7.

I must say I'm happy to hear that as it rules out the std::chrono
changes.  If that turns out to be wrong and you do think its
std::chrono, let me know and I'll look into it.  I have a couple of
overlapping series recordings (every weeknight) on my dev system that
haven't shown any problems with recording times.  I record off a
HDHomerun box, but that shouldn't matter if the problem is in the
scheduler.

David