[mythtv-users] Help needed for scripted recordings

Klaas de Waal klaas.de.waal at gmail.com
Mon Mar 30 19:15:18 UTC 2020


On Mon, 30 Mar 2020 at 00:27, Bill Meek <keemllib at gmail.com> wrote:

> On 3/29/20 5:01 PM, Klaas de Waal wrote:
>
> .
>
> > Hi Bill,
> >
> > I got a hacked version of vcr.py now running in the way I want it.
> > What I do now is to start the recording with second accuracy instead of
> > minute accuracy by effectively removing the code that forces the seconds
> > field to be zero.
> > This is only in the code that uses UTC time when the specified time is in
> > the past. So I do specify a time in the past and that works OK.
> > I can now create a recording that starts immediately instead of on the
> next
> > minute and this then enables waiting a number of seconds before deleting.
> > I also hard-coded the delays for the time being.
> > A bit strange is that the backend has not yet crashed with all this
> > testing... but that is for another day.
> >
> > Thanks!
> > Klaas.
>
> I was reading the ticket a back on the 27th and noticed that deleting
> remotely was
> part of it: "The program is deleted from the frontend on a separate
> machine".
>
> Using the API, even remotely still sends the delete request to the
> backend. Don't
> know if that's important - just thought I'd mention it.
>
>
> The ticket describes the situation in which the records are deleted with
the mythtfrontend GUI on a separated frontend machine. The mythtv-backend
then crashes.
Since then I have been able to reproduce the problem on my development
system which is a combined backend/frontend. Also here the
recordings-in-progress are deleted via the GUI.
With the GUI I can now reproduce the problem with on average 5 deletes of a
recording in progress, almost never more than 10.
I have now the script running, three in parallel, each script doing a start
recording and a delete in about 25 second. This has been running for hours
without making it crash. Somehow this is good of course but not what I
wanted....

As I understand it mythfrontend communicates with mythbackend with the
so-called mythprotocol, while the ServicesAPI is implemented as a HTTP
webserver on the same mythbackend. This suggests that the frontend, via the
mythprotocol, does something different than the ServicesAPI.
This definitely needs some thinking, it might even be necessary to
understand what the code is actually doing.... which is usually not need to
fix crashes. Just find the null pointer....

Thanks,
Klaas.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mythtv.org/pipermail/mythtv-users/attachments/20200330/66327be8/attachment.htm>


More information about the mythtv-users mailing list