[mythtv-users] uPnP chaos

[Michael T. Dean]

On 02/01/2014 09:37 PM, Michael T. Dean wrote:
> On 02/01/2014 09:06 PM, Tony Gould wrote:
>> if the optimize script doesn't work, my fix to the same problem was 
>> to look in the recorded table for records with deletepending=1, and 
>> then delete these records and the corresponding records in the 
>> related tables recordedprograms and recordseek.You'd need to know 
>> some SQL to do this, and you'd have to backup your database first.
>
> That's a very dangerous thing to do.  There's a much better approach 
> for cleaning any of that up, and I'm happy to give it to anyone off 
> list once they find that there are a bunch of recordings that are 
> stuck in that state (i.e. were deleted days or longer ago and are 
> still there, even after a full system (including master backend) 
> restart--making sure to wait until mythbackend processes finish 
> shutting down, which may take much longer than your system start 
> scripts are willing to wait).

Oh, and off list because the process of cleaning up the brokenness is 
dangerous--even if done properly--and should /never/ be performed unless 
there actually is a problem, and must be performed in a very specific 
way.  So, I don't want to post the instructions to the list where people 
are likely to see it and may attempt it when they shouldn't/when it will 
cause problems/when it won't help with whatever problem they're 
attempting to solve.

And, FWIW, eventually I'll have MythTV automatically fix the issue 
whenever it occurs (and safely)--just need to get some time to finish 
the patch.  (Oh, and this issue won't occur except in an unlikely race 
condition that may occur when MythTV isn't properly shut down.)

Mike