[mythtv-users] Before I tear my hair out can a slave BE...

Michael T. Dean mtdean at thirdcontact.com
Thu Feb 5 19:54:20 UTC 2009


On 02/05/2009 09:29 AM, Mike Perkins wrote:
> Michael T. Dean wrote:
>> He is.  It will be in MythTV 0.22 at the soonest.
>>
>> If you have more than 1 remote backend, you can stagger runs of the 
>> script
>> where the first wakes up the first remote backend, then a few minutes 
>> later,
>> the second script checks if more capture cards are required and then 
>> wakes
>> the second remote backend, etc.
>>
> This procedure assumes, to my understanding, that the tuners on the 
> slave(s) are all of the same type as those on the master. What happens 
> when the tuner types are different? E.g. firewire vs OTA vs satellite?
>
> Chris, will your proposed change be able to wake up the correct slave 
> if that proves necessary? 

Don't know what Chris's plans are, but worst case (with the approach 
I've given--that works now ;)--and even if Chris's approach just starts 
waking up backends without knowing which has which capture card as my 
approach does--the host will wake up, then the scheduler re-runs, then 
there's still a conflict, so the next run wakes another host, the 
scheduler re-runs and the conflict is resolved.  Then, the master 
backend and the 2nd remote backend start recording.  Meanwhile, the 1st 
remote backend is idle, so it shuts itself down.  Sure, that did wake it 
up "for nothing", but it didn't take /that/ much energy (a couple 
hundred watts for a few minutes).

It would actually be easy to modify the script I posted to account for 
which channel has the conflict and--using a lookup table in the 
script--wake the appropriate remote backend.  The 
myth_upcoming_recordings.pl script allows output of information in any 
format, so just have it output information about the channel in an 
easy-to-parse location that the script can access.  I'll leave the code 
as an exercise for the reader.  :)

Mike


More information about the mythtv-users mailing list