[mythtv-users] Moving MythTV backend to another computer

Michael T. Dean mtdean at thirdcontact.com
Wed Apr 2 10:43:40 UTC 2014


On 04/01/2014 06:02 PM, Mike Perkins wrote:
> On 01/04/14 22:05, jedi wrote:
>> On Tue, Apr 01, 2014 at 02:51:57PM -0400, Michael T. Dean wrote:
>>> On 04/01/2014 11:48 AM, Tony Lill wrote:
>>>> I don't know
>>>> why they don't make it official since they just have to not break what
>>>> already works.
>>>
>>> Because parts of MythTV's functionality (involving file and file
>>> system management) break when you have any tunerless backends.  If
>>> you're going to do things wrong, please don't recommend that other
>>> people do the same.  And, really, you should at least create demo
>>> tuners so that your system won't be broken.
>>
>>      Given the requirement that content be streamed through a backend
>> rather accessed through a lower level protocol, that requirement seems
>> a bit out of date.

mythmediaserver is for systems that have content (Videos, etc.) to serve 
but no tuners.

And, FWIW, mythjobqueue is for systems that run jobs but have no tuners.

Eventually (if I get my way), we'll have separate mythrecorder and 
mythmaster applications, too.

>> If a machine has all of the disks but doesn't have
>> any tuners, then running a tunerless backend on that box should not be
>> problem. Just a basic operational requirement.
>>
> If a machine has all of the disks and doesn't have a tuner, it doesn't 
> need to be a backend. You might just as well treat the box as a NAS 
> since there is no benefit running a backend on it.
>

Actually, going beyond that, you have to treat the box as a NAS.

"If a machine has all of the disks but doesn't have any tuners," then 
running a mythbackend or a mythmediaserver is still useless--since we 
don't write recordings via the protocol.  Therefore, in the described 
instance you have to use some sort of network file system 
(NFS/CIFS/NBD/iSCSI/...).

Mike


More information about the mythtv-users mailing list