[vdr] Re: two VDR boxes

["Huber, Matthias RD-CP1" <Matthias.Huber@heidelberg.com>]

>I would prefer a simple 'remote-timer' feature inside the VDR core.
>The timer menu could be extended by one entry which would 
>describe on which machine the recording should be done. If it 
>is set to a remote machine, then the entry could be sent via 
>svdrp to the specified computer. All this could be done inside 
>VDR with some simple patches I think (optimisticly). 
>Or extend the timer menu in a way that you directly modify the 
>remote timers.conf instead of the local one. 
>As to EPG, I think that there is no urgend need to have a 
>shared epg. any client can have its own epg.data file in its 
>private config dir for example.
>and the video dirs can of course be shared via nfs so all 
>recordings are available to any clients.
>
>As I see it, the only problem is handling the timers. (and 
>have the channels.conf in sync on all VDRs, but that can be 
>done manually)
>
>So is anyone volunteering to dig into the timer handling issue 
>? :) my c++ is rather limited...

Hi,
I'm working on a "ClientMode"-Patch for VDR, for use in a client-server
VDR environment with no DFB cards in the clients. In fact, my clients
are diskless VDRs without recording capability, streaming via streamdev
plugin.
The patch adds a "ClientMode" entry to the setup. If set to yes, the
timers list will be synchronized with the server and every modification
on the client side results in an immediate change on the server side
(via SVDRP). Recording of the client side will be suppressed, every
record command (even timeshifting) leads to recording on the server.
This way, the VDR clients act as fully transparent frontends to the VDR
server.
The patch is ready to test, but I had not yet time to test it.
Any comments are wellcome.
Greetings,
Matthias Huber