Bernd Helmle wrote:
>
> On Wed, 29 Nov 2006 18:20:31 -0600, "Shaun Thomas" <[EMAIL PROTECTED]> wrote:
>   
>> This is probably due to me doing something stupid, but I figured I'd ask
>> anyway.
>>
>> I've got a configuration that looks roughly like this:
>>
>> Node 1 - schema foo, host foo, master of set 1
>> Node 2 - schema bar, host foo, master of set 2
>> Node 3 - schema foo + schema bar, host baz, wants set 1 and 2.
>>
>> Node 1, set 1 -> Node 3 = works fine.
>> Node 2, set 2 -> Node 3 = el nada.
>>
>> The idea here is that I have two schemas on different postgresql
>> logical databases on another machine I'm combining into a single logical
>> database so I can actually access all of the tables in a single query.
>> There's no table overlap, and each set is self-contained on the node
>> defined as the set provider.
>>
>>     
>
> I'm not sure i understand you correctly, but i assume you are trying to
> have two origin's in your Slony-I cluster...
>
> You need to define two separate clusters which could then replicate their 
> master nodes 
> to the same database on your slave (so node 3 is becoming a slave within both 
> clusters), 
> but you cannot have two masters within one slony cluster.
>
>   
That's not true.  You can have multiple sets with multiple origins
within a cluster.

A particular table can have only one origin.

I'd expect this to be able to work, and I'm not quite sure what's busted
about it.
_______________________________________________
Slony1-general mailing list
[email protected]
http://gborg.postgresql.org/mailman/listinfo/slony1-general

Reply via email to