In 2.0.3 you can:
- create the database on the soon-to-be master node (without slony
turned on)
- take a pg-dump of that database
- load the dump onto one or more soon-to-be slaves
- initialize slony on the master
- initialize slony on each slave
* When you issue the 'subscribe set' command on the slaves, you add the
new (in the 2.0.x version) 5th parameter - "omit copy = yes"
Works just fine.
Michael
On 3/5/10 7:17 AM, Steve Singer wrote:
> Jeff Amiel wrote:
>> I'm sure this is in the docs/archives somewhere but I can't seem to find
>> it.
>>
>> Assuming I am doing a dump/restore of a brand new master node and also
>> restore the same data into a soon-to-be slave/subscriber node, is there
>> any way I can set up slony so that it only replicates 'new' activity?
>
> No not really. With 1.2.x the subscribers need to be populated from the
> provider after the set subscription.
>
> With 2.0.x the CLONE PREPARE and CLONE FINISH commands should allow you
> to take a pg_dump of the provider after you've setup slony and restore
> it on the subscribers. I don't think there is a way you can use a
> pg_dump taken before slony is installed and use it to populate both the
> provider and subscribers.
>
>
>
>> _______________________________________________
>> Slony1-general mailing list
>> [email protected]
>> http://lists.slony.info/mailman/listinfo/slony1-general
>
>
_______________________________________________
Slony1-general mailing list
[email protected]
http://lists.slony.info/mailman/listinfo/slony1-general
