- Server "A" will be the master node for 50 databases, and "B" their
slave node.
- Server "B" will be the master node for 50 other databases, and "A"
their slave node.
Do you think it's a viable setup ?
Doesn't sound like what I had in mind when designing the system. It
will be complicated to administer to say the least.
Hi,
I'm writing specific scripts to automate to the maximum the
administration tasks, handling each database separately, and sequentially.
So, the complexity is reduced for me.
Actually, I think that the pain will come from the memory/connexion
overhead...
It would be interesting to be able to group several databases into one
cluster.
For the above situation, two clusters would then be suficient. (Note
that if I take into account the advices I received, this setup could be
more or less achieved with two big databases and using schemas, but with
some drawbacks).
However, as far a I'm concerned (and I think it would apply to any mass
hosting service), in order to keep flexibility, and scalability, the
best solution would be to keep 100 distinct clusters, but directed by a
minimum of slon processes...
Do you think any of the above solutions are achievable with Slony ?
Thank you,
Philippe Ferreira.
_______________________________________________
Slony1-general mailing list
[email protected]
http://gborg.postgresql.org/mailman/listinfo/slony1-general
