On Sat, May 8, 2010 at 12:12 AM, Mike Christensen <mike(at)kitchenpc(dot)com> wrote:
> What's the best way to do this? Looks like something like pgPool
> might be what I want, but I haven't looked into it deeply yet.
I don't think your requirement and postgres are consistent with each
other. Unless your data volume is *so* tiny that copying it takes
just a few seconds, this concept just won't work. Besides the fact
that I don't think there is a master-master solution that does not
impose a lot of overhead and will deal gracefully with nodes
disappearing and appearing arbitrarily.