On Wed, August 6, 2008 11:06 pm, Decibel! wrote:
> Call me paranoid, but I don't trust our users to not modify data on
> our slaves. Currently I've got a trigger that I've applied that
> prevents this, but the solution is an ugly hack, and of course fights
> with londiste's idea of how triggers should work.
>
> So one thing I'd like is to have londiste normally add triggers to
> slaves to prevent updates. Are there any cases anyone can think of
> where updates *should* be allowed?

A clean way of achieving this is to use PgPool on all the slaves to
transparently redirect UPDATES/DELETES/INSERTS to the master.

Regards
Henry

_______________________________________________
Pgpool-general mailing list
[email protected]
http://pgfoundry.org/mailman/listinfo/pgpool-general

Reply via email to