On Wed, Feb 4, 2009 at 8:35 AM, Gerd König <[email protected]> wrote:
> 2.) get back to sync after database failure [..] > => works, but isn't that nice since this causes database downtime and I > think there are better ways ;-) You may want to try online recovery with PostgreSQL's PITR. > 3.) make pgpool-II high available > --------------------------------- > to get rid of a single-point-of-failure I'm thinking of installing > pgpool-II on both database-nodes and let heartbeat check the > availability of the "main" pgpool-II installation and switch the virtual > IP in the case of a failure. It is "THE" way to go in my opinion. It is the way I've implemented it and it's working flawlessly here. I wrote a tutorial on it (in Spanish) which is ready to be published. As soon as my editor publishes it on paper I'll upload it to my website and post a link here. If you can read Spanish, I can forward the relevant parts to you. -- Jaume Sabater http://linuxsilo.net/ "Ubi sapientas ibi libertas" _______________________________________________ Pgpool-general mailing list [email protected] http://pgfoundry.org/mailman/listinfo/pgpool-general
