Flinn Mueller on 18 Jan 2008 20:07:52 -0800 |
Whatever you find out will you post it back to this list? I spent a while investigating a similar setup a while back then never implemented but I may need to revisit this in the near future. In short I liked pgcluster's multi-master setup and didn't like slony's single master multi slave setup but that's a pretty superficial analysis. On Jan 18, 2008, at 10:51 PM, Toby DiPasquale wrote: Hi all, Sorry in advance for the crosspost. I know its bad netiquette but I wanted to hit the best in Philly with one fell swoop. We're having an issue with some code we picked up that's tied to PostgresSQL. This code currently does not run any kind of replication for its Postgres instance which we find to be unacceptable for what would seem to be obvious reasons. The kind of replication we'd like to do is master-slave asynchronous replication, which is quite simple using MySQL as replication is built in. We've investigated some of the options for Postgres replication (in particular Slony, but most others are pretty old) but the best we can come up with so far is master-slave *synchronous* backup. We'd rather not have to stop the master while we are backing up to the slave if possible. Is anyone out there using PostgresSQL in a production environment using master-slave asynchronous replication? If so, what package/setup are you using? If anyone's got any info on this I'd really appreciate it. Thanks in advance :) -- Toby DiPasquale _______________________________________________ To unsubscribe or change your settings, visit: http://lists.phillyonrails.org/mailman/listinfo/talk _______________________________________________ To unsubscribe or change your settings, visit: http://lists.phillyonrails.org/mailman/listinfo/talk
|
|