Replication: specific ?s on failed transmissions

From: Alex Sokol <asokol_at_bcgi.net>
Date: 10 Dec 2001 11:13:53 -0800
Message-ID: <c2f336c7.0112101113.3b8a00db@posting.google.com>

I'm looking into a possibility of setting up multi master and multi slave (is there such thing as multi slave?) databases all running 8.1.7 on Sun Solaris 2.6. I couldn't locate any info on where server stores info about transactions that didn't make it for targets that were down (would that be some tables created by catrep*.sql scripts (REPCAT$_) and how it re-submits failed ones. I guess for LOG based replication it would be in the archived redo logs, but what about snapshot based?

Also, what are the limits on # of masters and/or slaves?

Any suggestions are appreciated.
Thanks, Alex Received on Mon Dec 10 2001 - 13:13:53 CST