> I was just now reading through the homepage slides, which include an > intro to pgpool-II 3.0's support for 9.0 replication. I noticed > there's a monitoring tool that uses polls the standby's > pg_last_xlog_receive_location() and alerts if it's too far behind > pg_current_xlog_location(). Could a similar approach be used for a > simple implementation of synchronous replication? Any major pitfalls > here (besides the need to balance polling frequency and latency)?
It's definitely possible by using similar approach used in pgpool-II 3.0. Only a pitfall I can think of is, in a busy sytem you would never be able to send query to standbys since in such a system standby will not likely catch up the primary usually. -- Tatsuo Ishii SRA OSS, Inc. Japan English: http://www.sraoss.co.jp/index_en.php Japanese: http://www.sraoss.co.jp > (I realize that synchronous replication is in the works for 9.1.) > > Thanks in advance. > -- > Yang Zhang > http://yz.mit.edu/ > _______________________________________________ > Pgpool-general mailing list > [email protected] > http://pgfoundry.org/mailman/listinfo/pgpool-general _______________________________________________ Pgpool-general mailing list [email protected] http://pgfoundry.org/mailman/listinfo/pgpool-general
