High Availability, Load-balanced, Replicating MySQL
The what, why, who, when and how of Replication with MySQL utilising High Availability and Load Balancing
I (Oliver Hookins) was fortunate enough to have been selected to give a talk at Linux.Conf.Au 2008 not once but twice on the subject of High Availability, Load-balancing and Replication in the context of MySQL. Here is the slideshow from the talk in OpenOffice Presentation format and a number of helper scripts that I reference in the talk, but were too long or complex to include in the slides.
Script to change a slave into a master, and redirect a slave to a new master
MySQL replication check script, for use with Nagios NRPE and lvs-helper
Example DRBD 0.7.x configuration file for the master/standby
Script to initialise a new slave, using either the master or another slave as the data source
Slightly altered IPaddr2 resource script from Heartbeat R1 to allow usage with tagged VLANs
Example MySQL configuration file for the slave running on the standby master
Slightly altered additional LSB init script for the MySQL slave running on the standby master
Script to run on the master which inserts timestamps for checking replication delay on the slaves
Related sites
http://mysqlmusings.blogspot.com/2007/06/replication-poll-and-our-plans-for.html
http://mysqlha.blogspot.com/2007/11/how-to-keep-mysql-replication-in-sync.html
http://www.xaprb.com/blog/2007/11/08/how-mysql-replication-got-out-of-sync/
http://www.mysqlperformanceblog.com/2007/11/07/filtered-mysql-replication/
http://mysqldump.azundris.com/archives/71-Replication-now-and-then.html
http://blogs.linbit.com/florian/2007/10/01/an-underrated-cluster-admins-companion-dopd/