Re: Slow Slave Replication Catchup

[Simon]

Hi All,

Thanks for all the great replies! Its taken me a bit to get organsied to reply. :)

As far as percona backup, yep we are using that for our larger databases, but these
particular master>slave servers are our primary web hosting servers. In this case we
have a huge mix of all types of databases and we need to run a mysqldump of each
individual database so that we can provide them to our customers as downloadable
archives. 

So.. i changed from "stop slave" to "STOP SLAVE SQL_THREAD" (as per
http://dev.mysql.com/doc/refman/5.1/en/replication-solutions-backups-mysqldump.html). The
backup started at 8:08am, and completed at approx 9:32am... approx 2 mins later the slave
recovered! 

http://www.nzlocal.com/simon/mysqlbackup.txt

Has changing from "stop slave" to "STOP SLAVE SQL_THREAD" made such a difference?

Simon