Comments on: MySQL: Using CONNECT to Quickly Verify Replication Health

By: Phil

Phil — Wed, 02 Feb 2011 16:26:28 +0000

chetanM > Thank you for that. Personnaly, I use “Exec_Master_Log_Pos” instead of “Read_Master_Log_Pos” (our tests have shown that Read_Master_Log_Pos isn’t what we want to detect loss of replication!)

By: chetanM

chetanM — Mon, 09 Nov 2009 23:01:26 +0000

slave_pos=”`mysql -uroot -ppasswd -e ‘show slave status \G;’ |grep -r ‘Read_Master_Log_Pos’ |awk ‘{ printf ” %s “, $2 }’ |xargs`”

master_pos=”` mysql -uroot -ppasswd -e ‘show master status;’ | grep mysql-bin |awk ‘{ printf ” %s “, $2 }’ |xargs`”

By: Gerry Narvaja

Gerry Narvaja — Thu, 29 Jan 2009 21:15:37 +0000

Pretty concept. I guess that the awk and perl experts can take it and add some warning messages based on the info. Thanks for sharing.

By: Stefan Hinz

Stefan Hinz — Wed, 28 Jan 2009 11:10:36 +0000

For my small replication network, I’m using this shell script:

#! /bin/sh
echo “Master servers ————————————————–”
for host in themis iconnect2; do
echo $host
mysql -h $host -e “show master status\G” | grep Position
done
echo “Slave servers —————————————————”
for host in iconnect2 artemis astraia athena aphrodite hermes; do
echo $host
mysql -h $host -e “show slave status\G” | grep Exec_Master_Log_Pos
done

The output is pretty concise and straightforward. :-)