There are many uses in which this is a non-issue. For example, I use the master-slave relationship for logging. All logs are written to the master, log queries go to the slave. When someone runs a report against the logs, they might get data that's 1-2 seconds behind maximum, and that doesn't matter.
That said, typically on modern machines, replication is almost always less than a second behind.
----- Original Message ----- From: "Fagyal Csongor" <[EMAIL PROTECTED]>
To: "Mysql (E-mail)" <mysql@lists.mysql.com>
Sent: Sunday, April 24, 2005 3:42 PM
Subject: Replication - is there a "server lag"?
Hi,
I am new to replication so excuse me if my question is stupid.
The manual recommends that a nice scenario to take advantage of replication in MySQL is to send all updating queries to the master server, and reading from the slave. I would like to use this setup (as usual, I have many more selects than inserts/updates) but I am a little concerned what happens if the slave is behind the master in updating its DB.
Say I do like this: 1. update something set `a`=1 where c=d (using the master server) 2. update something set `a`=2 where c=d (using the master server) and then immediately 3. select `a` from something where c=d (using the slave)
What if #3 fetches the value of `a` from the slave before `a`=2 takes place? Is it possible that I get `a`==1? Or does replication take care of that?
Other than that: does anybody here have a Nagios script that checks if replication is running O.K.? :-)
Thanks, - Csongor
-- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED]
-- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED]
