Few questions on MySQL Replication (1 reply)

Dear Sir/Madam,

I have few questions on MySQL Replication.

1. Skip table in replication process
Says I have a `TEMP` table and has to be moved to `ACTUAL` table after the update process. What will be if I skip the `TEMP` in replication process? Will it generate 'table not exists' error?
Sample Process:
INSERT INTO `TEMP`;
UPDATE `TEMP` BLA BLA;
INSERT INTO `ACTUAL` SELECT FROM `TEMP`;

replicate-ignore=TEMP


2. There are many paper suggest 'MASTER TO SLAVE' replication. WRITE/UPDATE/DELETE on master, READ on slave.
I am wondering how do we perform this kind of configuration instead of identify in our coding, says WRITE/DELETE call master db, READ call slave db in a script.
I feel this will be very confusing and hard to determine.
Can I say, this kind of configuration only applicable in blogs? Or some other Web application that can differentiate READ & WRITE/UPDATE clearly?

3. As I understand that it is not recommended to have replication setup in different MySQL versions.
We are not recommended to have replication setup in between 5.1.45 and 5.5.12.
What if to have replication setup in between 5.1.45 and 5.1.61 (where the last version digit may differ)?

4. Would the INSERT ... SELECT query slow down the replication process?


Thank you!