Setting up Circular replication in existing replication infrastructure

[Phil Marsay]

I have the following MySQL replication topology set up for fault tolerance/recovery with master server A and slaves B,C:

A -> B
A -> C

Both B and C are fully up to date with everything from A and have nothing in addition to that (i.e. there is nothing in B,C that is not in A).

I would like to introduce some circular replication such that:
A <-> B
A -> C

So changes made in B would replicate back to server A. The reason for this is to allow me to get more use out of a secondary server by spreading the load over two primary servers.

Primary use of each individual database on the servers would be limited to a specific server. So, e.g. Users of DB "bookingsystem" always access via server B, users of DB "enquirysystem" always use server A. So there is no danger of having to deal with changes on the same DB in two different locations. The replication is purely for fault tolerance should there be a problem with either server.

So, my question is, can I simply set server A to slave from server B and have this working straight away, or is there further preparatory, or subsequent work needed?

Thanks!

Phil