MySQL replication is based on the master server keeping track of all changes to your database (updates, deletes, etc) in the binary log (see Binary log). Each slave server receives from the master the saved queries that the master has recorded in its binary log, so that the slave can execute the same queries on its copy of the data.
It is very important to realize that the binary log is simply a record starting from a fixed point in time (the moment you enable binary logging). Any slaves that you set up will need copies of the databases on your master as they existed at the moment you enabled binary logging on the master. If you start your slaves with data that is not the same as what was on the master when the binary log was started, your slaves may fail.
Starting from 4.0.0, you can use LOAD DATA FROM MASTER to set up a slave. Be aware that LOAD DATA FROM MASTER currently works only if all the tables on the master are MyISAM type. Also, this statement acquires a global read lock, so no writes are possible while the tables are being transferred from the master. When we implement lock-free hot table backup (in MySQL 5.0), this global read lock will no longer be necessary.
Due to these limitations, we recommend that at this point you use LOAD DATA FROM MASTER only if the dataset on the master is relatively small, or if a prolonged read lock on the master is acceptable. While the actual speed of LOAD DATA FROM MASTER may vary from system to system, a good rule of thumb for how long it is going to take is 1 second per 1 MB of the datafile. You will get close to the estimate if both master and slave are equivalent to 700 MHz Pentium and are connected through a 100 MBit/s network. Note that this is only a rough estimate.
Once a slave is properly configured and running, it will simply connect to the master and wait for updates to process. If the master goes away or the slave loses connectivity with your master, it will keep trying to connect periodically until it is able to reconnect and resume listening for updates. The retry interval is controlled by the --master-connect-retry option. The default is 60 seconds.
Each slave keeps track of where it left off. The master server has no knowledge of how many slaves there are or which ones are up-to-date at any given time.