MySQL Replication

MySQL Replication is the built-in mechanism by which a MySQL server (the master) streams changes to one or more replicas, enabling read scaling, failover, and geographic distribution. The replication stream is derived from the Binary Log (binlog), a logical log of statements or row changes that the master writes and replicas consume. Despite being the default replication architecture for the world's most popular database, MySQL replication has well-known consistency weaknesses that its documentation treats as features.

The core problem is asynchrony: by default, a transaction committed on the master is not guaranteed to have reached any replica. If the master fails immediately after commit, the transaction is lost — a failure mode known as replication lag that has caused production outages at every scale. Semi-synchronous replication mitigates this by waiting for at least one replica to acknowledge receipt, but this increases commit latency and does not eliminate the possibility of divergent states.

MySQL's replication model prioritizes availability over consistency, a valid systems choice but one that operators often discover only after a failover reveals missing data. It is a reminder that the default configuration of infrastructure encodes a theory of failure, and that the theory is not always the one you would choose.