Clustering for MySQL DBAs
MySQL clustering means running several MySQL servers as one highly available group, using Group Replication for consensus based replication and InnoDB Cluster to combine it with MySQL Shell and MySQL Router. Clients keep working when a member fails because another member takes over.
Stage: High Availability · Part of the MySQL DBA roadmap · Last reviewed 19 November 2026
What to learn in Clustering
- Create an InnoDB Cluster with MySQL Shell dba.createCluster
- Explain single-primary versus multi-primary mode in Group Replication
- Route application traffic through MySQL Router
- State how many members a group needs to tolerate failures
- Recover a cluster that lost quorum
Clustering interview questions
How does Group Replication keep members consistent?
Transactions are certified by the group using a Paxos based protocol: a majority of members must agree before a commit is applied. Conflicting writes are detected during certification and the later one is rolled back.
What does MySQL Router do in InnoDB Cluster?
MySQL Router is a lightweight proxy that reads cluster metadata and routes client connections, sending writes to the primary and reads to secondaries. When the primary changes, the router follows without the application changing its connection settings.
Where Clustering fits in the MySQL DBA roadmap
Clustering belongs to the High Availability stage (advanced level). Scale and survive.
Other units in this stage
Back to the full MySQL DBA roadmap