Everything You Ever Wanted to Know About MariaDB Replication but Were Afraid to Ask
summary
Overview of concrete replication setup and repair: demos build a three-source topology (DrApp ~1 txn/sec, Pandrea ticketing, Combi aggregator), use GTID positions and binary/relay logs, create users, and run MariaDB dump and MariaBackup to seed replicas, then start slaves with change master scripts. One table has ~129,000 rows; a broken replica missing rows 13813–13836 is repaired by dumping those rows as INSERTs and replaying them. All nodes resynchronize and continue growing; optional routing via MaxScale is shown.
AI-written summary · about
chapters (9)
- 0:00Intro
- 0:28What is Replication?
- 1:453 types of replication
- 3:14Key Concept: GTID
- 4:02Key Concept: Binary and Relay Logs
- 11:40Key Concept: Physical/Logical Copies
- 12:15Demo Setup
- 12:43Demo
- 28:42Review and Summary
related talks
50mMySQL/MariaDB Parallel Replication: inventory, use cases and limitations
51mReplication latest features: the community sceptic roundup
49mMySQL Replication — Advanced Features
40mMySQL at Scale: Booking MariaDB to MySQL Migration
58mDatabase High Availability
53mMySQL replication setup, benefits, limitations
22mMySQL Replication: Latest Developments
14mThe State of the Art of MySQL Replication