Running MySQL on K8s. Deployment, Scaling, Backups and advanced configuration
summary
Deployment uses the Percona Operator for MySQL to create a reproducible 3-node cluster via kubectl and YAML, with HAProxy default or optional ProxySQL frontends and PMM integration for monitoring. Backups use xtrabackup for physical snapshots plus binlog streaming for point-in-time recovery to S3 (recommended) and support scheduled jobs. Scaling is done by patching the CRD to change pod counts or resources; rebuilding large multi‑TB nodes triggers lengthy resyncs and may make huge datasets impractical without evaluation.
AI-written summary · about
chapters (10)
- 0:00MySQL on Kubernetes intro
- 1:04Why MySQL with K8s
- 2:30Deploying MySQL with operators
- 6:04Percona MySQL Operator features
- 7:52Prerequisites and steps
- 9:36Backups and services config
- 12:51Scaling MySQL clusters
- 15:21Backup strategies
- 17:08Advanced config options
- 20:38Q&A replication & large DBs
related talks
56mDeploying MySQL on Kubernetes & Openshift
41mDemocratizing MySQL: From Cloud Managed to Kubernetes Managed
35mRunning MySQL on Kubernetes [I]
36mCreate your Own MySQL-as-a-Service that Runs Anywhere Using Kubernetes Operators
27mDatabase Infrastructure with Open Source Kubernetes Database Operators
27mDIY DBaaS with Percona Kubernetes Operators
31mOpen Source Databases on Kubernetes: Best Practices
53mMySQL in the Cloud: DBaaS vs Kubernetes