Running MySQL on Kubernetes [I]
summary
Describes how to run MySQL on Kubernetes using StatefulSets, PVC templates and an MySQL operator that deploys a highly available Group Replication cluster with one kubectl command, watches CRDs, and does scheduled or on‑demand backups to object storage (backup appears as a .tar.gz). Shows a working demo: creates a 3‑node cluster, restores from backup, and performs Vitess operations—creates two shards, applies VSchema, scales one shard from 2→3 replicas, deletes a statefulset and master, and observes automatic promotion under Vitess sharding.
AI-written summary · about
chapters (25)
- 0:00Introduction
- 1:02Oracle Dyne
- 1:52Agenda
- 2:44MySQL and Kubernetes
- 3:40Database vs Kubernetes
- 4:29Ingredients
- 5:19Stateful Sets
- 6:26Deployment Replica Sets
- 7:07Deployment Stateful Sets
- 7:32Operators
- 9:29MySQL Operator Example
- 10:28MySQL Operator Overview
- 11:20MySQL Operator Features
- 11:57Creating a Cluster
- 13:14Demo
- 18:48The Test
- 20:18The Test Components
- 22:55Cross Shard Index
- 23:29Repairing
- 24:03Backups
- 24:49Demonstration
- 25:24Test
- 31:48Collect Beats
- 32:16Thank You
- 32:38Experience
related talks
23mRunning MySQL on K8s. Deployment, Scaling, Backups and advanced configuration
36mCreate your Own MySQL-as-a-Service that Runs Anywhere Using Kubernetes Operators
56mDeploying MySQL on Kubernetes & Openshift
31mVitess: Running MySQL with confidence on Kubernetes
35mHow to put a Database in Kubernetes
38mIntro: Vitess
35mBuilding a Database as a Service on Kubernetes
41mDemocratizing MySQL: From Cloud Managed to Kubernetes Managed