Decentralized Cluster Nodes
summary
ProxySQL provides a native clustering system that actually synchronizes concrete configuration tables—MySQL users, servers, query rules and variables—across nodes using checksum + timestamp validation and requires identical versions before peering. It supports auto-discovery and a writable core + satellite scale to thousands model so a few core nodes propagate config to many satellites, registers clients with persistent UIDs for monitoring, avoids a strict leader by design, and adds safeguards against race conditions from DNS, load‑balancers or concurrent config tools.
AI-written summary · about
chapters (18)
- 0:00Intro
- 2:32Main features
- 7:44ProxySQL Layer Deployment
- 9:54Distributed system
- 11:18Configuration Management
- 14:07Puppet
- 19:22ProxySQL Cluster Topologies
- 21:26Cluster core nodes
- 22:57What to synchronize?
- 31:36Poor design patterns
- 34:00Variables Synchronization
- 35:34Core/Satellites Layers
- 37:34Satellite nodes registration
- 40:43Why nodes registration?
- 42:11Todo: Leader election
- 45:59Synchronization slowdown
- 46:58Avoid config rollback
- 53:08Prevent incorrect sync
related talks
27mThe State of ProxySQL, 2020 Edition
50m#ProxySQL Use Case Scenarios
55mProxySQL Use Case Scenarios
55mProduction with ProxySQL and Orchestrator
37mScale MySQL beyond limits with ProxySQL
08mMySQL as a Layered Service: How to use Proxy SQL to Control Traffic and Scale Out
41mProxySQL 2021 Dev Submit
33mMySQL Cluster as Transactional NoSQL