Overview of Solutions for PostgreSQL High Availability
summary
Overview of PostgreSQL high-availability solutions as of 2019, covering older pre-cloud tools such as pgpool, Bucardo, and Postgres-XC/XtraDB/XL. The talk explains each solution’s shortcomings (and some advantages) and why these older approaches fell out of favor as systems moved toward cloud-era architectures.
AI-written summary · about
chapters (19)
- 0:00Introduction
- 1:06High availability problems
- 3:57Existing solutions overview
- 8:20Cloud solutions evolution
- 9:01Tools overview: Patroni, repmgr, Stolon
- 12:12Solution architecture features
- 15:28Patroni vs repmgr
- 17:17Stolon vs others
- 19:46Installation options
- 21:52Cluster config management
- 26:54Synchronous replication setup
- 32:00Incident logging and diagnostics
- 33:03Automatic failover
- 34:32Event handling customization
- 36:05Split-brain and fencing
- 39:33Failed node recovery
- 41:15App prep for failover
- 43:18Automation vs control
- 44:44Q&A
related talks
50mEvolution Of PostgreSQL Replication and High Availability (upto ~ 99.999%)
47mPostgreSQL High Availability: The Considerations and the Candidates
46mDeploying Maximum HA Architecture With Postgres
36mDesigning Layered High‑Availability Architectures for PostgreSQL on a Budget
50mMoving from MSSQL to pgSQL – High Availability and Disaster Recovery
55mHigh Availability with PostgreSQL and Pacemaker
58mDatabase High Availability
48mChoosing a Replication System for PostgreSQL