Moving from MSSQL to pgSQL – High Availability and Disaster Recovery
summary
Compares SQL Server patterns to concrete PostgreSQL equivalents and pitfalls: streaming replication provides synchronous or asynchronous replicas like mirroring but Postgres has no built‑in listener or automatic failover; WAL shipping (archive_timeout exists) gives DR but can leave hours of data if a WAL file hasn’t closed; replication slots force the primary to retain WAL and can bloat storage; pg_rewind can repair split‑brain after a bad manual promote; common failover tools are pg_auto_failover and Patroni (repmgr and Pgpool-II are alternatives).
AI-written summary · about
chapters (12)
- 0:00Introduction
- 2:09SQL Server HA options
- 6:03Failover clusters in SQL Server
- 12:45SQL Server vs PostgreSQL
- 15:57PostgreSQL HA tools
- 21:50Streaming replication for HA
- 27:26Log shipping vs streaming replication
- 32:08Failover automation in PostgreSQL
- 41:02Promoting standby vs failover
- 44:39HA pitfalls in PostgreSQL
- 47:14PostgreSQL failover extensions
- 49:42PostgreSQL HA SaaS options
related talks
22mSecuring Postgres with streaming replication
55mHigh Availability with PostgreSQL and Pacemaker
50mEvolution Of PostgreSQL Replication and High Availability (upto ~ 99.999%)
49mPostgreSQL Replication - The Most Exciting Technology on Earth
02h 26mTutorial
48mChoosing a Replication System for PostgreSQL
52mNew Data Replication Technologies in PostgreSQL
46mDeploying Maximum HA Architecture With Postgres