What can't pgBackRest do for you?
summary
pgBackRest demonstrates a working asynchronous archiving pipeline that uses multiple processes and a local spool status file to ship WAL segments ahead of streaming replicas (demo showed archiving ahead of pg_receivewal), reducing lost transactions and removing the need for a receiver in many cases. It supports zstd compression for gzip-like space with lz4-like speed, exposes JSON info for monitoring and a check plugin, and offers archive-push-queue-size to avoid PostgreSQL panic. Limits: no single-table in-place restore—use selective restore (DB include) or full restore then pg_dump, and restore intent (timeline/backup choice) must be specified.
AI-written summary · about
chapters (9)
- 0:00Introduction to pgBackRest
- 2:57Archiving: Command vs Receiver
- 7:05Timelines and recovery
- 11:05Optimizing archiving performance
- 18:06Troubleshooting and monitoring
- 24:10Logging and repository insights
- 31:28Advanced recovery techniques
- 36:41Selective restores and conclusions
- 40:17Q&A session
related talks
51mHeavy Duty Backup w/PgBackRest
41mpgBackRest: Elevating PostgreSQL Backup Game
56mPostgreSQL Backup Strategies
56mI Dropped My Database! Now What? - A Dive Into PostgreSQL Backup Using pgBackRest
50mPostgreSQL Backup and Recovery Methods
22mPostgreSQL Disaster Recovery Fundamentals
43mPostgreSQL 17 and beyond
49mIncremental Backup