Heavy Duty Backup w/PgBackRest
summary
pgBackRest targets very large Postgres installs and delivers parallel backups, efficient incrementals that can resume after failure, and asynchronous archive shipping instead of rsync. It uses a custom remote protocol with in‑stream compression and per‑thread SSH, stores each backup as a usable Postgres data directory (so a snapshot can be mounted), enforces retention/expiration automatically, and reports missing archive logs. Production use shows a 5.7 GB full backup in ~12 seconds with compression off. A restore command is pending but filesystem restores and ZFS snapshots work now.
AI-written summary · about
chapters (16)
- 0:00Introduction to pgBackRest
- 1:18Backup definitions
- 2:51Motivation for backups
- 4:37PostgreSQL backup methods
- 6:36Design of pgBackRest
- 7:41Technical architecture
- 10:20Archiving and remote protocol
- 12:48Backup structure and integrity
- 15:31Advanced backup features
- 17:50Live demo setup
- 21:20Performing a backup
- 26:33Incremental backups
- 30:40Configuration and command line
- 33:34Installation and multi-host
- 37:17Restoring and best practices
- 41:28Project status and Q&A
