DBA 3: PostgreSQL Backups and Replication
For PostgreSQL DBAs who already have solid operational experience and want to deepen their expertise in data protection, replication, and high availability strategies.
Upcoming dates
What you'll learn
This course is designed to equip you with the practical skills to manage PostgreSQL backups and replication for high availability and data protection. You'll learn how to use PITR tools, maintain standby servers, set up physical and logical replication, and handle failover procedures — while understanding the risks and challenges of automated failover.
The course covers
- Manage backups with PITR tools
- Maintain one or more standby servers
- Replicate PostgreSQL data both physically and logically
- Understand failover procedures
- Be aware of the pitfalls of automated failover
Prerequisites
You must either have completed DBA 2 (PostgreSQL Advanced) or possess equivalent hands-on experience.
For details see the course terms & conditions.
Syllabus
1Tools for PostgreSQL backups
This chapter begins with a recap of point-in-time recovery (PITR) fundamentals to show why tooling is essential for reliable recovery. It then explores the main tools available for PostgreSQL backups, with a focus on their capabilities, differences, and practical use cases.
- Understanding why backup tooling is essential for reliable recovery
- Overview of Barman, WAL-G, and pgBackRest
- How each tool handles WAL archiving, encryption, compression, and incremental backups
- Comparing features to choose the right fit for your infrastructure
2Streaming Replication in PostgreSQL
This training chapter explains PostgreSQL’s replication architecture, focusing on streaming replication and related physical, logical, and cascading replication concepts.
- Key replication terminology and patterns (asymmetric/symmetric, synchronous/asynchronous)
- How WAL-based replication works, including log shipping and streaming
- Step-by-step streaming replication setup (primary and standby)
- Standby promotion, failover safety, and re-synchronisation
- Monitoring replication health, lag, and conflicts
- Preventing standby dropout with WAL archiving or replication slots
3Logical Replication in PostgreSQL
This chapter introduces PostgreSQL’s logical replication. It explains how logical decoding works, what can and cannot be replicated, and how to configure and operate publications and subscriptions.
- Key differences between physical and logical replication
- Architecture, terminology, and the streaming process
- Setup steps: creating publications and subscriptions
- Supported granularity: by table, schema, row, or operation
- Limitations (DDL, sequences, large objects) and how to work around them
- Monitoring and troubleshooting replication errors
- Using logical replication for major PostgreSQL upgrades
4High Availability and automated failover best practices
This course chapter introduces the principles and best practices of high availability in PostgreSQL. It explains why availability matters, how to set clear objectives, and what compromises often arise between service continuity and data durability.
- Service vs. data availability and their trade-offs
- RPO and RTO as key objectives to define
- Backups, streaming replication, and automated failover as technical means
- Risks of automation, such as split-brain and the added operational complexity following failover
- Anti split-brain mechanisms: fencing, quorum, watchdog
- Overview of major PostgreSQL HA solutions
- Practical introduction to Patroni operations
Frequently asked questions
What prerequisites do I need for DBA3?
You should have completed DBA2 (PostgreSQL Advanced) or have equivalent hands-on experience with PostgreSQL operations, WAL configuration, and basic replication concepts.
Will I receive a certificate?
Yes. Participants who attend all five sessions receive a Certificate of Attendance from Data Egret.
Which backup tools does this course cover?
The course covers pgBackRest, Barman, and WAL-G in depth, including comparison of their features and practical configuration. Our instructor Stefan Fercot is a pgBackRest contributor.
Does the course cover Patroni for high availability?
Yes. The course covers Patroni for automated failover, including its integration with pgBackRest and the practical considerations of automated vs. manual failover.
Are sessions recorded?
No. Sessions are live and interactive. They are not recorded.
Can my company book this for a group?
Yes. Contact us at contact@dataegret.com for group bookings and dedicated sessions.
Instructors
Over 15 years of experience across Oracle, DB2, MySQL, SQL Server and PostgreSQL. A recognised PostgreSQL contributor, active in the community through committees, user groups and international conferences.
A specialist in monitoring, high availability, performance tuning and scalable system architecture, with a background in banking, finance and telecommunications.
Ready to book, or still deciding?
Live, small-group, taught by the people who run PostgreSQL in production. Questions first? Write to us — no sales pitch.