Presented by:

Shaun Thomas
Shaun has spent decades working in the Postgres ecosystem, specializing in architecture and high availability. His "PostgreSQL High Availability Cookbook" serves as a treatise to the lessons he learned over that time. Perhaps you've read something from his PG Phriday blog series over the years?
Currently he serves as a Senior Software Engineer and Postgres SME at pgEdge, striving to help make Postgres the distributed cluster-aware platform he knows it can be!
No video of the event yet, sorry!
Postgres major version upgrades are a lot of work! Before you start reaching for that handy pg_upgrade utility and accepting the inevitable downtime as part of the equation, let us describe a Better Way™. You didn’t set up an entire High Availability cluster just to shut it down for an upgrade, did you? Of course not!
Postgres 10 introduced native logical replication, and in doing so, secretly brought fully-online upgrades to the platform. But this comes at a cost: it requires a bit of expertise and finesse. Luckily you’re at a Postgres conference! It’s slightly more complicated than setting up a publication and subscription, in ways that may surprise and confound.
This talk will discuss all of the necessary steps to upgrade a sample 3-node Postgres cluster from version 16 to 17. From initial creation of a new cluster and efficiently copying the data, to throwing the migration switch and declaring the operation successful, we’ll mock up the entire procedure. The best part is that these steps work with any version of Postgres after version 10.
Don’t get stuck on an EOL version of Postgres due to HA concerns. Upgrade today!
- Date:
- 2025 March 20 13:30 EDT
- Duration:
- 50 min
- Room:
- Space Coast 1&2
- Conference:
- Postgres Conference 2025
- Language:
- Track:
- Ops
- Difficulty:
- Medium