Postgres Summit US2026

Something's wrong with your Postgres database; a query is stuck, the app is timing out, or it's just "slow" and nobody can say why. This session is the diagnostic toolkit for exactly that moment.

We'll cover how to find and safely stop the queries and locks causing trouble, the timeout settings that act as guardrails so it doesn't happen again, and the handful of metrics worth actually watching. Then we'll spend real time on the single most useful skill for understanding why a query is slow: reading Postgres's EXPLAIN output.

You'll see how to read query costs and actual execution times, recognize the difference between a sequential scan and an index scan, watch a CREATE INDEX visibly change a query plan in real time, and walk away with a checklist of common query anti-patterns and how to fix them.

This is a mental model and a checklist, not deep tuning mastery, everything here is meant to be useful the next time something misbehaves, whether or not you've seen EXPLAIN output before.