Neues vom PostgreSQL Planet

Say goodbye to the old way of distributing Postgres extensions as part of the main pre-built operand image. Leveraging the Kubernetes ImageVolume feature, CloudNativePG now allows you to mount extensions like pgvector and PostGIS from separate, dedicated images. This new declarative method completely decouples the PostgreSQL core from the extension binaries, enabling dynamic addition, easier evaluation, and simplified updates without ever having to build or manage monolithic custom container images.

 

This blog provides my opinion on how Postgres, according to the annual Stack Overflow developer survey, became the most admired, desired and used database by developers. This is part three in my series about the Postgres journey to the top with developers. Here are the first parts of this story:

 

And our last 2025 recording is here! Check out Jay Miller’s talk!

PostgreSQL 19 changes this week debug_exec_backend GUC to show EXEC_BACKEND state pg_buffercache view pg_buffercache_os_pages added following functions added: pg_buffercache_mark_dirty() pg_buffercache_mark_dirty_relation() pg_buffercache_mark_dirty_all() pg_replication_slots column slotsync_skip_reason added pg_stat_replication_slots columns slotsync_skip_count and slotsync_skip_at added Planner will now replace COUNT(ANY) with COUNT(*) , when possible PostgreSQL 19 articles Teaching Query Planner to See Inside C Functions (2025-11-27) - Robins Tharakan Settling COUNT(*) vs COUNT(1) debate in

Seattle Postgres Users Group (SEAPUG) maintained the PostgreSQL booth at PASS Data Community Summit 2025 from November 17-21, 2025:

• Lloyd Albin
• Jeremy Schneider
• Harry Pierson
• Ben Chobot
• Deon Gill
• Rick Lowe
• Pavlo Golub

Speakers:

PASS Data Community Summit 2025 wrapped up last week. This conference originated 25 years ago with the independent, user-led, not-for-profit “Professional Association for SQL Server (PASS)” and the annual summit in Seattle continues to attract thousands of database professionals each year. After the pandemic it was reorganized and broadened as a “Data Community” event, including a Postgres track.

Introduction

Recently, I read an article titled "Alias Settings of Engineers Around the World" (in Japanese). As a PostgreSQL engineer, this got me thinking: "If they are customizing their bash aliases, how are they configuring their psql environments?"

Driven by curiosity, I decided to investigate GitHub repositories to see how developers commonly configure their psqlrc files.

Recently(ish) I had a conversation on one of PostgreSQL support chats (IRC, Slack, or Discord) about efficient storage of JSON data, which compression to use, which datatype. Unrelated to this, some people (at least two over the last year or so) said that they aren't sure if PostgreSQL doesn't optimize storage between columns, for example, … Continue reading "Using JSON: json vs. jsonb, pglz vs. lz4, key optimization, parsing speed?"

The Postgres planner just got a new superpower: X-ray vision for your black-box functions. For years, Postgres has been able to "inline" simple SQL functions. If you wrote CREATE FUNCTION ... LANGUAGE SQL AS 'SELECT ...', the planner would often rip out the function call and paste the raw SQL directly into the main query. This was magic. It meant that WHERE clauses could be pushed down, indexes

Back in October, before PGConf.EU, I explained the issues impacting the prolonged wait for TDE in PostgreSQL 18. Explanations were needed as users were buzzing with anticipation, and they deserved to understand what caused the delays and what the roadmap looked like.

A recent commit to the PostgreSQL master branch brings a nice quality-of-life optimization for a very common SQL pattern - improving performance by up to 64% for SELECT COUNT(h) where h is a NOT NULL column. If you've ever wondered whether you should use COUNT(*) or COUNT(1), or if you've been dutifully using COUNT(id) on a non-null column, this change is for you. Note: This feature is currently