· 2 min · PostgreSQL

All your GUCs in a row: allow_in_place_tablespaces

allow_in_place_tablespaces This obscure PostgreSQL setting exists only for the test suite to simulate replication on a single machine. You will never need it in production.

0 comments

· 6 min · PostgreSQL

Hints, Part 3: Advice, Not Orders

PostgreSQL 19 proposes `pg_plan_advice`: a pluggable plan-advice mechanism that lets you capture plan shapes via EXPLAIN and apply them without touching SQL.

0 comments

· 2 min · PostgreSQL

All your GUCs in a row: allow_alter_system

PostgreSQL 17 added allow_alter_system to prevent accidental configuration drift when superusers reach for ALTER SYSTEM instead of your configuration…

0 comments

· 5 min · PostgreSQL

Hints, Part 2: Features We Do Not Want

PostgreSQL has officially rejected Oracle-style hints for two decades.

0 comments

· 4 min · PostgreSQL

Hints, Part 1: The State of the Art Everywhere But Here

Query hints exist in every major database except PostgreSQL — until now.

0 comments

· 1 min

At last.

After far too long, this blog’s conversion to PostgreSQL and Django is complete.

0 comments

· 3 min · PostgreSQL

Lies, Damn Lies, and LLM Output.

A viral Medium article claims PostgreSQL 18 is revolutionary with ten incredible new features.

3 comments

· 1 min · PostgreSQL

Do not expose port 5432 to the public Internet

Exposing PostgreSQL port 5432 to the public Internet invites brute-force attacks, DoS vulnerabilities, and potential data corruption.

0 comments

· 2 min · PostgreSQL

VACUUM (INDEX_CLEANUP OFF) Considered Harmful

PostgreSQL 12's INDEX_CLEANUP option can tank your query performance if misused. Learn when you absolutely need it, and why you almost certainly don't.

0 comments

· 1 min · Python

.pyc files and their discontents

Stale .pyc files from deleted Python modules can cause mysterious bugs in production. Always clean them out during deploys.

0 comments