Source: pg-stat-plans Section: database Priority: optional Maintainer: Debian PostgreSQL Maintainers Uploaders: Christoph Berg , Build-Depends: architecture-is-64-bit , debhelper-compat (= 13), libkrb5-dev, libzstd-dev, postgresql-all , postgresql-server-dev-all (>= 217~), Standards-Version: 4.7.2 Rules-Requires-Root: no Homepage: https://github.com/pganalyze/pg_stat_plans Vcs-Browser: https://salsa.debian.org/postgresql/pg-stat-plans Vcs-Git: https://salsa.debian.org/postgresql/pg-stat-plans.git Package: postgresql-18-pg-stat-plans Architecture: any Depends: ${misc:Depends}, ${postgresql:Depends}, ${shlibs:Depends}, Description: Track per-plan call counts, execution times and EXPLAIN texts in PostgreSQL pg_stat_plans is designed for low overhead tracking of aggregate plan statistics in Postgres, by relying on hashing the plan tree with a plan ID calculation. It aims to help identify plan regressions, and get an example plan for each Postgres query run, slow and fast. Additionally, it allows showing the plan for a currently running query. . Plan texts are stored in shared memory for efficiency reasons (instead of a local file), with support for zstd compression to compress large plan texts. . Plans have the same plan IDs when they have the same "plan shape", which intends to match EXPLAIN (COSTS OFF). This extension is optimized for tracking changes in plan shape, but does not aim to track execution statistics for plans, like auto_explain can do for outliers.