As of 2025-Dec, you can choose to use PGSQL and/or Prometheus for the data sink of the metrics, and Grafana dashboards for both cases are provided
GH feature request: Need official Grafana dashboards for the coroot pg agent #7
GH issue: Metrics name conflict overlaps with prometheus-community/postgres_exporter #22
Pg-agent – a Postgres exporter for Prometheus focusing on query performance statistics
Missing metrics required to gain visibility into Postgres performance
Not designed as an exporter, output format: human, json or csv
But commercial product https://pgdash.io/ is designed for using pgmetrics
Percona pg_stat_monitor (an extension), not compatible/tested on AWS RDS
https://github.com/percona/pg_stat_monitor
Percon’s pg_stat_monitor is not in AWS supported list of modules (as of 2025-Dec)
https://docs.percona.com/pg-stat-monitor/index.html#availability
pgexporter and pgexporter extension
https://github.com/pgexporter/pgexporter
As of 2025-Dec, no official container images are provided but instruction to build it is provided
PostgreSQL Snapper - The PostgreSQL Snapper tool enables periodic collection (snapping) of PostgreSQL performance related statistics and metrics. The config file used by the tool can be customized to add and remove database dictionary views and queries to be snapped as required. Snapper collects and stores the PostgreSQL database metrics in separate OS level files to have minimal impact on the database. These files can be loaded into another PostgreSQL instance by the loader script for doing analysis. In this lab, you will configure Snapper to collect PostgreSQL statistics every 1 minute, generate load on Aurora PostgreSQL using pgbench, package and import Snapper Output to a PostgreSQL database for analysis and then derive insights from the PostgreSQL statistics data using sample SQLs included in the tool.