Skip to content

Instantly share code, notes, and snippets.

@varun-dhawan

varun-dhawan/AzurePostgresFlexibleServer-PgBouncerMetrics.md

Last active February 24, 2023 05:07
Show Gist options

  • Select an option

    Learn more about clone URLs
  • Save varun-dhawan/e86a6c18066c25fc9ca069503ee680d3 to your computer and use it in GitHub Desktop.

Select an option

Learn more about clone URLs
Save varun-dhawan/e86a6c18066c25fc9ca069503ee680d3 to your computer and use it in GitHub Desktop.
Download ZIP
Raw
AzurePostgresFlexibleServer-PgBouncerMetrics.md
title description ms.service ms.subservice ms.topic ms.author ms.date
Azure Database for PostgreSQL Flexible Server - PgBouncer metrics (Preview)
Metrics to monitor PgBouncer
postgresql
flexible-server
conceptual
varun.dhawan
11/01/2022

Azure Postgres Flexible Server - PgBouncer metrics (Preview)

PgBouncer monitoring

Postgres is a powerful and popular open-source relational database management system (RDBMS). However, as the number of connections to a Postgres database grows, managing and monitoring those connections can become a challenge. That's where connection pooling comes in.

PgBouncer probably the most popular connection pooler in postgrtes. It is a lightweight connection pooler that helps to improve the performance and scalability of your Postgres databases by managing and reusing database connections. By pooling connections, PgBouncer reduces the number of connections that need to be opened and closed, reducing the load on your Postgres server and improving overall performance.

Azure Database for PostgreSQL – Flexible Server offers PgBouncer as a built-in option, wherein customers can easily enable PgBouncer from Azure Postgres server parameters and update the app connection string to talk to PgBouncer on port 6432, rather than directly to the database server.

When using PgBouncer, it's important to monitor the performance of PgBouncer to ensure that it's functioning as expected and to identify any potential issues. Here are some key metrics to monitor when using PgBouncer with Azure Postgres Flexible Server:

List of PgBouncer metrics

Display Name Metrics ID Unit Description Dimension Default enabled
Active client connections (Preview) client_connections_active Count Connections from clients which are associated with a PostgreSQL connection DatabaseName No
Waiting client connections (Preview) client_connections_waiting Count Connections from clients that are waiting for a PostgreSQL connection to service them DatabaseName No
Active server connections (Preview) server_connections_active Count Connections to PostgreSQL that are in use by a client connection DatabaseName No
Idle server connections (Preview) server_connections_idle Count Connections to PostgreSQL that are idle, ready to service a new client connection DatabaseName No
Total pooled connections (Preview) total_pooled_connections Count Current number of pooled connections DatabaseName No
Number of connection pools (Preview) num_pools Count Total number of connection pools DatabaseName No

Enabling PgBouncer metrics

  • Most of these new metrics are disabled by default
  • To enable these metrics, please turn ON the server parameter metrics.pgbouncer_diagnostics
    • This parameter is dynamic, and will not require instance restart.
  • For Pgbouncer metrics to work, the pgbouncer.enabled and metrics.pgbouncer_diagnostics have to be enabled

References

Team

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment