How do I calculate the ideal database pool size?

A common starting formula is (concurrent requests / average queries per request) * 1.5, while ensuring the total count across all app instances doesn't exceed the database server's max_connections.

Why should I reuse an aiohttp ClientSession?

Reusing a ClientSession allows the application to keep connections open (keep-alive) and pool them, significantly reducing the overhead of creating new TCP connections for every request.

What is the benefit of pool_pre_ping in SQLAlchemy?

It performs a 'pessimistic' check to ensure a connection is still valid before using it, preventing 'connection closed' errors at the cost of a small amount of latency.

How do I detect connection leaks?

The skill provides SQLAlchemy event listeners to log checkouts and checkins, helping you identify code paths where connections are not being returned to the pool.

Can I use this skill with FastAPI?

Yes, it includes specific patterns for using FastAPI's lifespan events to initialize pools on startup and gracefully close them during shutdown.

Connection Pooling for Python

Name: Connection Pooling for Python
Author: yonatangross

byyonatangross

•

Desarrollo de API

Optimizes database and HTTP performance using production-ready async connection pooling patterns for Python applications.

This skill provides specialized guidance for implementing and managing connection pools in high-concurrency Python environments. It covers critical configurations for SQLAlchemy, asyncpg, and aiohttp, focusing on preventing resource exhaustion, handling stale connections, and implementing proper lifecycle management within FastAPI frameworks. By applying these patterns, developers can ensure system stability, reduce latency, and scale backend services effectively while avoiding common anti-patterns like per-request session creation.

Características Principales

01Prometheus-based monitoring and troubleshooting patterns

02Aiohttp ClientSession lifecycle management

0369 GitHub stars

04Asyncpg and SQLAlchemy connection pool configuration

05Automatic health checks and connection validation (pre-ping)

06Pool sizing formulas for high-concurrency workloads

Casos de Uso

01Configuring a high-performance database layer for a FastAPI service

02Diagnosing and fixing 'QueuePool limit reached' or connection leak errors

03Optimizing external API calling efficiency with reused HTTP sessions

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add yonatangross/orchestkit connection-pooling

For use in Claude.ai and ChatGPT

Download Skill