TechPrufer
Get started
Airbnb
Airbnb logoAirbnb · Feature trail

Database Traffic Capture and Replay

1 post · Oct 2026

What this trail covers

This post introduces a new system for capturing and replaying real production database workloads. This system addresses challenges in load testing, de-risking database upgrades and migrations, and performance debugging by providing a client-agnostic capture point using ProxySQL. It includes components for log moving, processing (parsing, grouping, timestamp-based bucketing, query rewriting for auto-increment consistency), and replaying traffic in 'replay only' (load testing) or 'replay and compare' (compatibility testing) modes. The replay system ensures accurate timing and concurrency by using a shared start signal for tasks.

All Airbnb trails
  1. Beyond synthetic testing: Capturing and replaying real database workloads at AirbnbThis post details the design and implementation of a novel database traffic capture and replay system at Airbnb. Key technical contributions include: 1. Leveraging ProxySQL as a client-agnostic capture point without application changes. 2. A Log Mover sidecar for efficient log transfer to object storage. 3. A Log Processor that parses binary logs, groups by backend, reassembles transactions, buckets by time window, and rewrites INSERT statements to pin auto-increment IDs for compatibility testing. 4. A Log Replayer with 'replay only' and 'replay and compare' modes, featuring an API server, task scheduler, and workers that use a shared start signal to reproduce original timing and concurrency.