Burrow is a monitoring service for Apache Kafka that checks consumer lag without requiring manual threshold configuration.
Burrow solves the problem of tracking whether Kafka consumers are keeping up with their topics by automatically monitoring committed offsets across all consumer groups and calculating their status on demand. Rather than requiring operators to set fixed lag thresholds, it evaluates consumer health over a sliding window, allowing the system to adapt to normal patterns of lag variation. The tool exposes an HTTP endpoint for querying consumer group status, broker information, and consumer details, and includes configurable notifiers that can send alerts via email or HTTP webhooks when issues are detected.
Burrow suits organizations running Apache Kafka who want automated consumer lag monitoring without the operational burden of tuning thresholds per group. It handles multiple Kafka clusters from a single deployment and automatically discovers all consumers using Kafka-committed offsets, with optional support for Zookeeper-committed and Storm-committed offsets. The HTTP API makes it easy to integrate with existing monitoring and alerting systems.
The project maintains active development with regular updates to its codebase. The build system is kept current with recent Go versions, and the team provides multiple deployment options including Docker and Docker Compose configurations for quick local testing. Documentation is centralized in a project wiki covering configuration details, and the repository includes a docker-compose setup with pre-configured test topics to help new users get started quickly.
