Installer burrow avec Homebrew, apt

Liens du projet

• Page d'accueil
• Docs amont
• Page du gestionnaire de paquets

Commandes et alias

• burrow

Historique du projet

LinkedIn announced Burrow on June 12, 2015 as an open-source project from its Data Infrastructure SRE team. The launch post explained that standard Kafka lag metrics were not enough for wildcard consumers, mirror makers, and large consumer groups because partition-level lag and fixed thresholds produced noisy or incomplete alerts.

The initial open-source design monitored Kafka-committed offsets, exposed consumer state over HTTP, and reduced many per-partition signals into an OK, warning, or error status. The repository changelog records 0.1.0 as the initial release on October 7, 2015, 0.1.1 in May 2016 with Zookeeper and Storm offset checking, and later 1.x releases that tracked newer Kafka and Go ecosystems.

LinkedIn revisited Burrow in May 2018 for Burrow 1.1, describing a rewrite that added tests, modularized components, improved topic-deletion handling, moved configuration toward Viper-supported formats such as TOML and JSON, and made the project easier to maintain and embed.

Historique d'adoption

Burrow was adopted quickly by other companies after LinkedIn's release, according to LinkedIn's 2018 follow-up post. The input metadata records packages for Homebrew, Debian, and Ubuntu, and the GitHub project remains a common reference point for Kafka consumer-lag monitoring.

Modes d'utilisation

Operators run Burrow with a configuration directory, often mounted at `/etc/burrow`, then query HTTP endpoints for Kafka cluster, consumer group, broker, and status information. Burrow can also notify via email or HTTP integrations, making it suitable for alert pipelines that want health states rather than raw lag counters.

Pourquoi les passionnés de paquets s'y intéressent

Burrow is important to package nerds because it captures a very specific operational lesson from large Kafka deployments: good monitoring packages often encode domain-specific judgment, not just metrics collection. It also illustrates the Go-era pattern of distributing infrastructure daemons as relatively easy-to-package single binaries with TOML config.

Chronologie

• 2015-06-12: LinkedIn announced Burrow as an open-source Kafka consumer monitoring project.
• 2015-10-07: Burrow 0.1.0 initial release recorded in the changelog.
• 2016-05-01: Burrow 0.1.1 added Zookeeper and Storm offset checking.
• 2018-05-15: LinkedIn announced Burrow 1.1 and described the rewrite and modularization work.
• 2022-2026: 1.x releases continued to update Go, dependencies, and Kafka compatibility.

Related projects

• Burrow is related to Apache Kafka, Kafka MirrorMaker, ZooKeeper, Storm, Sarama, Viper, Prometheus exporters, and Kafka management or monitoring tools such as Kafka Manager/CMAK and broker-level metrics systems.

Sources

• https://engineering.linkedin.com/apache-kafka/burrow-kafka-consumer-monitoring-reinvented
• https://github.com/linkedin/Burrow
• https://raw.githubusercontent.com/linkedin/Burrow/master/CHANGELOG.md
• https://raw.githubusercontent.com/linkedin/Burrow/master/README.md
• https://www.linkedin.com/blog/engineering/open-source/revisiting-burrow-burrow-1-1-
• source_facts.package-manager