cassandra-reaper.io

What It Is

Reaper for Apache Cassandra is a centralized, stateful, and highly configurable repair management tool designed to replace manually maintained crontab jobs and nodetool repair workflows. Built for multi-site Cassandra clusters, it can schedule, coordinate, and track repair tasks in a unified way, and is maintained by The Last Pickle and the open-source community.

Core Capabilities

Functionally, Reaper splits a repair run into smaller segments and applies back-pressure based on running repairs and pending compactions, helping prevent repair operations from putting excessive load on the cluster. It supports fine-grained scheduling at the keyspace, table, and even token range level, and allows tasks to be paused, cancelled, and tracked precisely. Multi-data-center support is one of its key strengths: a Cassandra backend can act as the coordination layer, allowing multiple Reaper instances to run concurrently, which makes it suitable for high-availability and cross-DC deployments.

Tech Stack and Deployment

In terms of compatibility, the documentation states that it supports Apache Cassandra 1.2 through the latest versions, with incremental repair supported on Cassandra 2.1 and later. Starting with Reaper v4, Java 11 is required. Source builds use Maven, while the UI build also depends on Node.js v10 and npm v9. Deployment options are fairly comprehensive: JAR, Docker, DEB, RPM, and source builds are all supported. For the backend, users can choose in-memory/local storage or a Cassandra persistent backend; for production, the latter—or local mode with a persistent path—is generally recommended.

API, Ecosystem, and Documentation

Reaper provides a Web UI, the bin/spreaper command-line tool, and a full REST API, making it suitable for integration into automated operations workflows. For monitoring, it supports metrics integrations such as Prometheus and Graphite. The documentation covers installation, Docker, backends, configuration references, authentication, metrics, REST API, multi-DC usage, and Sidecar mode, and includes command and YAML examples. Overall quality is good, though it still has a learning curve for users who are new to Cassandra operations.

Pricing, Pros, and Cons

The collected text does not mention commercial pricing or a hosted edition. As an open-source tool, it offers very strong value for money. Its strengths are its focused domain, deep functionality, self-hosting support, multi-DC support, and API-driven automation. Its limitations are that it only addresses the vertical problem of Cassandra repair; production configuration involves JMX, backend keyspaces, YAML, permissions, and monitoring, so complexity is not low. The text also does not provide information about commercial support or SLAs.

Who It’s For and Access from China

It is best suited for DBAs, SREs, and platform teams already running Cassandra, especially in multi-cluster or multi-data-center environments. It is less suitable for small teams without Cassandra operations experience. Access from mainland China cannot be confirmed from the text alone; however, because it relies on download sources such as GitHub, Docker Hub, and Cloudsmith, pulling images or packages may be affected by network conditions. Alternatives include using nodetool repair directly, crontab scripts, or an internal enterprise scheduling system.