• Downloads
  • Install and Run
  • Building from Source
  • Docker
• Core Concepts
• Backends
  • In Memory
  • H2
  • Postgres
  • Cassandra
• Configuration Reference
  • Reaper Specific Settings
  • Backend Specific Settings
  • Docker Variables
  • Shiro Authentication
• Metrics
  • Graphite Reporter
• Using Reaper
  • Adding a Cluster
  • Checking a Cluster's Health
  • Running a Cluster Repair
  • Scheduling a Cluster Repair
  • Multi DCs with One Reaper
  • Multi DCs with Multi Reapers
  • Sidecar Mode
  • Cassandra Diagnostic Events
• Rest API
• Community
• Development

Downloads and Installation

Packages

The current stable version can be downloaded in the following packaging formats :

• Deb package
• RPM
• Tarball

The current development version can be downloaded in the following packaging formats :

• Deb package
• RPM
• Tarball

Quick Installation Guide

For a docker image, please see the Docker section.

Once the appropriate package has been downloaded, head over to the Install and Run section.

Upgrading from 1.2.0/1.2.1 to 1.2.2

We unfortunately had to break schema migrations for upgrades to 1.2.2 from 1.2.0 and 1.2.1. Here is the upgrade procedure for each Reaper backend:

Cassandra

• Stop all Reaper instances
• Run the following DDL statement on the backend cluster :

ALTER TABLE reaper_db.repair_unit_v1 DROP repair_thread_count;

• Upgrade Reaper to 1.2.2 and start it

Note: by doing so the number of threads for each existing schedule will revert back to 0, which will be translated to a single repair thread. If you previously had defined a specific number of threads you will need to recreate these schedules or update the repair_unit_v1 table manually (this field is only useful with Cassandra 2.2, there is no benefit in Cassandra <=2.1 or >=3.0).

H2

• Stop Reaper
• Pointing to your current Reaper jar file (located in /usr/share/cassandra-reaper for packaged installs), run the following command : java -cp /usr/share/cassandra-reaper/cassandra-reaper-1.2.*.jar org.h2.tools.Shell When asked, provide the JDBC URL from your Cassandra Reaper yaml file (for example : jdbc:h2:~/reaper-db/db;MODE=PostgreSQL), use the default Driver and leave username/password empty. Then run the following statements:

ALTER TABLE REPAIR_UNIT DROP COLUMN repair_thread_count;
ALTER TABLE REPAIR_SEGMENT DROP COLUMN TOKEN_RANGES;

• Upgrade Reaper to 1.2.2 and start it

Note: by doing so the number of threads for each existing schedule will revert back to 0, which will be translated to a single repair thread. If you previously had defined a specific number of threads you will need to recreate these schedules or update the repair_unit table manually (this field is only useful with Cassandra 2.2, there is no benefit in Cassandra <=2.1 or >=3.0).

Postgres

• Stop Reaper
• Connect to your Postgres database using the psql shell (or any other client) and run the following statements:

ALTER TABLE "repair_unit" DROP COLUMN "repair_thread_count";
ALTER TABLE "repair_segment" DROP COLUMN "token_ranges";

• Upgrade Reaper to 1.2.2 and start it

Note: by doing so the number of threads for each existing schedule will revert back to 0, which will be translated to a single repair thread. If you previously had defined a specific number of threads you will need to recreate these schedules or update the repair_unit table manually (this field is only useful with Cassandra 2.2, there is no benefit in Cassandra <=2.1 or >=3.0).