Skip to content

Latest commit

 

History

History
51 lines (32 loc) · 1.54 KB

File metadata and controls

51 lines (32 loc) · 1.54 KB

Getting started with Gemini

Quick Start

  1. Download the latest release from GitHub Releases

  2. Set up two Scylla clusters - we'll call them ORACLE_CLUSTER (reference) and TEST_CLUSTER (system under test)

  3. Run Gemini:

    # Single node per cluster
    ./gemini --oracle-cluster=192.168.1.10 --test-cluster=192.168.1.20
    
    # Multiple nodes (comma-separated IP addresses)
    ./gemini --oracle-cluster=192.168.1.10 --test-cluster=192.168.1.20,192.168.1.21,192.168.1.22

    The --oracle-cluster and --test-cluster flags accept IP addresses. For multi-node clusters, provide comma-separated addresses.

That's it! Gemini will generate a random schema and start running tests.

Test-Only Mode

If you don't have an oracle cluster, you can run in test-only mode:

./gemini --test-cluster=192.168.1.20,192.168.1.21,192.168.1.22

This runs mutations without validation - useful for stress testing.

Using Docker

docker run -it scylladb/gemini:latest \
  --oracle-cluster=192.168.1.10 \
  --test-cluster=192.168.1.20,192.168.1.21,192.168.1.22

CLI Arguments

Run ./gemini --help to see all available options.

For detailed explanations, see CLI arguments documentation.

Next Steps