tl;dr

Start an OpenFGA server, simulate some traffic, check the metrics and traces.

Pre-requisites

• Docker: https://docs.docker.com/engine/install/
• FGA CLI: https://github.com/openfga/cli#installation

Demo

1. Start OpenFGA with tracing and metrics enabled: docker-compose up --detach
2. Create a store: export FGA_STORE_ID=$(fga store create --model model.fga | jq -r .store.id)
3. Write the authorization model: fga model write --file model.fga
4. Write tuples: fga tuple write --file tuples.yaml
5. Make Check calls: fga model test --tests demo.fga.yaml
6. See metrics in Prometheus: http://localhost:9090/. E.g. go_sql_open_connections
7. See metrics in Grafana: http://localhost:3001/
8. See traces in Jaeger: http://localhost:16686/search

Cleanup

1. Stop OpenFGA and remove data: docker-compose down --volumes

Resources

• Learn more about OpenFGA: https://openfga.dev / https://openfga.dev/blog.

• Engage with the OpenFGA community in Discord or Github.

• Follow us on Twitter: @openfga

And please, don't forget to star our Github Repository!