Skip to content

Services and ports

./iquana.sh status shows what is running, on which port, and at which commit.

Your browser where you annotate frontend :3000 the web UI backend :8000 REST API, exports ai-service :8004 model inference postgres :5432 container redis :6379 broker, container mlflow :5000 runs, model registry backend-worker batch inference runs ai-worker model training loads the app REST inference enqueue consume HTTP request job queue, via Redis
The eight processes and how they reach each other. Everything except your browser runs on the machine you installed on.
Service What it does Default port
frontend the web UI you annotate in 3000
backend REST API, database, exports 8000
ai-service model inference for all AI-assisted tasks 8004
ai-worker Celery worker for model training —
backend-worker Celery worker for batch inference runs —
mlflow model and experiment tracking 5000
postgres the database (container) 5432
redis task broker for both workers (container) 6379

Logs are written to logs/, one file per service.

Containers

postgres and redis run as containers and cannot start without a container runtime. The installer creates:

  • the volume iquana-pg-data — the database
  • the containers iquana-pg and iquana-redis

These are the only things installed outside the iquana-tool/ directory.

Changing ports

If a port is already in use, the installer warns and lets you pick another one. To change ports later:

./install.sh --reconfigure