DataOmbudsman / dataplatform

dataplatform

My progress through the Manning LiveProject course Create a Data Platform for Real-time Anomaly Detection.

Notes

Some useful commands:

docker build -t af/lp-jupyter .  # build image

docker run -p 8889:8889 af/lp-jupyter  # create container from image

docker run -p 8889:8889 --mount type=bind,source="$(pwd)",target=/src af/lp-jupyter  # ... with bind mount

docker-compose up --build  # boot with building image

docker-compose up --build  # boot with building image and detached mode

docker-compose ps  # list of running containers

docker-compose stop  # stops containers

docker-compose down  # stops and removes containers

docker-compose down && docker-compose up --build  # reboot

ADMIN_USER=admin ADMIN_PASSWORD=admin docker-compose down && docker-compose up --build -d  # start with setting default username/password

Components

• FastAPI: web framework
• Prometheus: monitoring system
• Grafana: dashboard
• Caddy: reverse proxy service used for authentication

Access

• http://localhost:8000/metrics/ — Prometheus raw metrics
• http://localhost:9090/graph — Prometheus console
• http://localhost:8000/docs — API docs
• http://localhost:3000/?orgId=1 — Grafana