System Status

Simple FastAPI status page backed by Prometheus queries.

Pipeline (end to end)

1. A service exposes health/metrics data (/metrics) directly, or via an exporter (for example cAdvisor, blackbox, postgres-exporter).
2. Prometheus scrapes that target based on prometheus/prometheus.yml (scrape_configs).
3. Prometheus stores those time series (for example up, probe_success, container_last_seen).
4. system_status/services.yml defines which PromQL query each row (or child row) should use.
5. system_status/server.py runs each query against Prometheus (/api/v1/query_range) for the last 24 hours.
6. The app converts query results into UP/DOWN/NO DATA + history.
7. system_status/templates/index.html renders the final status page.

What to edit when adding a new service

1. Make sure the service has a metric source (native /metrics or an exporter).
2. Add/confirm a scrape job in prometheus/prometheus.yml.
3. Add a row in system_status/services.yml with a query that returns 0/1 style status.
4. Restart affected containers:

docker compose restart prometheus system-status

Run

docker compose up -d --build system-status

Open: http://localhost:9110

JSON mode: http://localhost:9110/?json=1