Argus

Lightweight async endpoint uptime monitor. Pings a configurable list of HTTP targets on a set interval and exposes uptime, latency, and request counts as Prometheus metrics.

Stack

Component	Technology
Runtime	Python 3.11
API	FastAPI + Uvicorn
HTTP client	httpx (async)
Metrics	prometheus-client
Config	pydantic-settings + PyYAML
Observability	Prometheus + Grafana

Metrics exposed

Metric	Type	Labels
argus_endpoint_up	Gauge	url, name
argus_endpoint_latency_seconds	Histogram	url, name
argus_endpoint_requests_total	Counter	url, name, status_code

Quick start

# Clone and run the full stack
git clone https://github.com/malik-builds/argus
cd argus
docker compose up --build

• Argus metrics: http://localhost:8000/metrics
• Prometheus: http://localhost:9090
• Grafana: http://localhost:3000 (admin/admin)

Configuration

Edit config.yaml to set targets and scrape interval:

scrape_interval: 30

targets:
  - name: github
    url: https://github.com
  - name: my-api
    url: https://api.example.com/health

Screenshot

Add Grafana dashboard screenshot here