PyMonitor

A Prometheus-like monitoring server built in Python.

Setup

1. Install Docker and start ClickHouse:

   sudo systemctl start docker
   sudo systemctl enable docker
   sudo usermod -aG docker $USER
   # Logout and login again
   docker compose up -d

2. Install dependencies:

   pip install -e .

3. Create config file scrape_config.json:

   {
     "jobs": [
       {
         "name": "demo",
         "targets": ["http://localhost:9090/metrics"]
       }
     ],
     "scrape_interval": 10
   }

4. Run the scraper:

   python -m pymonitor.scrape.scrape_targets

5. Run the API server:

   python -m pymonitor.main

Query API

Query metrics at http://localhost:8000/api/v1/query_range?query=cpu_usage&start=1731177600&end=1731181200

Service Discovery

PyMonitor supports service discovery for finding scrape targets. Configure via environment variables:

• PYMONITOR_JOB: Job name (default: "default")
• PYMONITOR_DISCOVERY_MODE: "static", "file", or "dns" (default: "static")
• PYMONITOR_TARGETS: Comma-separated list of endpoints (for static mode)
• PYMONITOR_SCRAPE_INTERVAL: Scrape interval in seconds (default: 60)

Example usage:

PYMONITOR_JOB=demo PYMONITOR_DISCOVERY_MODE=static PYMONITOR_TARGETS="http://localhost:9090/metrics" python -m pymonitor.scrape.scrape_targets

For DNS/file-based discovery, see src/pymonitor/service_discovery/discovery.py.

Repo Structure

• All source code is under src/pymonitor/
• Ready for packaging (see pyproject.toml)