Prometheus Has a Free Open-Source Monitoring System

#cloud #monitoring #opensource #devops

Prometheus is a free, open-source systems monitoring and alerting toolkit. It is the standard for cloud-native monitoring.

What Is Prometheus?

Prometheus collects metrics from your services and stores them as time series data. Combined with Grafana, it creates a powerful monitoring stack.

Key features:

Pull-based metrics collection
PromQL query language
Multi-dimensional data model
Built-in alerting (Alertmanager)
Service discovery
No external dependencies
CNCF graduated project
Thousands of exporters available

Quick Start

docker run -d -p 9090:9090 \
  -v prometheus-data:/prometheus \
  prom/prometheus

Dashboard at http://localhost:9090.

Monitor Anything

Exporters available for:

Node (Linux/system metrics)
MySQL, PostgreSQL, MongoDB, Redis
Nginx, Apache, Traefik
Docker, Kubernetes
AWS, GCP, Azure
Custom apps (client libraries for every language)

PromQL

# CPU usage rate
rate(node_cpu_seconds_total{mode!="idle"}[5m])

# HTTP request rate
rate(http_requests_total[1m])

# 95th percentile latency
histogram_quantile(0.95, rate(http_request_duration_seconds_bucket[5m]))

Instrument Your App

from prometheus_client import Counter, Histogram, start_http_server

REQUESTS = Counter("http_requests_total", "Total requests", ["method", "path"])
LATENCY = Histogram("http_request_duration_seconds", "Request latency")

@LATENCY.time()
def handle_request(method, path):
    REQUESTS.labels(method=method, path=path).inc()
    # your code

start_http_server(8000)  # Prometheus scrapes this