Scrape Configuration

Each scrape_config defines a job. A job represents a group of similar targets that share the same configuration. For example, all instances of a web service or all database nodes can belong to one job.

static_configs lists targets manually. You explicitly define the endpoints Prometheus should scrape, such as:

• localhost:9090
• app-server:8080

This is simple and useful for small setups or local development.

Service discovery finds targets automatically. In dynamic environments, Prometheus can automatically discover targets using integrations like:

• Kubernetes
• Consul
• DNS
• EC2 and other cloud providers

This is essential in containerized systems where instances frequently start and stop.

metrics_path and scheme have sensible defaults.

• metrics_path defaults to /metrics
• scheme defaults to http

These can be customized if your application exposes metrics on a different path or uses HTTPS.