Prometheus

Prometheus at a glance

Prometheus is the de-facto standard for metrics-based monitoring and alerting in cloud-native environments (a CNCF project). Prometheus ships regular releases; as a core observability component, a deliberate update rhythm is sensible.

For admins updates are usually low-risk (swap the binary/container), but larger jumps can bring configuration or storage changes — check the changelogs. The scrape and API endpoints belong on the internal network behind an auth proxy, since metrics can reveal a lot about the infrastructure. Prometheus pairs with Alertmanager (for routing) and many exporters — keep their versions compatible. After updating, verify that scraping and alerting still work; broken alerting is more dangerous than a visible outage. Track the version and its support status.