Observability lab
A real Prometheus, Grafana, Alertmanager and OpenTelemetry Collector, all running and all reachable in your browser. Learn mode has you scrape targets, write PromQL, build a dashboard and wire an alert; Break-fix mode has you find where a signal stopped.
Why this matters: This was the thinnest topic on the whole site — OpenTelemetry appeared zero times across 4,021 cards. It is also the requirement most likely to come up as 'how would you tell?' in an infrastructure interview. Monitoring is not a thing you can learn from slides; you have to watch a target go down and go find out why.
Observability — getting started
New to monitoring? Each walkthrough spins up a real Prometheus, Grafana, Alertmanager and OpenTelemetry Collector — you scrape a target, write PromQL, build a dashboard you can actually open in your browser, wire an alert, and follow a trace through a collector pipeline.
Loading scenarios…