Observability

Observability is how well you can understand what is happening inside a system by looking at the data it emits. That data is usually metrics, logs and traces. Monitoring tells you that something you anticipated is wrong. Observability lets you ask new questions during an incident, so you can find out why it went wrong.

How it works

Services are instrumented to emit telemetry. Metrics are numbers aggregated over time, such as a 5xx error rate. Logs are timestamped records of events. Traces follow one request across services. Shared attributes, such as a trace ID or service name, let an engineer move from a spike on a dashboard to the slow request and then to the log line that explains it. OpenTelemetry is the common open standard for producing this data.

Why it matters in production

In a system of many services, the failure you hit is often one you never built a dashboard for. Good observability shortens the time to isolate a fault, which is usually the longest part of an incident.