Summary

In this chapter we’ve learned a lot about Telemetry and Observability. The three telemetry types or signals are logs, metrics and traces which provide valuable insights into the system observed from slightly different perspectives. An observable system is the one that is constantly monitored where we know the state based on the telemetry data that serves as evidence.

We’ve also learned about projects such as OpenTelemetry that can help with instrumentation and simplify the work needed to implement telemetry. Had a quick introduction to projects such as Zipkin and Jaeger for tracing and had a closer look at Prometheus – a fully featured monitoring platform.

Prometheus supports both Push and Pull operating models for metric collection, but dominantly uses Pull model to periodically scrape the metric data at (/metrics endpoint) and save it in TSDB in time-series format. Having metrics in TSDB allows us visualizing the data in software such as Grafana...