From Green Dashboards to Real Answers: OpenTelemetry, SLOs, and the Craft of Cloud-Native Reliability - Softcover

Inhaltsangabe

At 20:41 on a Friday, Tarun got a message from a friend asking whether the app was broken. He was on call. He had not been paged. Every panel on the dashboard was green, and it stayed green for the six hours and twenty-six minutes it took to find the problem.

Nothing had crashed. That is the point.

This book follows one team through the year after that night. A food-delivery company in Pune, one checkout service that is "sometimes slow", and thirteen chapters that each add a layer to the same service until the outage that took six hours takes seventeen minutes.

What you build along the way:

• OpenTelemetry instrumentation, starting with the zero-code version you can ship in an afternoon and nobody has to approve
• Traces, metrics and structured logs that share a trace ID, so you can walk from a spike on a graph to the log line that explains it in two clicks
• A Collector pipeline you control, where dropping a costly label is a config reload instead of five pull requests
• Tail sampling that keeps every error and every slow trace at 8% of the storage
• SLIs and SLOs defined on things a customer would notice, with the error-budget arithmetic worked out in full
• Burn-rate alerts that replaced forty-seven rules with six, derived rather than guessed

There is also a chapter on what all of it costs, with a real monthly invoice broken down line by line, and the four levers that cut it by 60% without losing the data that saves you at 3 am.

Written for backend and DevOps engineers who inherited a Grafana instance from someone who left and suspect it is not telling them much. No prior OpenTelemetry. No PromQL. Working Go, YAML and PromQL examples throughout, with every threshold derived from something rather than copied from a blog post.

By the end you will have one number on a wall that means something.

Die Inhaltsangabe kann sich auf eine andere Ausgabe dieses Titels beziehen.