Book Image

Cloud-Native Observability with OpenTelemetry

By : Alex Boten
Book Image

Cloud-Native Observability with OpenTelemetry

By: Alex Boten

Overview of this book

Cloud-Native Observability with OpenTelemetry is a guide to helping you look for answers to questions about your applications. This book teaches you how to produce telemetry from your applications using an open standard to retain control of data. OpenTelemetry provides the tools necessary for you to gain visibility into the performance of your services. It allows you to instrument your application code through vendor-neutral APIs, libraries and tools. By reading Cloud-Native Observability with OpenTelemetry, you’ll learn about the concepts and signals of OpenTelemetry - traces, metrics, and logs. You’ll practice producing telemetry for these signals by configuring and instrumenting a distributed cloud-native application using the OpenTelemetry API. The book also guides you through deploying the collector, as well as telemetry backends necessary to help you understand what to do with the data once it's emitted. You’ll look at various examples of how to identify application performance issues through telemetry. By analyzing telemetry, you’ll also be able to better understand how an observable application can improve the software development life cycle. By the end of this book, you’ll be well-versed with OpenTelemetry, be able to instrument services using the OpenTelemetry API to produce distributed traces, metrics and logs, and more.
Table of Contents (17 chapters)
1
Section 1: The Basics
3
Chapter 2: OpenTelemetry Signals – Traces, Metrics, and Logs
5
Section 2: Instrumenting an Application
10
Section 3: Using Telemetry Data

Enriching the data

You may have noticed in output from the previous examples that each span emitted contains a resource attribute. The resource attribute provides an immutable set of attributes, representing the entity producing the telemetry. resource attributes are not specific to tracing. Any signal that emits telemetry leverages resource attributes by adding them to the data produced at export time. As covered in Chapter 1, The History and Concepts of Observability, the resource in an application is associated with the telemetry generator, which, in the case of tracing, is TracerProvider. The resource attribute on the span output we've seen so far is automatically provided by the SDK with some information about the SDK itself, as well as a default service.name. The service name is used by many backends to identify the services sending traces to them; however, as you can see, the default value of unknown_service is not a super useful name. Let's fix this. The following...