Instrumenting with OpenTracing
Now that we have run the application and got a pretty-looking trace, it is time to get to the main point of this chapter and discuss how it was instrumented to achieve this. As we discussed earlier, the primary goal of the OpenTracing project is to provide an API that allows for the creation of open source instrumentation for other open source projects and frameworks. In the Tracing Talk application, we rely heavily on such open source instrumentation for Kafka, Redis, and the Spring framework, to the point that there is very little manual instrumentation in the code, and that small amount is only due to the relative immaturity of the respective instrumentation libraries, rather than a fundamental limitation. We also made this example completely agnostic to which tracing library we are using. Unlike Chapter 4, Instrumentation Basics with OpenTracing, where we had explicit code instantiating the Jaeger tracer, this time we are using the "tracer resolver" mechanism...