Tracing protocols and best practices

Tracing, or as it is more commonly referred to, distributed tracing, tracks application requests as they are made between services of a system. It allows you to follow a single request through an entire system or look at the aggregate data over requests to better understand distributed behavior.

This capability provides software developers (Diego), operators (Ophelia), and service managers (Steven) with valuable tools that enable an understanding of the flow of logic that is essential for troubleshooting. Instrumenting your code by adding traces helps you easily pinpoint almost any issue or at least have a clear indicator of where the problem could be. Distributed tracing uses the concepts of spans and traces to capture this data. Let’s examine these in more detail.

Spans and traces

The trace record is the parent object that represents the data flow or execution path through the system being observed. Each trace will contain one...