11 Logging frameworks

In the previous chapter, we looked at how we can create log events that can be used to give the most meaning and value. Another significant way we can easily derive more value from logs is by using logging frameworks for our application development. Most programming languages these days will be able to use a logging framework. In some cases, the third-party ones predate the language native feature and become something of a de-facto standard. Other logging frameworks have come as part of an application container or platform to address weaknesses perceived or proven in the native solutions.

This chapter will explore the logging framework landscape as there is a range of commonalities in their capability and often in design. A general understanding of this will help us appreciate the 'art of the possible' and make informed decisions when choosing a framework. We will also look at whether the more dominant frameworks for different languages can support the ability to connect directly to Fluentd. Fluentd has also helped us in this space by providing logging libraries for multiple languages, so we'll look at those to understand how they may fit into the options we have.

11.1  Value of logging frameworks

11.2  Typical structure of a logging framework

11.2.1  Logger Context

11.2.2  Appender

    Logger

11.2.4  Filter

11.2.5  Formatter

11.2.6  Configuration

11.2.7  Logger Config

11.3  Appender structures

11.4  Logging framework landscape

11.5  Choosing a framework

11.5.1  Putting optimizing application logging into action

11.6  Fluentd's own logging and appenders

11.7  Illustrations of an application logging directly to Fluentd

11.7.1 Python with logging framework – using Fluentd library

11.7.2  Invoking Fluentd Appender directly

11.7.3  Illustration with only Python's logging

11.7.4  Illustration without Python's logging or Fluentd library

11.7.5  Porting the Fluentd calls to another language into action.

11.7.6  Using generic appenders – the takeaways

11.8  Summary