Preface

I started working with BEA Tuxedo version 8 in 2003 by writing code in the C programming language. Tuxedo seemed like a magical framework that just worked and enabled several teams to develop microservices (we called them components then) and integrate them into a single distributed system in the end. Despite being developed in the 1980s, Oracle Tuxedo still runs a significant portion of critical infrastructure and is not going away any time soon. Yet the API provided by Oracle Tuxedo is still based on the same C programming language. Developing business logic in the C programming language is overkill in 2021; we should be using better tools for that. For developer productivity, nothing beats Python.

After 17 years, I have developed my own mental model of Tuxedo by writing and debugging code every day. On one hand, it is on a level of abstraction higher than the Tuxedo API and I operate in terms of C++ and Python wrappers for the lower-level Tuxedo API. On the other hand, I dive below Tuxedo abstractions into UNIX operating system processes, message queues, and other UNIX APIs. For me, that explains the application behavior and quirks I have observed. I have also used the same approach when teaching new recruits about Tuxedo.

So, this book combines what I wish I knew when I started developing with Tuxedo and the programming language and library I wish I had in my toolbox. I hope it gives you a different viewpoint than the Tuxedo documentation does and makes your Tuxedo journey more pleasant.