Recently, while working on a workshop titled Testing Your Pull Request on Kubernetes with GKE, and GitHub Actions, I faced twice the same issue: service A needs service B, but service A starts faster than service B, and the system fails. In this post, I want to describe the context of these issues and how I solved them both with the same tool. Waiting in Kubernetes It might sound strange to wait in Kubernetes. The self-healing nature of the Kubernetes platform is one of its biggest benefits.

In my previous post Chopping the Monolith, I explained that some parts of a monolith are pretty stable and only the fast-changing parts are worth being 'chopped.' I turned the post into a talk and presented it at several conferences. I think it’s pretty well received; I believe it’s because most developers understand, or have direct experience, that microservices are not a good fit for traditional organizations, as per Conway’s Law.

In my previous blog post Chopping the monolith, I explained my stance on microservices and why it shouldn’t be your golden standard. However, I admitted that some parts of the codebase were less stable than others and had to change more frequently. I proposed 'chopping' these few parts to cope with this requirement while keeping the monolith intact. As Linus Torvalds once wrote: Talk is cheap, show me the code!

If you attend conferences or read technical articles, you could think that microservices are the correct and only way to build a system at the moment. Despite some pushback from cooler heads, the default architecture is microservices. In this post, I’d like to argue why it’s wrong. I’ll first get back to the origin of microservices and the fundamental reason to use them. Then, I’ll describe why microservices don’t fit most organizations' structures.

During the pioneer area of the World Wide Web, the content was static. To serve it, a group of developers created a web server, which is now known as the Apache Web Server. The Apache Web Server is built around a module architecture. Developers created a module to run CGI scripts to add dynamic content to the lot. Users wrote early CGI scripts in Perl.