Storefront

This project is a barebones partial implementation of an online game store platform.
The goal of this project was to present a variety of skills in creating microservices, which include:

• Architecture design
• Documentation
• Testing
• Library usage
• External tool usage
• Fault tolerance
• Monitoring
• Streamlined building
• Maintainability

My priority was to present an example in each of those categories. Due to time constraints, some functionalities like product drafts and authorization had to be left out, as it's trivial to add them later on, and they don't add to the primary goal of showcasing my skills.

Below, I have listed an optimal way of reviewing this project:

1. Figjam architecture overview
2. How to run the project
3. How to use the API
4. Remaining TODOs are listed below

Remaining TODOs

Here is a list of remaining things to do that have not been finished due to time constraints:

• Set up ElasticSearch to add intelligent search to the products and orders services
• Set up Kibana for advanced monitoring
• Set up Spring Cloud Config to externalize sensitive service configurations for different environments
• Pass principal information in the headers to allow services to use the gateway security context.
• Add role-based authorization on the gateway
• Add product drafts to the products service
• Add the image storage service to host product image content
• Add the binary storage service to host product binary content
• Add angular frontend
• Add Nginx instances to host the frontend
• Route Zipkin through the gateway as a protected route.
• Add a dev jib docker build for macOS with aarch64 architecture (faster execution on macOS) for development
• Add custom error responses using @ControllerAdvice/@ExceptionHandler instead of the standard spring boot 500 error message.
• Set up k8s
• Add Postgres & MongoDB replication
• Migrate from Sleuth to Micrometer once Micrometer reaches maturity, which will allow migration to Spring Boot 3.0.0

Known issues

• On M1 macOS, the docker environment might sometimes hang, and one or two containers might freeze. This might be due to the arm64 arch of the docker images produced by the maven build explained in the RUNNING file. In such a case, restart the affected containers. A second build configuration just for m1 macOS would fix the issue.