This repo contains a sample application based on an ecommerce web application I named AfrikaMart to explore several architecture concepts like;
- Microservices
- CQRS
- PUB/SUB pattern
- Clean Architecture
and how to use container technologies like;
- Docker
- Docker-compose
- Kubernetes (Not yet implemented)
- Centralized Distributed Logging (using the ELK (Elastic search, Log stash, and Kibana) stack)
I've created a software architecture diagram which shows all the moving parts in the application. Services communicate synchronously using gRPC while asynchronous communication are done with RabbitMQ Publish/Subscribe Topic Exchange Model.
This is the front end of the system. Users can create account, login, view products, purchase product. (Not yet developed)
This service provides an API to manage products, perform CRUD operations on products. It uses MongoDB to persist products.
This service provides an API for managing user basket. When you add a product to cart. It uses a Redis database for persistence. This service also publish the event;
- BasketCheckoutEvent
This is a gRPC service that is use for calculating discounts, and getting products as well as user specific discounts. It uses Postgres Sql for persistence.
This service is responsible for managing orders. Performing CRUD on Orders . It also consumes the "BasketCheckoutEvent" event from Basket Service. it makes use of CQRS pattern with MediatR, FluentValidation and AutoMapper packages.
This will be responsible for handling notifications, user specific notifications as well as order notification. (Not yet implemented). Notificatios are currently handled in the Order service when orders are created, but that should be extracted into its own service.
This will be a gRPC service responsible for handling payments (Not yet implemented). Currently, payments are being handled in Order service upon creation of orders, this will be extracted into a seperate service to handle and store payments.
It use Ocelot Api Gateway and serves as a unified point of entry into the system using the BFF pattern.
This describes the technology and libraries used to build this application.
This application is built completely using .Net5 and Asp.net5. You can learn more here.
This is a light weight ORM used in Discount service. You can learn about Dapper here.
For the Order service, database-access is implemented using Entity Framework Core code-first. See here for more info.
Every service within the system and all infrastructural components (database, message-broker) are run in a Docker container. In this solution, I only used Linux based containers. Docker Compose is used to orchestrate the application and connect all the components. See here for more info about Docker.
RabbitMQ is used as message-broker. I use a default RabbitMQ Docker image (including management) from Docker Hub (rabbitmq:3-management). See here for more info about RabbitMq.
The database server used to host all databases is MS SQL Server running on Linux. I use a default 2017 SQL Server Docker image from Docker Hub (mcr.microsoft.com/mssql/server:2017-latest).
AutoMapper is used (only where it adds value) to map between POCOs. This is primarily handy when mapping commands to events, events to events or events to models. See automapper for more info.
Swashbuckle is used for auto-generating Swagger documentation and a test-ui for the ASP.NET Web APIs. See swashbuckle for more info.
I will be updating this as I build this project.
I will be assuming you are on a windows machine. To run this project, Make sure you have the latest version of Docker Desktop running smoothly on your machine. I only used Linux based containers.
To build the images, make sure you have cloned the project to your machine, then run do the following;
using the terminal, Change direcroty to the "src" folder
cd src
Then, run the folowing docker command in the "src" directory
docker-compose -f docker-compose.yml -f docker-compose.override.yml up -d
The above commands will build the images and run the container on your machine. These might take quite some time depending on your internet as there are many external images it will have to pull. When it completes, You can check docker desktop that your images has been built and container running. Optionally, you can navigate to the portainer service I added through http://host.docker.internal:9000 -- admin/P@ssw0rd portainer, Portainer makes it easier to manage Docker containers, images, networks, and volumes from a web-based Portainer dashboard.
These are the set hostname and port for each of the services;
Catalog Service -> http://host.docker.internal:8000/swagger/index.html
Basket Service -> http://host.docker.internal:8001/swagger/index.html
Discount Service -> http://host.docker.internal:8002/swagger/index.html
Order Service -> http://host.docker.internal:8004/swagger/index.html
Shopping.Aggregator -> http://host.docker.internal:8005/swagger/index.html
API Gateway -> http://host.docker.internal:8010/Catalog
RabbitMq Management Dashboard -> http://host.docker.internal:15672 -- guest/guest
Portainer -> http://host.docker.internal:9000 -- admin/P@ssw0rd
pgAdmin PostgreSQL -> http://host.docker.internal:5050 -- admin@philo.com/P@ssw0rd
I will update this readMe as I add more services and build out more features.
I learnt a lot from exploring Microsoft eShopOnContainers refrence app which you can find here.
I explored most, if not all of Mehmet Ozkaya Articles on implementing cloud native applications in .Net ecosystem.