This small project demonstrates the deployment of a RethinkDB database and an NGINX server using Docker-Compose and Kubernetes. It showcases the setup and orchestration of containers for a scalable and reliable real-time database solution. This project was completed during my third year of engineering studies (two years ago). 🎓
I enjoyed working on this project as it involved learning about Kubernetes and Docker and connecting systems, which was very hard at first, but it was very rewarding.
To run the project using Docker-Compose, navigate to the Archivos Docker directory and execute:
docker-compose up
This command will start the RethinkDB container followed by a script container. The script creates a collection in RethinkDB, inserts documents, and reads them. The setup ensures that the same documents are not re-inserted upon each execution by deleting and recreating the collection.
- The RethinkDB image used is the official one, while the script's image, created from an Alpine base with Python and the RethinkDB library installed, is hosted on my Docker Hub: korah91/rethinkdbscript-docker.
The Kubernetes setup replicates the Docker-Compose functionalities with additional features from Kubernetes. The cluster consists of three deployments, two Cluster-IP services, a persistent volume claim, and an Ingress object.
There are three deployments:
- RethinkDB: The main database.
- Script: Executes database operations.
- NGINX: Serves as a proxy to access the RethinkDB web interface.
The script's deployment image, similar to the Docker-Compose version, communicates with RethinkDB to perform database operations. The NGINX server verifies the functionality of the persistent volume by accessing it through its shell.
The Ingress object facilitates access to the NGINX server and the RethinkDB interface via defined routes.
Namespaces are used to logically separate the cluster, preventing conflicts between development and production environments. The inicializarTodo.sh script prompts for a namespace name, creating Kubernetes objects within the specified namespace.
Below is a visual representation of the Kubernetes cluster setup.
