In our company, we have an internal application that utilizes MongoDB. The application consists of multiple microservices, each interacting with its own MongoDB instance. While the application doesn’t experience heavy loads, it’s critical for our business, so we need it to be resilient. Creating multiple MongoDB instances for each service seemed impractical and costly. Therefore, we opted for a unified resilient MongoDB cluster. Since we currently don’t require sharding, we’ll only be using MongoDB replicas. In this article, I’ll describe how I set up replicas in AWS using Docker and how we use them in our application.