Irakli Khakhviashvili

Objective

To set up a static website - awssolutionsarchitect.com - hosted on Amazon S3, secured with an SSL certificate from AWS Certificate Manager (ACM), and delivered through Amazon CloudFront. The site should be accessible via both www and non-www domains, with proper DNS configuration using Route 53.

Challenges

• When creating a new ACM certificate, it remained stuck in "Pending validation" indefinitely.
• The certificate wouldn't validate because I had an A record pointing to the CloudFront distribution.
• Ensuring the site was accessible via both www and non-www domains was tricky.
• Properly routing the static website from an S3 bucket through CloudFront required additional setup.
• Each time I made an update to the site, I had to manually upload the files to the S3 bucket. This process was time-consuming and inefficient, and I wanted to automate it. My goal was to set up a workflow so that whenever I pushed changes to the main branch on GitHub, the changes would be automatically deployed to the S3 bucket. I needed to run the build command and upload the production files to S3 without manual intervention.

Solutions

• I added the required CNAME record in Route 53, allowing the certificate to validate.
• I deleted the A record pointing to CloudFront, and the certificate validated within minutes.
• I added both www and non-www versions as alternate domain names in the CloudFront distribution and created an A record in Route 53 pointing the www version to CloudFront.
• I set up a dedicated S3 bucket for my static website and routed it through CloudFront.
• To automate the deployment process, I created a new GitHub Action that connects to my S3 bucket and uploads the updated production files. First, I created a new IAM user in AWS and granted the necessary permissions to read and write to the S3 bucket. Then, I stored the IAM user's credentials as secret variables in GitHub. Finally, I updated the GitHub Action YAML file to use these credentials for authentication. Now, every time I commit and push changes to the main branch in GitHub, the build is automatically run, and the production files are uploaded to the S3 bucket without manual intervention. Easy and efficient! 🚀

Outcome

Successfully deployed a static website on S3 with CloudFront for content delivery and SSL security via ACM. The site is fully accessible through both www and non-www domains, with proper DNS configuration in Route 53. This project improved my understanding of Route 53, ACM, CloudFront, and static site hosting on AWS.

Objective

Design and deploy a highly available, secure, and scalable WordPress application on AWS. Leverage AWS services like Amazon RDS Aurora and AWS Elastic Beanstalk to ensure optimal performance, reliability, and security. Integrate load balancers and auto-scaling groups to enhance scalability and resilience, while isolating the database and securing communication between components using AWS security groups.

Live Demo: WordPress Deployment

Challenges

• Set up security auto groups to allow seamless communication between resources created with Beanstalk, such as EC2s in the public subnet and the RDS Aurora database.
• Faced issues while deploying the WordPress application via Beanstalk due to misconfigurations in the config files in .ebextensions.
• Encountered challenges querying data via the AWS Query Editor to test and retrieve data from the database.

Solutions

• Created custom security groups and correctly configured them to ensure seamless communication between EC2 instances in the public subnet and the RDS Aurora database.
• Configured the efs-create.config file with my custom VPC ID, which allowed the successful deployment of the WordPress application via Beanstalk.
• Discovered that the AWS Query Editor was only supported in certain DB engine versions, so I switched to Aurora MySQL 3.07.1 to query data from my database using the Query Editor, utilizing familiar SQL-like language.

Outcome

The project resulted in a highly secure, scalable, and high-performance WordPress application hosted on AWS. By leveraging AWS Elastic Beanstalk, RDS Aurora, and custom security groups, I was able to architect an infrastructure that optimized both security and performance. The decoupled architecture, with the database in a private subnet and the application in a public subnet, ensured robust isolation and security, while also enabling seamless communication between components. The successful configuration of AWS services allowed for automated scaling and load balancing, improving both reliability and user experience resulting in an efficient and reliable and cost-efficient WordPress hosting solution.

Objective

The objective of this project was to build a small serverless application using AWS services. By leveraging AWS Lambda, DynamoDB, API Gateway, and Cognito for user authentication, the app provides users with personalized movie recommendations based on TMDb data.

Live Demo: Should I Watch This?

Challenges

• Integrating AWS Services: Configuring and integrating multiple AWS services (Lambda, API Gateway, Cognito, DynamoDB) to work together, while ensuring each service is properly secured and optimized for serverless architecture with no server provisioning or management.
• Securing API Endpoints: Implementing and securing API Gateway endpoints using Cognito authorization to ensure that only authenticated users can interact with the application's backend services, adding an extra layer of security.
• GitHub Actions and CI/CD Pipeline: Setting up a reliable GitHub Actions pipeline to automate the build and deployment process to S3.
• CORS and Query Parameters Access Issues: Facing challenges with configuring Cross-Origin Resource Sharing (CORS) correctly and ensuring that query parameters are accessible while setting up API Gateway resources.

Solutions

• Learned a great deal about Lambda, API Gateway, Cognito, and DynamoDB to connect these services by creating a seamless integration between each server. I focused on setting up correct IAM user policies and role permissions, ensuring proper communication and security between the services. This allowed me to achieve a fully functional and secure serverless application.
• Configured Cognito user pools and added necessary authentication logic in API Gateway, enabling secure access control.
• I configured a GitHub Actions workflow that automatically triggers on code pushes to the main branch. It installs dependencies, builds the project, and securely deploys it to AWS S3 by using GitHub environment variables for AWS credentials. This automated the deployment process, reducing manual intervention.
• I carefully configured CORS settings in API Gateway to allow necessary origins, headers, and methods. I also ensured that query parameters were correctly passed through to Lambda functions by enabling Lambda proxy integration in API Gateway.

Outcome

I successfully built a simple yet fully functional serverless application using AWS services like Lambda, API Gateway, Cognito, and DynamoDB. The app allows users to securely log in, and manage their movie watchlists. By integrating these services with proper security and IAM configurations, I created a cost-effective, scalable solution with no servers to manage. The deployment process is also automated through GitHub Actions, ensuring seamless updates and maintenance. The app is now running efficiently on AWS, with CloudFront providing high availability, showcasing the power of serverless architecture. In this project I have learned a great deal about serverless architecture, its real use cases, and benefits.