Tasker is a project management tool designed to facilitate team collaboration and task tracking. It provides a range of functionalities to manage projects, assign tasks, and monitor progress.

As I am primarily a backend/DevOps engineer, I focused on the backend, infrastructure, and CI/CD aspects of the project. The backend is built using a serverless architecture using AWS services such as Lambda, DynamoDB, and Cognito. The infrastructure is defined using Terraform, ensuring the application is scalable and easy to manage. The CI/CD pipeline is implemented using GitHub Actions, automating the testing and deployment process. The frontend is built using Next.js and Tailwind CSS hosted on Amplify.

The frontend was originally built by Edward Roh. Special thanks for this amazing app! I lack frontend skills, so I had to borrow his work to make this project complete.

Introduction

Functionalities

Sign up, log in, and sign out.
Create projects and tasks.
Assign tasks to team members.
Update task status and track task progress.
Light and dark mode support.

Limitations

The application is still in the early stages of development and has several limitations. The following features are not yet implemented:

Changing user avatars and settings.
Changing task details.
Changing team members.
Uploading attachments.
Adding comments to tasks.
Deleting tasks and projects.

Backend and Infrastructure

The backend is built using a serverless architecture, with AWS Lambda functions attached to API Gateway endpoints.

The infrastructure is defined using Terraform, which ensures the application is scalable and easy to manage. Key infrastructure components include:

S3: Used for storing public images and other static assets.
DynamoDB: Table definitions for storing different entities such as projects, users, teams, and tasks.
Cognito: For user pool management and authentication.

System Manager's Parameter Store is used to store sensitive information and environment variables.

CI/CD Pipelines

Tasker uses GitHub Actions for its CI/CD pipeline, which automates the testing and deployment process. There are 2 main workflows:

Amplify Deployment: Deploys the frontend application to Amplify. This workflow is triggered on every push to the tasker-client directory of the repository.
Serverless Deployment: Deploys the backend application to AWS. This workflow is triggered on every push to the tasker-server directory of the repository. Although the architecture is serverless, the directory is named tasker-server for consistency.

GitHub OIDC is used to authenticate the GitHub Actions workflows with AWS.

Manual Deployment

Pre-requisites:

AWS credentials: Ensure you have AWS credentials configured on your machine.
Terraform.
Serverless Framework.
Node.js and npm.

Steps:

Set up the infrastructure using Terraform. Change the S3 bucket name in terraform/s3.tf to a unique name. Run the following commands:

npm run infra:deploy

Deploy the backend using the Serverless Framework. Run the following commands:

cd tasker-server
npm install
npm sls:deploy

(Optional) To populate the database with sample data, run the following command:

sls invoke local function -f populateSeedData

Deploy the frontend using Amplify. It is recommended to use Amplify Console for this. Follow the instructions in the Amplify Console to deploy the frontend. Set the following environment variables in the Amplify Console to the values created in System Manager's Parameter Store:

NEXT_PUBLIC_API_BASE_URL
NEXT_PUBLIC_COGNITO_USER_POOL_ID
NEXT_PUBLIC_COGNITO_USER_POOL_CLIENT_ID
NEXT_PUBLIC_S3_PUBLIC_IMAGE_URL

The repo is a monorepo with the frontend in tasker-client, so set AMPLIFY_MONOREPO_APP_ROOT to tasker-client.