msull / s3-upload-with-presigned-using-sam

Upload file to s3 with generated presigned url with SAM AWS SAM

Upload files to s3 with presigend URLs with AWS SAM

This project is to create a presigned url to upload files to s3 directory. Whole stack will be created using AWS SAM.

Architecture is as follows:

Prerequisites

• Need to have sam cli and npm installed

How to install

1. Within the backend directory, run:

sam build
sam deploy -g

• Parameter SnsNotificationEmail : Add an email address to recieve SNS notification from uploaded file processed events

2. This will create:

• S3 bucket for static web site hosting
• S3 bucket to upload files
• Lambda function to generate pre-signed url
• API gateway to trigger above lambda
• Lambda function to trigger when file is uploaded into the S3 bucket
• Lambda destination SNS topic for above function when success
• Lambda destination SNS topic for above function when failed

** While creating the stack, you will receive two emails to subscribe to above sns topics. Confirm the subscriptions.

Once the stack is created, note the values of below outputs:

• S3WebsiteURL - This will be the final url to access the system.
• S3WebsiteBucket - S3 bucket for static website contents.
• S3FileUploadBucket - S3 bucket to store the uploaded files
• PresignedUrlApi - API endpoint to generate the presigned url.

3. Go to the frontend directory. Update the variable VUE_APP_GENERATE_URL in the .env file with the value of PresignedUrlApi.

4. Run below commands to install dependancies and build the frontend

npm install
npm run build

This will generate the project files to deploy in the dist directory.

5. To copy the built contents to s3, run:

aws s3 cp dist s3://<S3WebsiteBucket> --recursive

6. Access the web site using the S3WebsiteURL value.

7. Upload a file and you should receive a sns notification to the given email.

Deleting the stack

1. First remove the data in the both s3 buckets:

aws s3 rm s3://<S3FileUploadBucket> --recursive
aws s3 rm s3://<S3WebsiteBucket>  --recursive

2. Then run:

aws cloudformation delete-stack --stack-name <Stack Name>