Home

Awesome

Serverless Lambda S3 Demonstration

This project demonstrates how the Serverless Framework can be used to deploy a NodeJS Lambda function that responds to events in an S3 bucket.

In our demonstration, the Lambda function responds to .csv files uploaded to an S3 bucket, transforms the data to a fixed width format, and writes the data to a .txt file in an output bucket.

sampleInputData.csv
First,Last,Email,Phone
John,Munson,john.munson@gmail.com,343-231-3893
Ed,Karisch,edward.karisch@gmail.com,680-236-1187
Josh,Stevens,josh.stevens@gmail.com,851-990-4343
sampleOutputData.txt
John                Munson              john.munson@gmail.com                             343-231-3893
Ed                  Karisch             edward.karisch@gmail.com                          680-236-1187
Josh                Stevens             josh.stevens@gmail.com                            851-990-4343

Getting Started

You will need to have the Serverless Framework installed globally with npm install -g serverless. In addition, you will need to configure Serverless Framework with your AWS credententials.

Next, go ahead and clone the project and install package dependencies.

Because the serverless.yml file is configured to provision any AWS resources that the Lambda function is dependent on, and because S3 bucket names must be globally unique, you will need to change CSV-BUCKET-NAME-CHANGE-ME and FIXED-WIDTH-BUCKET-NAME-CHANGE-ME in serverless.yml to something that is meaningful but still unique.

At this point, the only thing left to do is deploy our function!

serverless deploy -v

Testing it out

If everything went according to plan, you should be able to login to the AWS S3 console and upload a .csv file to the input bucket. It will need to match the schema that schema.js is expecting. Feel free to use the sampleData.csv file provided with this repo.

The deployed Lambda function will be triggered and should generate a fixed width file that gets saved in the output bucket. Whether the function succeeded or failed, there should be some sort of output in AWS Cloudwatch.

Cleaning Up

If you were just playing around with this project as a learning exercise, you may want to perform a bit of cleanup when you're all finished.