serverless-s3-local

serverless-s3-local is a Serverless plugin to run S3 clone in local. This is aimed to accelerate development of AWS Lambda functions by local testing. I think it is good to collaborate with serverless-offline.

Installation

Use npm

npm install serverless-s3-local --save-dev

Use serverless plugin install

sls plugin install --name serverless-s3-local

Example

serverless.yaml

service: serverless-s3-local-example
provider:
  name: aws
  runtime: nodejs12.x
plugins:
  - serverless-s3-local
  - serverless-offline
custom:
# Uncomment only if you want to collaborate with serverless-plugin-additional-stacks
# additionalStacks:
#    permanent:
#      Resources:
#        S3BucketData:
#            Type: AWS::S3::Bucket
#            Properties:
#                BucketName: ${self:service}-data
  s3:
    host: localhost
    directory: /tmp
resources:
  Resources:
    NewResource:
      Type: AWS::S3::Bucket
      Properties:
        BucketName: local-bucket
functions:
  webhook:
    handler: handler.webhook
    events:
      - http:
        method: GET
        path: /
  s3hook:
    handler: handler.s3hook
    events:
      - s3: local-bucket
        event: s3:*

handler.js

const AWS = require('aws-sdk');

module.exports.webhook = (event, context, callback) => {
  const S3 = new AWS.S3({
    s3ForcePathStyle: true,
    accessKeyId: 'S3RVER', // This specific key is required when working offline
    secretAccessKey: 'S3RVER',
    endpoint: new AWS.Endpoint('http://localhost:4569'),
  });
  S3.putObject({
    Bucket: 'local-bucket',
    Key: '1234',
    Body: new Buffer('abcd')
  }, () => {} );
};

module.exports.s3hook = (event, context) => {
  console.log(JSON.stringify(event));
  console.log(JSON.stringify(context));
  console.log(JSON.stringify(process.env));
};

Configuration options

Configuration options can be defined in multiple ways. They will be parsed with the following priority:

• custom.s3 in serverless.yml
• custom.serverless-offline in serverless.yml
• Default values (see table below)

Feature

• Start local S3 server with specified root directory and port.
• Create buckets at launching.
• Support serverless-plugin-additional-stacks
• Support serverless-webpack
• Support serverless-plugin-existing-s3
• Support S3 events.

Working with IaC tools

If your want to work with IaC tools such as terraform, you have to manage creating bucket process. In this case, please follow the below steps.

1. Comment out configurations about S3 Bucket from resources section in serverless.yml.

#resources:
#  Resources:
#    NewResource:
#      Type: AWS::S3::Bucket
#      Properties:
#        BucketName: local-bucket

2. Create bucket directory in s3rver working directory.

$ mkdir /tmp/local-bucket

Triggering AWS Events offline

This plugin will create a temporary directory to store mock S3 info. You must use the AWS cli to trigger events locally. First, using aws configure set up a new profile, i.e. aws configure --profile s3local. The default creds are

aws_access_key_id = S3RVER
aws_secret_access_key = S3RVER

You can now use this profile to trigger events. e.g. to trigger a put-object on a file at ~/tmp/userdata.csv in a local bucket run: aws --endpoint http://localhost:4569 s3 cp ~/tmp/data.csv s3://local-bucket/userdata.csv --profile s3local

You should see the event trigger in the serverless offline console: info: PUT /local-bucket/user-data.csv 200 16ms 0b and a new object with metadata will appear in your local bucket.

See also

• s3rver
• serverless-offline
• serverless-webpack
• serverless-plugin-additional-stacks
• serverless-plugin-existing-s3

License

This software is released under the MIT License, see LICENSE.