Skip to content

Latest commit

 

History

History
 
 

aws-ts-s3-lambda-copyzip

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Deploy

Serverless App to Copy and Zip Objects Between Amazon S3 Buckets

This example sets up two AWS S3 Buckets and a single Lambda that listens to one and, upon each new object arriving in it, zips it up and copies it to the second bucket. Its architecture looks like this:

Architecture

This example is also featured in the blog post Easy Serverless Apps and Infrastructure -- Real Events, Real Code.

Deploying the App

To deploy your new serverless application, follow the below steps.

Prerequisites

  1. Ensure you have Node.js
  2. Install Pulumi
  3. Configure AWS Credentials

Steps

After cloning this repo, from this working directory, run these commands:

  1. Install Node.js dependencies, either using NPM or Yarn:

    $ npm install
  2. Create a new Pulumi stack, which is an isolated environment for this example:

    $ pulumi stack init

    This will ask you to give your stack a name; dev is a fine name to begin with.

  3. Configure the AWS region for this program -- any valid AWS region will do:

    $ pulumi config set aws:region us-east-1
  4. Deploy the application:

    $ pulumi up
  5. After about 20 seconds, your buckets and lambda will have been deployed. Their names are printed:

    Outputs:
        tpsReportsBucket: "tpsreports-21b7b7a"
        tpsZipsBucket   : "tpszips-c869600"
  6. Now copy a file to the tpsReportsBucket using the AWS CLI:

    $ aws s3 cp ./myTpsReport001.txt s3://$(pulumi stack output tpsReportsBucket)
  7. Tail the logs to see evidence the Lambda ran:

    $ pulumi logs -f
    Collecting logs for stack dev since 2019-03-10T10:09:56.000-07:00...
    2019-03-10T11:10:48.617-07:00[zipTpsReports] Zipping
        tpsreports-96458ef/tps001.txt into tpszips-edfde11/tps001.txt.zip
  8. ^C out of pulumi logs -f, and then download your new zipfile!

    $ aws s3 cp s3://$(pulumi stack output tpsZipsBucket)/myTpsReport001.txt.zip .
  9. Once you're done, destroy your stack and remove it -- eliminating all traces of running:

    # First, delete files so we can destroy the buckets (by default, bucket content isn't auto-deleted):
    $ aws s3 rm s3://$(pulumi stack output tpsReportsBucket)/myTpsReport001.txt
    $ aws s3 rm s3://$(pulumi stack output tpsZipsBucket)/myTpsReport001.txt.zip
    $ pulumi destroy --yes
    $ pulumi stack rm --yes