Setup complete AWS environment for Push Dashboard

[bbangert]

This includes setting up:

• Firehose for notifications landing into S3
• Lambda for processing S3 notification payloads
• EC2 instance for test autopush setup
• EC2 instance for developer dashboard
• Redis instance for lambda to dump recent messages into

Deployment:

• Deploy autopush with firehose running to EC2 test instance
  • Cleanup and modernize logging, provide file output autopush#419
  • Add firehose logger output autopush#421
• Deploy and configure S3 triggered push-processor
  • Tag a release and upload to S3 push-processor#4
  • Update JSON format to handle standard JSON record delimiter push-processor#3
• Deploy push-messages API app
  • Setup initial project and API push-messages#1
  • Change status code to 204 when GET /messages/{key} is empty push-messages#6
  • Deploy live staging environment push-messages#7
  • Dockerize push messages for easy deployment push-messages#8

[groovecoder]

On the EC2 instance for developer dashboard, are we considering the existing Deis cluster (of 3 EC2 instances) to serve that purpose? Or do I need to create another deployment of the dashboard on new EC2 instance(s)?

[bbangert]

I was going to enable the Push Messages API to be accessible from anywhere so you wouldn't need to be in the same security group.

[groovecoder]

Just on the dev environment? I.e., in stage/prod it will be restricted to calls coming from the same security group so the outside world can't hit it?

[bbangert]

I'm not sure, we'll need to discuss with ops. I assume for stage it'll be restricted as ops deploys that.

[bbangert]

This is now complete.