Serverless Microservices As One (how to handle multiple API gateways with serverless offline)

This is a node server that allows to use multiple running serverless offline instances as a single backend, i.e. on a singe host and port.

It manages everything out-of-the-box, from starting up all the servers to routing the incoming requests to the relevant service, everything dynamically by looping the services folder in your project and parsing serverless.yml files.

Why?

When using the microservices architecture with serverless framework and the serverless offline plugin, you can't have more than one microservice running locally on the same host and port, unless you create an additional serverless.yml that combines all the other serverless.yml of each service, or solutions like that. As a frontend developer, this can be troublesome when using serverless offline as a local backend for the frontend, where it's not reasonable to have different urls for each microservice. As a backend developer, yes it's true you can start the microservice you are working on, but how many times you need to check a different endpoint in another microservice, or your colleague ask you to quickly check something? More than we'd like to admit probably, and having to stop and start the server every time is a nono for me.

That's why I've created this node server that acts as a proxy and allows you to handle multiple API Gateway as a single backend, exposing only one url and port.

How it works

This is basically what this script does:

scans your service directories and parses each serverless.yml file
starts serverless offline for every microservice found
spin up an express server that handles all the method + paths found in the serverless.yml files
proxies and routes every incoming request to the correct server matching method and endpoint of the request
show all console output of every service in its console

How to use

run node index.js --servicesFolderPath path/to/services

Command line options

--servicesFolderPath path to folder containing services
--appPort            port to listen to, default 3001
--port               port to start sls offline from
--stage              stage for sls offline

Please note

I've created this in less than 3 hours, so it's not so well written (but at least it's commented!)
It's my first project with node and express 😅

I'd like to make this more ready for a general use so in the next days I will add for sure support for passing arguments at startup and solve the TODOs left in the code.

Check the issue tab to see what's planned in details