Challenge Orchestrator

This is the base code for challenge orchestrator which setup two lumbda functions and a state machine on AWS platform using serverless.

Dependency

• Nodejs v12
• serverless

Environment Variables

Name	Description
REGION	AWS region
API_ENDPOINT	The endpoint for lambda function to call

The environment variables are passed using dotEnv, so there should be a .env file containing the variables in the root folder before deployment. You can find a example of .env in the root folder called example.env.

How to deploy

1. Install dependencies

   npm install

2. Run lint

   npm run lint
   npm run lint:fix  # with fix

3. Configure AWS account with serverless. Refer https://www.serverless.com/framework/docs/providers/aws/guide/credentials/ for details.

4. Deploy

npm run deploy

5. Remove resources after varification

   npm run destroy

Explainations

Currently, two lambda functions and one step function (state machine) is deployed for demostration.

Name	Type	Handler
submit	Lambda function	src/handler.submit
startExec	Lambda function	src/handler.startExec
submitMachine	state machine	N/A

The workflow is:

• user send http request to trigger startExec
• startExec starts submitMachine
• stateMachine start from state CallSubmit, and execute submit as a task
• stateMachine stop as CallSubmit is the only state it has for now

Verification

1. Deploy the lambda functions and state machine as instructed above. Copying example.env to .env and keeping the environment variables is recommended.

2. Note that there is an endpoints in the console output for startExec.

3. Make a POST request to the endpoint'. The body of the POST request must have field "input" and "name". "name" should be unique for each execution. For example

   curl -X POST 'https://e0gbflxn58.execute-api.us-east-1.amazonaws.com/dev/start' \
   --header 'Content-Type: application/json' \
   --data-raw '{
       "input": {
           "message": "hello",
           "value": 123
       },
       "name": "name1"
   }'

   • Check status code 200 and message 'ok' from response

   • Go to https://console.aws.amazon.com/cloudwatch/home, click log groups from left menu, check logs from both lambda functions

   • Go to https://console.aws.amazon.com/states/home, click the stateMachine and check the executions and logs of it.

   • Go to https://webhook.site/#!/de92f4c4-dc9d-4d90-9b00-8bcb350d19e9/9ed801a9-e0e3-4fcf-b868-ef94e089cae0/1, check the request details from submit function. The body should be the same as your request.

     Note: this url is related to API_ENDPOINT variable. Any request sent to this endpoint will be shown on the page. If you change API_ENDPOINT, you have to check the request on your own.

4. Don't forget to remove resources after verification : )

Local Deploy

Lambda Functions

Lambda functions can be deployed locally with the help of plugin serverless-offline. Just run sls offline and you can invoke your lambdas.

You can also configure it to run in docker. Add --useDocker after the command. See https://www.npmjs.com/package/serverless-offline for details.

Step functions

There is a dockerized step function image provided by Amazon, but step functions have to be deployed to it using aws command line tool with template.yaml file and the file can't be generated from serverless.yml file. So another file has to be maintained which is not a good idea.

What's more, lambda functions and step functions use ARNs to interact with each other. I'm not sure if it will work with local-deployed lambda function.

With all the above being said, I will leave this part as TBD. https://docs.aws.amazon.com/step-functions/latest/dg/sfn-local-lambda.html might be a possible solution for your reference.