In the Serverless Framework – Building Web App using AWS Lambda Amazon API Gateway S3 DynamoDB and Cognito – Part 1, we’ve created and deployed a simple web application using a Serverless framework. In this article, we’ll improve our results and make the application deployment declaration even more clear.

We can replace API Gateway resources with the events: which are available on a Serverless framework.

You may find the final result, which we got at the end of the previous post at my GitHub repository. Please, use tag v1.0 as a starting point. Final result is available at tag v2.0.

Replacing API Gateway resources

First thing we need to do is to comment all resources, which has Type: AWS::ApiGateway::* in serverless.yaml file:

Also, you’ll need to comment WildRydesApiGatewayUrl in the Outputs: section, because we’re removed API Gateway declaration:

Now we can start adding the same configuration by using events: declaration in functions: section. Let’s publish our existing function RequestUnicorn using API Gateway with CORS enabled:

functions:
  RequestUnicorn:
    handler: handler.handler
    role: WildRydesLambdaRole
    events:
      - http:
          path: ride
          method: post
          cors: true

I removed all not necessary comments from the file to make the file more readable.

Let’s deploy our infrastructure using the following command:

sls deploy

Now we need to implement API Gateway Authorizer. I think, we can uncomment one of the previously commented resources and modify it’s reference to the API Gateway.

WildRydesApiGatewayAuthorizer:
  Type: AWS::ApiGateway::Authorizer
  Properties:
    Name: WildRydes
    RestApiId:
      Ref: ApiGatewayRestApi
    Type: COGNITO_USER_POOLS
    ProviderARNs:
      - Fn::GetAtt: [WildRydesCognitoUserPool, Arn]
    IdentitySource: method.request.header.Authorization

You may be interested, where I got ApiGatewayRestApi as a reference to the API Gateway, which we never declared. The reason is the Serverless framework which converts serverless.yaml file to the CloudFormation template which we deploying each time we’re calling sls deploy command. You may find it’s content in .serverless/cloudformation-template-update-stack.json file inside our project structure after the first deploy.

All we need to do is to find AWS::ApiGateway::RestApi (API Gateway itself) resource declaration and take its name as a reference.

And finally, we need to add declared authorizer to our WildRydesLambdaRole Lambda function like this:

functions:
  RequestUnicorn:
    handler: handler.handler
    role: WildRydesLambdaRole
    events:
      - http:
          path: ride
          method: post
          cors: true
          authorizer:
            type: COGNITO_USER_POOLS
            authorizerId:
              Ref: WildRydesApiGatewayAuthorizer

Let’s redeploy our stack to make sure everything’s working:

sls deploy

If you’ve already destroyed previous application, let’s upload web application static content on S3 once more again. Getting web application sources:

git clone https://github.com/awslabs/aws-serverless-workshops/

As you remember, we already described all needed Outputs: in resources: section of our serverless.yaml file. So, all we need to do is to execute the following command to get it:

sls info --verbose

Now we’re ready to edit the config.js file:

window._config = {
  cognito: {
    userPoolId: 'us-east-1_dADFuXCii', // e.g. us-east-2_uXboG5pAb
    userPoolClientId: '5dvnlid054tvjcrv3106762bh7', // e.g. 25ddkmj4v6hfsfvruhpfi7n4hv
    region: 'us-east-1', // e.g. us-east-2
  },
  api: {
    invokeUrl: 'https://c98nutbckd.execute-api.us-east-1.amazonaws.com/dev', // e.g. https://rc7nyt4tql.execute-api.us-west-2.amazonaws.com/prod',
  },
};

Now let’s upload web application content to S3 and remove it from our project sources:

aws s3 sync ./aws-serverless-workshops/WebApplication/1_StaticWebHosting/website s3://wildrydes-firstname-lastname
rm -rf ./aws-serverless-workshops

Redeploy the stack, if you did not do it earlier:

sls deploy

Testing

Now our application is up and running. All we need to do is to verify its functionality by opening the WildRydesBucketURL, registering new user using /register.html URL, verifying user manually using Cognito web interface and logging in using /ride.html URL. The whole testing process is described in my first post and original AWS tutorial.

Final result

Let remove all commented sections and take a look under the final result:

service: wild-rides-serverless-demo # NOTE: update this with your service name

provider:
  name: aws
  runtime: nodejs8.10
  stage: ${opt:stage, 'dev'}

functions:
  RequestUnicorn:
    handler: handler.handler
    role: WildRydesLambdaRole
    events:
      - http:
          path: ride
          method: post
          cors: true
          authorizer:
            type: COGNITO_USER_POOLS
            authorizerId:
              Ref: WildRydesApiGatewayAuthorizer

# you can add CloudFormation resource templates here
resources:
  Resources:
    WildRydesBucket:
      Type: AWS::S3::Bucket
      Properties:
        BucketName: wildrydes-andrei-maksimov
        WebsiteConfiguration:
          IndexDocument: index.html
    WildRydesBucketPolicy:
      Type: AWS::S3::BucketPolicy
      Properties:
        Bucket:
          Ref: 'WildRydesBucket'
        PolicyDocument:
          Statement:
            - Effect: 'Allow'
              Principal: '*'
              Action:
                - 's3:GetObject'
              Resource:
                Fn::Join:
                  - ''
                  - - 'arn:aws:s3:::'
                    - Ref: 'WildRydesBucket'
                    - '/*'
    WildRydesCognitoUserPool:
      Type: AWS::Cognito::UserPool
      Properties:
        UserPoolName: WildRydes
    WildRydesCognitoUserPoolClient:
      Type: AWS::Cognito::UserPoolClient
      Properties:
        ClientName: WildRydesWebApp
        GenerateSecret: false
        UserPoolId:
          Ref: 'WildRydesCognitoUserPool'
    WildRydesDynamoDBTable:
      Type: AWS::DynamoDB::Table
      Properties:
        TableName: Rides
        AttributeDefinitions:
          - AttributeName: RideId
            AttributeType: S
        KeySchema:
          - AttributeName: RideId
            KeyType: HASH
        ProvisionedThroughput:
          ReadCapacityUnits: 5
          WriteCapacityUnits: 5
    WildRydesLambdaRole:
      Type: AWS::IAM::Role
      Properties:
        RoleName: WildRydesLambda
        AssumeRolePolicyDocument:
          Version: '2012-10-17'
          Statement:
            - Effect: Allow
              Principal:
                Service:
                  - lambda.amazonaws.com
              Action: sts:AssumeRole
        Policies:
          - PolicyName: DynamoDBWriteAccess
            PolicyDocument:
              Version: '2012-10-17'
              Statement:
                - Effect: Allow
                  Action:
                    - logs:CreateLogGroup
                    - logs:CreateLogStream
                    - logs:PutLogEvents
                  Resource:
                    - 'Fn::Join':
                        - ':'
                        - - 'arn:aws:logs'
                          - Ref: 'AWS::Region'
                          - Ref: 'AWS::AccountId'
                          - 'log-group:/aws/lambda/*:*:*'
                - Effect: Allow
                  Action:
                    - dynamodb:PutItem
                  Resource:
                    'Fn::GetAtt': [WildRydesDynamoDBTable, Arn]
    WildRydesApiGatewayAuthorizer:
      Type: AWS::ApiGateway::Authorizer
      Properties:
        Name: WildRydes
        RestApiId:
          Ref: ApiGatewayRestApi
        Type: COGNITO_USER_POOLS
        ProviderARNs:
          - Fn::GetAtt: [WildRydesCognitoUserPool, Arn]
        IdentitySource: method.request.header.Authorization
  Outputs:
    WildRydesBucketURL:
      Description: 'Wild Rydes Bucket Website URL'
      Value:
        'Fn::GetAtt': [WildRydesBucket, WebsiteURL]
    WildRydesCognitoUserPoolId:
      Description: 'Wild Rydes Cognito User Pool ID'
      Value:
        Ref: 'WildRydesCognitoUserPool'
    WildRydesCognitoUserPoolClientId:
      Description: 'Wild Rydes Cognito User Pool Client ID'
      Value:
        Ref: 'WildRydesCognitoUserPoolClient'
    WildRydesDynamoDbARN:
      Description: 'Wild Rydes DynamoDB ARN'
      Value:
        'Fn::GetAtt': [WildRydesDynamoDBTable, Arn]

As you can see, now we have much less code.

Resource cleanup

To cleanup everything you need to call

aws s3 rm s3://wildrydes-firstname-lastname --recursive
sls remove

Summary

We hope, you’ve found this article helpful. If you have any questions, please, feel free to ask them in the comments section. Also, you may find additional examples of API Gateway integrations using the Serverless framework in its Events documentation.

Related articles

• What is Serverless computing
• Serverless Framework – Building Web App using AWS Lambda Amazon API Gateway S3 DynamoDB and Cognito – Part 1
• Top Frameworks For Serverless Applications Development
• AWS Lambda – How to process DynamoDB streams
• Security Best Practices For Serverless Applications