CDK Entest

Introduction#

In this post, I follow Elad's talk at re:invent 2018 and implement basic concepts such as build-int contructs and my-own constructs, setting up a CDK project, syntheize multiple stacks, apps.

Elad-Ben Israel - the creator of CDK joined Amazon in 2016. Then CDK first released in JUL 2019. Quoted from him: In early 2016, I joined a group of Amazon engineers who were exploring hardware utilization in one of the most important and large-scale services behind Amazon.com—Amazon Product Search. One piece of this solution required processing, in real time, all activity from Amazon.com to determine which products were being purchased...

Github

CDK Concept#

CDK CLI Workflow#

Bootstrap#

cdk bootstrap provision resources required in specified region. This is required first before we can deploy any stack in a region.

cdk bootstrap aws://AWS_ACCOUNT_ID/REGION

Enviornment Variables#

we can refere to cdk environment variables in our stack code as below

new MyDevStack(app, 'dev', {
  env: {
    account: process.env.CDK_DEFAULT_ACCOUNT,
    region: process.env.CDK_DEFAULT_REGION
  }
})

and we can provide/define them for a stack as well.

new MyDevStack(app, 'dev', {
  env: {
    account: 'MY_ACCOUNT_ID',
    region: 'REGION'
  }
})

Init a Project#

cdk init --language typescript

Project Structure#

|--build
   |--hello-cdk.ts
|--lib
   |--hello-cdk-stack.ts
|--lambda
   |--handler.py
|--cdk.out
   |--HelloCdkStack.template.json
|--cdk.json

Synthesize Templates#

synthesize cloudformation templates. Then templates stored in cdk.out

cdk synth

the default cdk synth synthesize a ts file specified by cdk.json

"app": "npx ts-node --prefer-ts-exts bin/hello-cdk.ts"

we can override it by (this means we can write multiple apps)

cdk --app 'npx ts-node --prefer-ts-exts bin/my-app.ts' synth

Deploy Stacks#

if there is one stack

cdk deploy

deploy a specific stack in multiple stacks

cdk deploy "HelloCdkStack"

if deploy all stacks in the app

cdk deploy --all

Destroy Stacks#

if there is only one stack

cdk destroy

destroy a specific stack in multiple stacks

cdk destroy "HelloCdkStack"

destroy all stacks in the app

cdk destroy --all

App, Stack, and Construct#

create an app

// create an app
const app = new cdk.App()

// a stack inside the app
new HelloCdkStack(app, 'HelloCdkStack', {
  env: {
    region: process.env.CDK_DEFAULT_REGION,
    account: process.env.CDK_DEFAULT_ACCOUNT
  }
})

create an stack

export class HelloCdkStack extends Stack {
  constructor(scope: Construct, id: string, props?: StackProps) {
    super(scope, id, props)

    // sqs
    const queue = new aws_sqs.Queue(this, 'HelloQueue', {
      queueName: 'HelloQueue'
    })

    // lambda
    const fn = new aws_lambda.Function(this, 'HelloFunction', {
      functionName: 'HelloLambda',
      runtime: aws_lambda.Runtime.PYTHON_3_8,
      code: aws_lambda.Code.fromAsset(path.join(__dirname, './../lambda')),
      handler: 'handler.handler'
    })

    // lambda event source sqs queue
    fn.addEventSource(new aws_lambda_event_sources.SqsEventSource(queue))
  }
}

here aws_sqs is a L2 construct.

Create your own construct#

aws_devops-cdk-course drawio (1)

export interface QueueRecorderProps {
  inputQueue: aws_sqs.Queue
}

export class QueueRecorder extends Construct {
  constructor(scope: Construct, id: string, props: QueueRecorderProps) {
    super(scope, id)

    // lambda
    const fn = new aws_lambda.Function(this, 'HelloFunction', {
      functionName: 'HelloLambda',
      runtime: aws_lambda.Runtime.PYTHON_3_8,
      code: aws_lambda.Code.fromAsset(path.join(__dirname, './../lambda')),
      handler: 'handler.handler'
    })

    // lambda event source - sqs queue
    fn.addEventSource(
      new aws_lambda_event_sources.SqsEventSource(props.inputQueue)
    )

    // dynamodb table
    const table = new aws_dynamodb.Table(this, 'HelloDynamoDb', {
      tableName: 'HelloTable',
      removalPolicy: RemovalPolicy.DESTROY,
      partitionKey: {
        name: 'id',
        type: aws_dynamodb.AttributeType.STRING
      }
    })

    // table name into lambda env
    fn.addEnvironment('TABLE_NAME', table.tableName)

    // grant lambda to write to table
    table.grantWriteData(fn.role!)
  }
}

Refactor the code#

export class HelloCdkStack extends Stack {
  constructor(scope: Construct, id: string, props?: StackProps) {
    super(scope, id, props)

    // sqs
    const queue = new aws_sqs.Queue(this, 'HelloQueue', {
      queueName: 'HelloQueue'
    })

    // queuerecorder construct
    new QueueRecorder(this, 'QueueRecorder', {
      inputQueue: queue
    })
  }
}

lambda write to a dynamodb table

def handler(event, context):
    """
    lambda handler
    """
    # log
    logger.info("Hello Lambda")
    # table
    dynamodb = boto3.resource("dynamodb")
    table = dynamodb.Table(os.environ['TABLE_NAME'])
    # put item to the table
    for record in event.Records:
        try:
            body = record['body']
        except:
            body = ''
        table.put_item(
            Item={
                'id': record['messageId'],
                'text': body
        }
    )