Using the AWS CDK to send a serverless Slack message

In this particular tutorial, I am sharing how I used the AWS SDK to send a basic Slack message using a cron job.

This is a simple proof-of-concept that I am planning to take further by swapping out the "send Slack message" with a number of other future project plans I have for do x.

In order to follow along, you'll should have the following:

Setting up the infra dependencies

In a new file, we need to setup the npm package:

# initialise with base settings
npm init -y
npm i @aws-cdk/aws-events @aws-cdk/aws-events-targets @aws-cdk/aws-lambda @aws-cdk/aws-s3-assets @aws-cdk/core dotenv
npm i --save-dev typescript @types/node @types/dotenv
copy

These dependencies are all essentially used for the CloudFormation file synthesis, with the addition of dotenv so that we can add in any secrets we want locally and not store it on git.

Gitignore file

Create a .gitignore file and add the following:

cdk.out/
node_modules/
.env
copy

This will keep out the pesky secrets we don't want stored in Git.

Setting up TypeScript

You can initialise with your preferred settings, however this is a pretty handy one you can add after creating a tsconfig.json file:

{
  "compilerOptions": {
    "target": "ES2018",
    "module": "commonjs",
    "lib": ["es2016", "es2017.object", "es2017.string"],
    "strict": true,
    "noImplicitAny": true,
    "strictNullChecks": true,
    "noImplicitThis": true,
    "alwaysStrict": true,
    "noUnusedLocals": true,
    "noUnusedParameters": true,
    "noImplicitReturns": true,
    "noFallthroughCasesInSwitch": false,
    "inlineSourceMap": true,
    "inlineSources": true,
    "experimentalDecorators": true,
    "strictPropertyInitialization": false
  }
}
copy

Setting up the infrastructure code

Create a simple index.ts file and add the following code:

import events = require("@aws-cdk/aws-events");
import targets = require("@aws-cdk/aws-events-targets");
import lambda = require("@aws-cdk/aws-lambda");
import cdk = require("@aws-cdk/core");
import assets = require("@aws-cdk/aws-s3-assets");
import path = require("path");

require("dotenv").config();

export class LambdaCronStack extends cdk.Stack {
  constructor(app: cdk.App, id: string) {
    super(app, id);

    // The following JavaScript example defines an directory
    // asset which is archived as a .zip file and uploaded to
    // S3 during deployment.
    // See https://docs.aws.amazon.com/cdk/api/latest/docs/aws-s3-assets-readme.html
    // @ts-ignore - this expects Construct not cdk.Construct :thinking:
    const myLambdaAsset = new assets.Asset(this, "LambdaCronJobExampleZip", {
      path: path.join(__dirname, "lambda"),
    });

    const lambdaFn = new lambda.Function(this, "LambdaCronJobExample", {
      code: lambda.Code.fromBucket(
        myLambdaAsset.bucket,
        myLambdaAsset.s3ObjectKey
      ),
      timeout: cdk.Duration.seconds(300),
      runtime: lambda.Runtime.NODEJS_12_X,
      environment: {
        SLACK_CHANNEL: process.env.SLACK_CHANNEL as string,
      },
      handler: "slack-message.handler",
    });

    // Run every day, every minute (UTC time)
    // See https://docs.aws.amazon.com/lambda/latest/dg/tutorial-scheduled-events-schedule-expressions.html
    const rule = new events.Rule(this, "Rule", {
      schedule: events.Schedule.expression("cron(0/1 * * * ? *)"),
    });

    rule.addTarget(new targets.LambdaFunction(lambdaFn));
  }
}

const app = new cdk.App();
new LambdaCronStack(app, "LambdaCronExample");
app.synth();
copy

So how does it work? We are simply importing all the required CDK packages to create a lambda function which occurs within the constructor for the LambdaCronStack class.

The main lambda function code comes from here:

const lambdaFn = new lambda.Function(this, "LambdaCronJobExample", {
  code: lambda.Code.fromBucket(myLambdaAsset.bucket, myLambdaAsset.s3ObjectKey),
  timeout: cdk.Duration.seconds(300),
  runtime: lambda.Runtime.NODEJS_12_X,
  environment: {
    SLACK_CHANNEL: process.env.SLACK_CHANNEL as string,
  },
  handler: "slack-message.handler",
});
copy

In the second argument, the code property takes a few options, but here we are using fromBucket. This allows us to bundle the dependencies that we need from NPM packages.

The lambda code itself is added from this code snippet:

// The following JavaScript example defines an directory
// asset which is archived as a .zip file and uploaded to
// S3 during deployment.
// See https://docs.aws.amazon.com/cdk/api/latest/docs/aws-s3-assets-readme.html
// @ts-ignore - this expects Construct not cdk.Construct :thinking:
const myLambdaAsset = new assets.Asset(this, "LambdaCronJobExampleZip", {
  path: path.join(__dirname, "lambda"),
});
copy

We are telling the CDK here to bundle the lambda folder in the root of our project and store it in a bucket as an asset. We will create this folder in the next section.

To set up the cron job, we have the following code:

// Run every day, every minute (UTC time)
// See https://docs.aws.amazon.com/lambda/latest/dg/tutorial-scheduled-events-schedule-expressions.html
const rule = new events.Rule(this, "Rule", {
  schedule: events.Schedule.expression("cron(0/1 * * * ? *)"),
});

rule.addTarget(new targets.LambdaFunction(lambdaFn));
copy

This basically just says to invoke the lambda function every minute. See the AWS docs for more info.

The last bit of importance here is to note that we are adding an environment variable SLACK_CHANNEL that is being loading from the .env file through require("dotenv").config() at the beginning of the file.

The SLACK_CHANNEL env var we are using here is simply just a URL for a Slack webhook to send into our file. You can find out how to create that on the Slack API documentation.

Create the .env file and add that webhook:

# replace the value with that of your webhook
SLACK_CHANNEL=https://hooks.slack.com/services/whatever/your/webhook/url/is
copy

Creating the Lambda function

Create a new lambda folder to use. Then, change into that folder an initialise a new npm project and add axios.

We will use axios to send the request to the Slack webhook. This is mainly to demonstration for how to bundle an npm package for Lambda with the CDK.

cd lambda
npm init -y
npm i axios
copy

Create a new file slack-message.js here where we will write the function. You can name the file whatever you want, but it needs to match the handler: "slack-message.handler" property we wrote in the CDK configuration.

touch slack-message.js
copy

Add the following code to that slack-message.js file:

const axios = require("axios");

/**
 * Send a basic text message into Slack.
 *
 * @param {*} message
 * @param {*} [channel=process.env.SLACK_CHANNEL]
 * @returns
 */
function sendText(message, channel = process.env.SLACK_CHANNEL) {
  return new Promise((resolve, reject) => {
    const data = {
      text: message,
    };
    postToChannel(data, channel)
      .then((res) => resolve(res))
      .catch((err) => reject(err));
  });
}

/**
 * Post the Slack data to a channel.
 *
 * @param {Object} data
 * @param {String} [channel=process.env.SLACK_CHANNEL]
 * @returns
 */
async function postToChannel(data, channel = process.env.SLACK_CHANNEL) {
  return await axios
    .post(channel, data, {
      headers: {
        "Content-Type": "application/json",
      },
    })
    .then((res) => res.data)
    .catch((err) => err);
}

exports.handler = async function (_, context) {
  await sendText("Lambda Cron job message");
};
copy

In the above, I basically have two helper functions for Slack sendText and postToChannel which creates a basic text message in the appropriate format for Slack from the API documentation and uses axios to make that request for us.

The exports.handler is our function that we want Lambda to invoke when it has been called. Here, we will simply call sendText("Lambda Cron job message") to send out message into Slack!

The name of the export also aligns to the handler: "slack-message.handler" property in our CDK setup (it basically goes file.exportedFunctionWeWantToInvoke).

Running the CDK

We are now ready to run the CDK!

npm install -g aws-cdk # if you have not already installed aws-cdk globally
npm install
npm run build # convert the infra TS file
copy

Once the above is done, we can synthesize, deploy or destory using the cdk command.

You can run cdk help for more information, but basically what we will want to do is run the following:

cdk synth # synthesise the file
cdk deploy
copy

You may also be required to run cdk bootstrap if it mentions it.

If successful, we will now have our lambda function deployed to AWS! Check Slack within the next minute or so to confirm things are working as expected.

Successful Slack message from the invoked lambda function

Congratulations! You can now run cdk destroy to teardown the AWS resources (unless you are so inclined to receiving that message every minute on Slack and enjoying the bill).

This can be reapplied for you to do whatever Cron jobs you would like! Life is fantastic. Happy coding!

See the completed project on GitHub.

Resources and Further Reading

A list of further reading or resources referenced throughout the post.

Image credit: Caspar Camille Rubin