Skip to content

Latest commit

 

History

History
167 lines (129 loc) · 5.63 KB

07-lambda-layers.md

File metadata and controls

167 lines (129 loc) · 5.63 KB
Raw

Lambda Layers

This article outlines the steps involved in building a node js lambda using lambda layers for library dependencies.

The example will build a lambda function that will return current time using momentjs library. The lambda will not bundle moment.js via node_modules, as we did in previous tutorial but, will use lambda layers.

What is a lambda layer ?

(Source: AWS Docs) :

  • A layer is a ZIP archive that contains libraries, a custom runtime, or other dependencies.
  • With layers, you can use libraries in your function without needing to include them in your deployment package.

(1) Publish the layer

We will bundle the node_modules for the moment.js library and publish to AWS as a layer.

(1.1) Create the lambda layer
  • The package.json in samples/07/momentjs-lambda-layer folder declares moment.js as a dependency.
  • The packaging should use the folder structure shown below for nodejs lambda layers. This folder structure is a requirement from lambda layers.
  • Move the node_modules into nodejs subdirectory and package them as a zip archive file.

Following steps will bundle the archive required for creating the lambda layer. Run the following commands from samples/07/momentjs-lambda-layer folder (or) you can run the bundle-layer.sh script in the same folder.

Note: Here we are moving the package.json to a dist/nodejs folder, installing the package and bundling the nodejs folder in an archive named momentJSLambdaLayer.zip

#!/bin/sh
mkdir -p dist/nodejs
cp package.json dist/nodejs
cd dist/nodejs
npm install
cd ..
zip -r /tmp/momentJSLambdaLayer.zip nodejs
cd ..
rm -rf dist

(2) Bundle the lambda

This section will refer to the source in samples/07/current-time-lambda folder and currentTimeLambda.js source (below).

const moment = require('moment');

exports.handler = (event, context, callback) => {
  const time = moment().format('MMMM Do YYYY, h:mm:ss a');
  callback(null, { time });
};

Note :

  • The lambda source has reference to moment.js, but moment.js is not included in package.json.
  • We will attach the lambda layer that was created earlier to this lambda in following steps.

Let's bundle the lambda source without its dependencies from samples/07/current-time-lambda folder using the following command (or) the script bundle-lambda.sh in samples/07/current-time-lambda.

#!/bin/sh
zip /tmp/currentTimeLambda.zip currentTimeLambda.js

(3) Terraform script for currentTimeLambda and momentJSLambdaLayer

Terraform script in samples/07/main.tf (below) will create the following resources

  1. momentJSLambdaLayer (lambda layer)
  2. currentTimeLambda (lambda with its role)
provider "aws" {
  region = var.aws_region
}

locals {
  layer_name = "momentJSLambdaLayer"
  layer_payload = "/tmp/momentJSLambdaLayer.zip"
  
  lambda_name = "currentTimeLambda"
  zip_file_name = "/tmp/currentTimeLambda.zip"
  handler_name = "currentTimeLambda.handler"
}

module "moment_js_lambda_layer" {
  source = "../modules/lambda-layer"
  lambda_layer_name = local.layer_name
  lambda_layer_payload = local.layer_payload
}

module "lambda_tf_way_role" {
  source = "../modules/lambda-role"
}

module "current_time_lambda" {
  source = "../modules/lambda"
  lambda_function_name = local.lambda_name
  lambda_function_handler = local.handler_name
  lambda_role_arn = module.lambda_tf_way_role.lambda_role_arn
  lambda_zip_filename = local.zip_file_name
  lambda_tf_way_layer = module.lambda_tf_way_layer.lambda_tf_way_layer_arn_with_version
}
  • Module moment_js_lambda_layer creates a lambda layer with payload pointing to /tmp/momentJSLambdaLayer.zip
  • Module current_time_lambda creates a lambda with payload pointing to /tmp/currentTimeLambda.zip. The layer is attached to this lambda by passing the ARN of the lambda layer moment_js_lambda_layer created earlier.

(4) Terraform Apply

Now we will run terraform script to create the MomentJS Lambda layer, and Current time lambda. You need to be in the samples/07 folder to run the script.

Note: The AWS_PROFILE is configured as lambda-tf-user

export AWS_PROFILE=lambda-tf-user
terraform init
terraform apply --auto-approve

(5) Invoke lambda

(5.1) Invoke the lambda

We will invoke the currentTimeLambda via AWS CLI to test the deployment of lambda with momentjs lambda layer.

aws lambda invoke \
    --function-name currentTimeLambda \
    --profile "$AWS_PROFILE" \
    --log-type Tail \
    --cli-binary-format raw-in-base64-out \
    --payload '{}' outputfile.txt

Successful invocation should list a response similar to the following.

{
  "LogResult": "A1BCDefghiJKLmnOPQRSt2OWZmMzMwMS0yY2IxLTQ5Y2ItOGJlMS0yMWQwNGZ...TYyLjk2IG1zCQo=",
  "ExecutedVersion": "$LATEST",
  "StatusCode": 200
}

(5.2) View the output

  • To ensure the current time is responded by currentTimeLambda, view the contents of outputfile.txt file. You should see an output similar to the one below.
> cat outputfile.txt
{"time":"January 1st 2020, 01:00:00 pm"}

(6) Teardown

Now, we will delete the currentTimeLambda and momentJSLambdaLayer, using terraform destroy. From the samples/07/ folder run the following command.

export AWS_PROFILE=lambda-tf-user
terraform destroy --auto-approve

🏁 If you have seen the timestamp, Congrats ! You got your first lambda layer working using Terraform 🏁

Next: Integrate with S3