How to use terraform to process SQS messages in a lambda

Published on

Last Updated on

Estimated Reading Time: 2 min

In the last post, we saw how to create an SNS topic to publish the bounce notification to an SQS queue. In this post, we will set up a lambda to process the available messages on the queue.

Step 1: Create an AWS lambda

We will create the lambda function using Resource: aws_lambda_function.

resource "aws_lambda_function" "SESBouncesLambda" {
  filename         = "./zips/"
  function_name    = "SESBouncesLambda"
  role             = aws_iam_role.ses_bounces_lambda_role.arn
  handler          = "Lambda::Lambda.Function::FunctionHandler"
  source_code_hash = filebase64sha256("./zips/")
  runtime          = "dotnetcore3.1"
  • Set the file name (including the path) for the deployment package. Create a temporary zip file. We will look at the actual lambda in a later post.
  • Set the unique name for the lambda.
  • Set the IAM role for the Lambda function so that we can give it permissions. We will look at this below.
  • Set the entry-point. This is similar to the main function in a console app.
  • Set the source code hash so that we can trigger updates for when the function changes.
  • Set the runtime to .Net Core 3.1 so that AWS knows that the lambda function is built using .Net core 3.1.

Step 2: Create an IAM role for the lambda

We will use Data Source: aws_iam_policy_document to allow the lambda service to assume an IAM role to get temporary credentials.

We will use Resource: aws_iam_role to set the IAM role for the lambda.

data "aws_iam_policy_document" "ses_bounces_lambda_role_iam_policy" {
  statement {
    actions = ["sts:AssumeRole"]
    principals {
      type        = "Service"
      identifiers = [""]

resource "aws_iam_role" "ses_bounces_lambda_role" {
  name               = "SESBouncesLambdaRole"
  assume_role_policy = data.aws_iam_policy_document.ses_bounces_lambda_role_iam_policy.json
  • Use AssumeRole, which allows AWS services to assume an IAM Role.
  • Constrain the policy so that only lambdas can assume an IAM role.
  • Attach the policy to the role.

Step 3: Give the lambda access to read messages from SQS queues

We will use Resource: aws_iam_role_policy_attachment to attach an IAM policy to allow the lambda to read messages from the SQS queue.

resource "aws_iam_role_policy_attachment" "lambda_sqs_role_policy" {
  role       =
  policy_arn = "arn:aws:iam::aws:policy/service-role/AWSLambdaSQSQueueExecutionRole"
  • Set the role to which this policy should be attached
  • Add the AWS managed policy AWSLambdaSQSQueueExecutionRole which has the permissions that the function needs to read items from Amazon SQS and write logs to CloudWatch Logs.

Step 4: Give the lambda access to the queue

We will use Resource: aws_lambda_event_source_mapping to allow our lambda to get events from the SQS queue

resource "aws_lambda_event_source_mapping" "event_source_mapping" {
  event_source_arn = aws_sqs_queue.ses_bounces_queue.arn
  function_name    = aws_lambda_function.SESBouncesLambda.arn
  • Set the queue to ses_bounces_queue.
  • Set the lambda to SESBouncesLambda.

Next Steps

You can see the code here.

So far, we have configured AWS to publish bounce notifications to an SQS queue. In the next post, we will see how to write a lambda function to send a custom message to a Slack channel.