Skip to main content

Deploy to AWS

So far we've tested our program locally using our Wing Console. Next we'll see how you can deploy your program to AWS using Terraform.

Wing compiler supports multiple platforms​

The Wing compiler supports multiple compilation platforms. Each resource in the Wing ecosystem can support any number of backends. For example, the cloud.Queue resource we used in our app can be backed by Amazon SQS by Azure Queue Storage or by RabbitMQ by CloudAMQP.

A platform represents both the cloud provider and the provisioning engine. For example, the tf-aws platform will compile your program to a set of AWS resources, using Terraform as the provisioning engine.

Under Construction

🚧 We plan to also support Azure and Google Cloud as platforms out of the box. In addition, we are planning support for other provisioning engines such as AWS CloudFormation and Kubernetes.

Click 👍 on the relevant issue and tell us what you think.

Prerequisites​

In order to deploy to AWS, you will need:

  • Terraform
  • AWS CLI with configured credentials. See here for more information.

Compile to Terraform/AWS​

We will use the tf-aws platform to tell the compiler to bind all of our resources to the default set of AWS resources and use Terraform as the provisioning engine.

wing compile --platform tf-aws main.w

Examine the output​

Our target directory now contains a main.tfaws directory which has all of the Terraform configuration for this application.

Initialize Terraform state​

Now, let's deploy our program to AWS.

Credentials Required

You will need to have AWS credentials configured on your system. See here for more information.

Let's change the working directory to where our Terraform configuration is and initialize the state file:

cd ./target/main.tfaws
export AWS_REGION=us-east-1 # or any other region
terraform init

Terraform deploy​

Now we are ready to deploy to our AWS account:

terraform apply

You should expect output similar to this: 

Terraform used the selected providers to generate the following execution
plan. Resource actions are indicated with the following symbols:
  + create

Terraform will perform the following actions:

  # aws_dynamodb_table.cloudCounter will be created
  + resource "aws_dynamodb_table" "cloudCounter" {...}

  # aws_iam_role.cloudQueue-SetConsumer-cdafee6e_IamRole_2548D828 will be created
  + resource "aws_iam_role" "cloudQueue-SetConsumer-cdafee6e_IamRole_2548D828" {...}

  # aws_iam_role_policy.cloudQueue-SetConsumer-cdafee6e_IamRolePolicy_37133937 will be created
  + resource "aws_iam_role_policy" "cloudQueue-SetConsumer-cdafee6e_IamRolePolicy_37133937" {...}

  # aws_iam_role_policy_attachment.cloudQueue-SetConsumer-cdafee6e_IamRolePolicyAttachment_45079F65 will be created
  + resource "aws_iam_role_policy_attachment" "cloudQueue-SetConsumer-cdafee6e_IamRolePolicyAttachment_45079F65" {...}

  # aws_lambda_event_source_mapping.cloudQueue_EventSourceMapping_41814136 will be created
  + resource "aws_lambda_event_source_mapping" "cloudQueue_EventSourceMapping_41814136" {...}

  # aws_lambda_function.cloudQueue-SetConsumer-cdafee6e will be created
  + resource "aws_lambda_function" "cloudQueue-SetConsumer-cdafee6e" {...}

  # aws_s3_bucket.Code will be created
  + resource "aws_s3_bucket" "Code" {...}

  # aws_s3_bucket.cloudBucket will be created
  + resource "aws_s3_bucket" "cloudBucket" {...}

  # aws_s3_bucket_public_access_block.cloudBucket_PublicAccessBlock_5946CCE8 will be created
  + resource "aws_s3_bucket_public_access_block" "cloudBucket_PublicAccessBlock_5946CCE8" {...}

  # aws_s3_bucket_server_side_encryption_configuration.cloudBucket_Encryption_77B6AEEF will be created
  + resource "aws_s3_bucket_server_side_encryption_configuration" "cloudBucket_Encryption_77B6AEEF" {...}

  # aws_s3_object.cloudQueue-SetConsumer-cdafee6e_S3Object_8868B9FB will be created
  + resource "aws_s3_object" "cloudQueue-SetConsumer-cdafee6e_S3Object_8868B9FB" {...}

  # aws_sqs_queue.cloudQueue will be created
  + resource "aws_sqs_queue" "cloudQueue" {...}

Plan: 12 to add, 0 to change, 0 to destroy.

This is a good opportunity to observe how much complexity the Wing compiler was able to abstract away for you when you wrote your Wing code. Just saying...

If you choose to proceed, Terraform will do its magic and will create all of these resources in your account.

Apply complete! Resources: 12 added, 0 changed, 0 destroyed.

Explore your app on AWS​

After our application has been deployed to AWS, you should be able to interact with it through the AWS Management Console.

  1. Open the Amazon SQS Console
  2. Select your AWS region
  3. You should be able to see that you have a queue there prefixed with cloud-Queue-
  4. Click Send and receive messages.
  5. In the Message Body box type cloud and hit Send message.
  6. Jump over to the S3 Console
  7. There should be some buckets prefixed with cloud-bucket-.
  8. Cycle through the buckets until you find one that contains wing-1.txt.
  9. Click wing-1.txt then click the Open button.
  10. The file should contain Hello, cloud.

Exploring the counter on AWS​

  1. Open the Amazon Dynamo DB Console, and search for a table prefixed with wing-counter-cloud.
  2. Use the Explore table items button to view the value of the counter id attribute, which represents the current value of the counter.
  3. Optional - repeat step 5 from the above list as many times as you want. While doing it, see how the bucket fills and how the counter id increments.

Cleanup​

Terraform doesn't allow destroying a non-empty bucket by default. To prepare for easy cleanup, you may delete the newly created file(s) by marking the checkbox next to the bucket name on the S3 console, clicking the Empty button, typing permanently delete in the confirmation box and clicking the Empty button.

Once you're done, you can destroy all of the resources that were created on your AWS account by running:

terraform destroy

{...}

Destroy complete! Resources: 12 destroyed.