Skip to main content

How to add a parameter to AWS Terraform

This guide details how to expand the existing Terraform setup to include a new configuration parameter: jit_capacity. This parameter allows you to control the maximum number of concurrent JIT streams allowed in your environment. See Services - Orchestrator for available environment variables.

To achieve this, you will propagate a new variable from your Terraform inputs down to the user_data script running on the Orchestrator instance.

Prerequisites​

Define the Input Variable​

Navigate to the terraform/aws folder in the jit-deployment-examples repository.

Begin with declaring the new variable so that Terraform can accept it as an input.

Open variables.tf. This file currently holds variables such as vpc_id and target_version. Append the following block to define jit_capacity as a number with a default value (e.g., 10), allowing you to override it at runtime if necessary.

variables.tf
variable "jit_capacity" {
    description = "Max number of concurrent JIT streams"
    type        = number
    default     = 10
}

Pass the Variable to the Orchestrator​

Next, you need to pass this variable into the Orchestrator's configuration template. The Orchestrator resource is defined in compute.tf, which uses a templatefile function to generate the user data script.

Locate the aws_instance resource named orchestrator in compute.tf. Inside the user_data block, add the jit_capacity key to the variable map. This map already contains variables like jit_ami_id and jit_orchestrator_port.

compute.tf
resource "aws_instance" "orchestrator" {
# ... existing configuration ...

    user_data = templatefile("${path.module}/user_data.yaml.tpl", {
        jit_ami_id              = data.aws_ami.agent_image.id
        jit_agent_profile       = aws_iam_instance_profile.agent_profile.name
        jit_security_group_id   = aws_security_group.agent_sg.id
        jit_key_name            = aws_key_pair.jit_key.key_name
        jit_orchestrator_port   = var.orchestrator_port
        jit_instance_agent_port = var.instance_agent_port
        jit_capacity            = var.jit_capacity
    })

    # ... existing configuration ...
}

Update the User Data Template​

Finally, you must update the template file to write this value into the environment file on the server. The user_data.yaml.tpl file currently writes configuration to /opt/jit/jit.env.

Open user_data.yaml.tpl and append the new environment variable JIT_CAPACITY to the content block.

user_data.yaml.tpl
#cloud-config
write_files:
  - path: /opt/jit/jit.env
    content: |
      JIT_AWS_AMI_ID=${jit_ami_id}
      JIT_AWS_AGENT_INSTANCE_PROFILE_NAME=${jit_agent_profile}
      JIT_AWS_SECURITY_GROUP_ID=${jit_security_group_id}
      JIT_AWS_KEY_PAIR_NAME=${jit_key_name}
      JIT_ORCHESTRATOR_PORT=${jit_orchestrator_port}
      JIT_INSTANCE_AGENT_PORT_MIN=${jit_instance_agent_port}
      JIT_INSTANCE_AGENT_PORT_MAX=${jit_instance_agent_port}
      JIT_CAPACITY=${jit_capacity}
    append: true
    owner: ubuntu:ubuntu

Verification​

To verify the changes, initialize and apply the Terraform configuration. You can pass a custom value for the new parameter to ensure it is being picked up correctly.

terraform apply -var "jit_capacity=20" -var "target_version=<insert version here>"

After the Orchestrator instance provisions, you can SSH into the instance and check the contents of the environment file:

cat /opt/jit/jit.env

You should see JIT_CAPACITY=20 (or your default value) at the bottom of the file.