In the previous post I spoke about infrastructure as code (IaC) and how Terraform can be used to provision and manage infrastructure. I also gave a simple example where I provisioned a simple Azure SignalR service using Terraform. In this post I am going to talk about how to make use of modules, input variables and output values in Terraform and improve our example in the previous post. Let’s get started.
Modules
Every software engineer knows about the DRY principle (I hope), which stipulates that you shouldn’t repeat yourself unless if it’s absolutely necessary. This principle also applies to Terraform and modules help in achieving DRY code. In Terraform, modules are containers of resources that are used together and can be shared and reused across multiple configurations. Here’s how you call a module in your configuration:
module "[MODULE_NAME]" {
source = "[link_to_module_source]"
}
Input Variables
In order for modules to be shared and reused in multiple configurations they need to be configurable and customizable. For example, you may need to use the same module for your development environment and production environment but with slightly different setups. This is where input variables come in. They serve as parameters for a module, allowing the module to be customized without altering the module’s source code. Here’s the syntax for declaring input variables:
variable "[VARIABLE_NAME]"{
type = [VARIABLE_TYPE]
description = "[DESCRIPTION]"
default = [DEFAULT_VALUE]
}
Accessing input variables in your configuration is very easy:
value = var.[VARIABLE_NAME]
The are many ways to populate your variables in Terraform. These are:
- Load from command line flag:
terraform plan -var 'variable=value' - Load from file – Create a
terraform.tfvarsfile and add your values therein:[variable_name] = value - Load from environment variables:
export TF_VAR_[variable_name]=value - Use default values when you declare the variables
Output Values
In most cases you may want to get values out of your modules after running terraform apply, for instance, database connection strings. This is made possible in Terraform by using output values. In simple terms, output values are the return values of a Terraform module and have several uses as shown in the docs. Here’s the sytanx for declaring output values:
output "[NAME]" {
value = [RESOURCE.PROPERTY]
}
Example
Let’s improve on our example from the previous post and make use of modules, input variables and output values. All the source code can be found on GitHub in case you want to get the complete code. First create the following directory structure:
terraform-example/
- modules/
- signalr/
SignalR Module
Go to the signalr directory and add the following files:
1. variables.tf
This file will contain all the input variables we need in our simple module:
variable "name" {
description = "The SignalR service name"
type = string
}
variable "resource_group_name" {
description = "The resource group to which the SignalR service belongs"
type = string
}
variable "location" {
description = "The location of the SignalR service"
type = string
}
2. outputs.tf
This file will contain all the output values that our module will return:
output "signalr_primary_key" {
value = "${azurerm_signalr_service.this.primary_access_key}"
}
output "signalr_secondary_key" {
value = "${azurerm_signalr_service.this.secondary_access_key}"
}
output "signalr_primary_connection_string" {
value = "${azurerm_signalr_service.this.primary_connection_string}"
}
output "signalr_secondary_connection_string" {
value = "${azurerm_signalr_service.this.secondary_connection_string}"
}
3. main.tf
This is where we declare all the resources that our module will contain. In this case we only have one resource – a SignalR service:
resource "azurerm_signalr_service" "this" {
name = var.name
location = var.location
resource_group_name = var.resource_group_name
sku {
name = "Free_F1"
capacity = 1
}
features {
flag = "ServiceMode"
value = "Serverless"
}
}
Configuration
We are done with our module. Now let’s make use of the module in our configuration. Go to the root directory and add the following files:
1. signalr-dev.tf
provider "azurerm" {
version = "=2.20.0"
features {}
}
resource "azurerm_resource_group" "this" {
name = "hones-rg-dev"
location = "North Europe"
}
module "signalr" {
source = "./modules/signalr"
name = var.signalr_name
resource_group_name = azurerm_resource_group.this.name
location = azurerm_resource_group.this.location
}
2. outputs.tf
output "primary_key" {
value = "${module.signalr.signalr_primary_key}"
}
output "primary_connection_string" {
value = "${module.signalr.signalr_primary_connection_string}"
}
We have all we need now. Let’s run the init command to initialize terraform and download all the required plugins:
terraform init
When all is initialized, run the following command to check if all is valid:
terraform validate
If everything checks out then you can run plan command:
terraform plan
and the apply command:
terraform apply
I am assuming that you have the Azure CLI installed on your machine and you’re already authenticated otherwise you won’t be able to apply your configuration.
Conclusion
In this post I spoke about how to use modules, input variables and output values in Terraform. In addition, I improved on the example from the previous post to use modules, input variables and output values. The source code can be found on GitHub. Please note that this post is part of the notes I gathered while I was learning about Terraform and is by no means a comprehensive guide. I’m also learning as I write this 
. Like always, thanks so much for taking your time to read and hopefully you have learnt something. Keep well and stay safe.
Comments