Simplifying Infrastructure Automation with For Loops in Terraform: A Practical Primer

In the recent years, most infrastructures have been slowly and totally moved from onsite servers over to the cloud. With this change, there has also been the need for creating infrastructures in the cloud.

One tool that has really helped in building our cloud infrastructure is Ansible even though it is a Configuration Management tool. However we recently switched over to Terraform and so far it has proven itself as being much better than expected. Today we will be talking about Terraform For Loops.

Terraform is the most popular infrastructure as a code tool which allows you to define and manage any and all cloud (AWS, GCP, Azure, Kubernetes, etc.) resources.

Like any language, Terraform also works with for loops. It was quite confusing when we started working with it. I recently decided to write about how we finally figured out how it works.

Like for Loops in any coding language, the for loops in Terraform provide a great mechanism for iterating over any list of items and accomplishing some actions and/or creating resources dynamically.

This functionality enables efficient and flexible infrastructure provisioning, which reduces manual work while increasing scalability.

In this post, I will walk you through the syntax and benefits of for loops in Terraform, providing practical insights and examples to help you properly exploit their power.

Understanding For Loops in Terraform

To utilise for loops, first define a variable that represents the list of things you wish to iterate over, then control the iteration with the for_each or count expressions.

The count statement is appropriate for iterating a fixed number of times, whereas the for_each expression is very beneficial for dealing with a dynamic set of elements, such as a map or set.

But what do I mean by "map" and "set"?

The for_each map is similar to Python dictionaries in that variables have attributes with keys translated to values, and the set is comparable to a list in that it is an ordered collection of objects.

Here's an example of Terraform code that shows how to use for_each and count expressions in different scenarios:

# Dynamic set of items using for_each
variable "dynamic_items" {
  type    = map(string)
  default = {
    ubuntu_server = "ami-01dd271720c1ba44f“ 
    windows_server = "ami-0274fd9e256dea7b1“ 
    rhel_server = "ami-013d87f7217614e10“ 
  }
}

resource "aws_instance" "servers" {
  for_each = var.dynamic_items

  ami           = each.value
  instance_type = "t2.micro"

  tags = {
    Name = each.key
  }
}

The for_eachexpression is used in this example to generate AWS EC2 instances depending on the dynamic set of items given in the dynamic_itemsvariable. Each map entry contains a server name and its associated type - machine image (ami). For each server in the map, an aws_instanceresource will be generated, and the instance tags will be set using the for_eachiterator.

The countexpression, on the other hand, is used to iteratively construct AWS EBS volumes based on the fixed count given in the fixed_countvariable. The aws_ebs_volumeresource will be generated three times, each with a unique name supplied via the count.index variable (DataVolume-1, DataVolume-2, DataVolume-3).

# Fixed number of iterations using count
variable "fixed_count" {
  type    = number
  default = 3
}

resource "aws_ebs_volume" "data_volume" {
  count          = var.fixed_count
  availability_zone = "us-west-1a"
  size           = 100
  volume_type    = "gp2"

  tags = {
    Name = "DataVolume-${count.index + 1}"
  }
}

Instead of manually specifying many resource blocks or modules, you can dynamically produce resources using for loops, which avoids the need for repetitious code and decreases the likelihood of errors.

They also provide for easy scalability, allowing you to adapt to changing requirements and handle larger infrastructure installations with ease. For loops provide a straightforward way for scaling your infrastructure, whether you need to launch numerous EC2 instances, provision different subnets, or configure multiple security groups.

In addition to automation, they offer conditional statements, which allow you to control the iteration process and make dynamic decisions based on specified criteria by incorporating if conditions and logical expressions into for loops.

Best Practices

Maintaining code clarity and organisation when dealing with for loops in Terraform is critical for long-term maintainability and collaboration. Consider the following best practises:

• Use descriptive variable names: To improve code readability, use descriptive variable names. This makes it easy for others (and you) to comprehend the loop's purpose and context.

• Include comments: Include comments in your code to explain and describe the intent behind the for loop. This assists others in understanding the logic and goal of the repetition.

• Indentation and formatting: Always use proper indentation and formatting when writing for loops.

• Keep loop complexity to a minimum: Avoid loops with nested iterations or lengthy conditional expressions. Code that is overly complex might be difficult to understand and maintain.

• Test and validate: Thoroughly test and validate the functionality of for loops before deploying them in production systems.

Finally

To summarise, Terraform's for loops enable you to automate resource creation and configuration while allowing flexibility via conditional expressions. Furthermore, the use of conditional statements enables you to react to various conditions, making your infrastructure settings more dynamic and adaptive to changing requirements.