Using the for each Meta Argument

The Terraform for_each meta-argument is a flexible alternative to count when you need to create multiple resources that are not identical. Unlike count, which iterates a numeric index, for_each iterates over a map or a set of strings. Each instance created by for_each is identified by a stable key (the map key or set element), which helps prevent accidental resource replacement when items are added or removed. Use for_each when you want:

Unique names, sizes, tags, or other attributes per resource.
Stable identities so Terraform can track each instance reliably across changes.

Prefer for_each when resources are unique or require stable identities. count is still useful for N identical resources, but for_each reduces accidental replacements when items are added or removed.

Example: define a variable that is a map of VM names to VM sizes, then use for_each to create one VM per map entry. Inside the resource block you can reference the map key with each.key and the value with each.value.

variable "vms" {
  type = map(string)
  default = {
    "vm1" = "Standard_D2s_v3"
    "vm2" = "Standard_D4s_v3"
    "vm3" = "Standard_B2ms"
  }
}

resource "azurerm_virtual_machine" "example" {
  for_each            = var.vms
  name                = each.key
  location            = "East US"
  resource_group_name = "my-resource-group"
  vm_size             = each.value

  os_profile {
    computer_name  = each.key
    admin_username = "adminuser"
  }

  # ...other required blocks (e.g., network_interface_ids, storage_os_disk)...
}

Referencing created instances

Use the map-keyed resource address: azurerm_virtual_machine.example["vm1"]
Or reference attributes: azurerm_virtual_machine.example["vm1"].id

Interpolation examples

You can combine each.key in strings:

os_profile {
  computer_name = "web-${each.key}"
  admin_username = "adminuser"
}

Types and conversions

for_each accepts:
- a map (map(string) etc.) — each.key is the map key and each.value is the corresponding value.
- a set of strings — each instance key is the set element and each.value equals that element.
If you have a list, convert it to a set to remove ordering and duplicates: for_each = toset(var.my_list)
If you need unique keys from a list, turn it into a map using zipmap or another transformation, for example:
for_each = zipmap(var.names, var.sizes)
Here var.names becomes the keys and var.sizes the values.

Comparison: for_each vs count

Use case	When to prefer	Example
`for_each`	Resources need unique identities or attributes (names, tags, sizes). Safe when adding/removing items.	`for_each = var.vms`
`count`	Create N identical resources where the individual identity doesn’t matter.	`count = 3`

Changing the keys used by for_each (for example renaming map keys) is a breaking change: Terraform treats the new key as a new resource and will destroy the old resource. Rename resources deliberately and consider lifecycle or terraform state commands for complex migrations.

Best practices and tips

Use meaningful map keys (like names or unique IDs). These keys are the stable identifiers Terraform uses to track instances.
Prefer maps when you need different attributes per instance; use sets for simple lists of unique names.
For large numbers of identical resources where identity isn’t important, count may be more concise.
When converting lists, remember toset() removes ordering — use it when ordering is irrelevant.
Use explicit depends_on or lifecycle blocks sparingly; stable keys usually reduce the need for complex lifecycle rules.

References and further reading

Terraform documentation: https://www.terraform.io/docs/language/meta-arguments/for_each.html
Transformations: toset(), zipmap(), and other builtin functions — see the Functions reference in Terraform docs.

Course Introduction

Preparing Your Environment

Terraform Foundations

The Core Terraform Workflow

Terraform CLI

Terraform File Structure and Organization

Terraform Configuration - Fundamentals

Understading and Managing Terraform State

Marking Code Reusable

Refactoring Terraform State

Terraform Modules

Managing Resource Behavior and Dependencies

Securing Terraform Configurations

Terraform Troubleshooting

HCP Terraform

Using the for each Meta Argument

Watch Video

Practice Lab