Terraform Tutorial — Resource Lifecycle

1 minute read

The lifecycle block gives you fine-grained control over how Terraform creates, updates, and destroys resources. These settings are essential for production infrastructure where you need to avoid downtime or prevent accidental deletion.

create_before_destroy

By default, Terraform destroys the old resource, then creates the new one. This causes downtime. create_before_destroy reverses the order:

resource "aws_instance" "web" {
  ami           = "ami-0c55b159cbfafe1f0"
  instance_type = "t3.micro"

  lifecycle {
    create_before_destroy = true
  }
}

When to use: Resources that can’t have downtime — instances behind a load balancer, databases, etc.

prevent_destroy

Blocks Terraform from destroying a resource, even during terraform destroy:

resource "aws_s3_bucket" "critical_data" {
  bucket = "company-critical-data"

  lifecycle {
    prevent_destroy = true
  }
}

If someone tries to destroy this resource:

Error: Instance cannot be destroyed

  Resource aws_s3_bucket.critical_data has lifecycle.prevent_destroy set,
  but the plan calls for this resource to be destroyed.

When to use: Databases, S3 buckets with important data, production infrastructure you never want accidentally deleted.

💡 Tip: prevent_destroy protects against terraform destroy and plan changes that would replace a resource. To remove the resource from Terraform’s management, first remove prevent_destroy, then remove the resource block.

ignore_changes

Tells Terraform to ignore changes to specific attributes. Useful when something outside Terraform modifies a resource:

resource "aws_instance" "web" {
  ami           = "ami-0c55b159cbfafe1f0"
  instance_type = "t3.micro"

  lifecycle {
    ignore_changes = [ami, tags]
  }
}

Now if someone updates the AMI or tags through the AWS console, Terraform won’t try to revert those changes.

Ignore All Changes

lifecycle {
  ignore_changes = all
}

This makes Terraform manage the resource’s creation and deletion but ignore all attribute changes. Useful for resources managed primarily outside Terraform.

replace_triggered_by

Force a resource to be replaced when another resource or value changes:

resource "aws_instance" "web" {
  ami           = "ami-0c55b159cbfafe1f0"
  instance_type = "t3.micro"

  lifecycle {
    replace_triggered_by = [
      aws_security_group.web.id
    ]
  }
}

If the security group is replaced, the instance is also replaced.

Lifecycle Summary

Setting Effect
create_before_destroy New resource created before old one is destroyed
prevent_destroy Blocks any plan that would destroy the resource
ignore_changes Ignores changes to specific attributes
replace_triggered_by Forces replacement when referenced resources change

What’s Next?

You understand lifecycle management. Next, let’s get an overview of all the meta-arguments available on resources — count, for_each, depends_on, provider, and lifecycle.

← Resource Dependencies Meta-Arguments Overview →

You may also enjoy