Skip to main content
In this article, we explain how to use Terraform’s taint and untaint commands to manage resource recreation effectively. These commands are especially useful when a resource fails during creation or when manual changes occur that necessitate a fresh deployment.

​
Overview

Terraform marks a resource as tainted when it encounters errors during creation, such as a failed provisioner command. A tainted resource is scheduled for replacement during the next apply. Conversely, you can use the untaint command to clear this status and prevent a replacement.
Using taint and untaint commands allows for efficient control of resource lifecycle without a complete destroy and reapply cycle.

​
Scenario: Tainted Resource due to Provisioner Failure

Consider a scenario where an AWS EC2 instance is provisioned using a local provisioner to store its public IP address in a file. If the provisioner command fails—perhaps because the file path is incorrect—the resource is marked as tainted, triggering its replacement on the next apply.

​
Resource Definition Example

resource "aws_instance" "webserver-3" {
  ami           = "ami-0edab43b6fa892279"
  instance_type = "t2.micro"
  key_name      = "ws"

  provisioner "local-exec" {
    command = "echo ${aws_instance.webserver-3.public_ip} > /temp/pub_ip.txt"
  }
}

​
Applying the Configuration

When executing the apply command, you might see output indicating that the provisioner has failed: 
$ terraform apply
Plan: 1 to add, 0 to change, 0 to destroy.

aws_instance.webserver: Creating...
aws_instance.webserver: Still creating... [10s elapsed]
aws_instance.webserver: Still creating... [20s elapsed]
aws_instance.webserver: Still creating... [30s elapsed]
aws_instance.webserver: Provisioning with 'local-exec'...
aws_instance.webserver (local-exec): Executing: ["cmd" "/C" "echo 35.183.14.192 > /temp/pub_ip.txt"]
aws_instance.webserver (local-exec): The system cannot find the path specified.

Error: Error running command 'echo 35.183.14.192 > /temp/pub_ip.txt': exit status 1. Output: The system
At this point, Terraform marks the “webserver” resource as tainted.

​
Verifying the Tainted Resource

Running the terraform plan command confirms that the tainted resource is scheduled for replacement: 
$ terraform plan
Refreshing Terraform state in-memory prior to plan...
The refreshed state will be used to calculate this plan, but will not be
persisted to local or remote state storage.

aws_instance.webserver: Refreshing state... [id=i-0dba2d5dc22a9a904]
------------------------------------------------------------------------------------------------

An execution plan has been generated and is shown below.
Resource actions are indicated with the following symbols:
-/+ destroy and then create replacement

Terraform will perform the following actions:

# aws_instance.webserver is tainted, so must be replaced
-/+ resource "aws_instance" "webserver-3" {
Double-check your resource configuration and provisioner commands to avoid unintentional resource replacement.

​
Forcing a Resource Rebuild

There are situations where you might want to deliberately force a resource rebuild. For example, if manual changes—such as updating the Nginx version—were made on an AWS instance, you can efficiently trigger the recreation of that resource without performing a full destroy and apply cycle.

​
Tainting the Resource

Run the following command to mark the resource as tainted: 
$ terraform taint aws_instance.webserver
Resource instance aws_instance.webserver has been marked as tainted.

​
Confirming the Change with Terraform Plan

After tainting, a terraform plan will show that the resource is scheduled for replacement: 
$ terraform plan
Refreshing Terraform state in-memory prior to plan...
The refreshed state will be used to calculate this plan, but will not be persisted to local or remote state storage.

aws_instance.webserver: Refreshing state... [id=i-0fd3946f5b3ab8af8]
--------------------------------------------------------------------------

An execution plan has been generated and is shown below.
Resource actions are indicated with the following symbols:
    -/+ destroy and then create replacement

Terraform will perform the following actions:

  # aws_instance.webserver is tainted, so must be replaced
  -/+ resource "aws_instance" "webserver" {

​
Reversing Taint: Using the Untaint Command

If you later decide that a resource should not be replaced, you can remove its tainted state by using the untaint command. This prevents Terraform from destroying and recreating the resource during the next apply. 
$ terraform untaint aws_instance.webserver
Resource instance aws_instance.webserver has been successfully untainted.

​
Summary Table

CommandActionDescription
terraform taintMark resource as taintedForces the resource to be replaced on the next apply
terraform untaintRemove taint from resourcePrevents resource replacement during the next apply
terraform planVerify resource replacement planConfirms which resources are marked for replacement
terraform applyApply configuration changesExecutes resource creation and replacement operations
By leveraging the taint and untaint commands, you can manage resource lifecycle events efficiently, ensuring that your infrastructure remains consistent with your desired state. For further details, explore the Terraform Documentation.

Watch Video