Amazon Web Services Question posted in
The official Amazon Web Services documentation can be found here.

Amazon web services – How to raise your own (custom) error in terraform?

My goal here is to raise a custom error instead of the error received from terraform for this sample code:

resource "aws_instance" "example" {
  ami           = "ami-0f"  # Invalid AMI ID intentionally causing an error
  instance_type = "t2.micro"
}

for this code I got an error output:

aws_instance.example: Creating...
╷
│ Error: creating EC2 Instance: InvalidAMIID.NotFound: The image id '[ami-0f]' does not exist
│       status code: 400, request id: 2d665c80-8918-4940-92cc-11f7a14ecd57
│ 
│   with aws_instance.example,
│   on main.tf line 27, in resource "aws_instance" "example":
│   27: resource "aws_instance" "example" {
│ 
╵

The above error is a proper error.. But I want to create a custom error like below:

Custom message: Failed to create the AWS instance.

Kindly give suggestions on:

  1. How to display my custom error instead of the proper error displayed by terraform?
  2. and also How to display my custom error along with the proper error displayed by terraform?

If you have any standard logics for error handling and exception handling in terraform please do share.

How to add my own custom error messages (like we do exceptions handling in python) for any type of errors caused by terraform ?

Thank you

Tags:

2

Answers


  1. 0

    One option would be to specify custom validation rules for your variables.

    Quick example with the variable validation:

    variable "ami" {
  description = "The AMI to use for the server"
  type        = string

  validation {
    condition     = can(regex("^ami-[a-f0-9]{17}$", var.ami))
    error_message = "The ami variable must be a valid AMI ID in the format ami-xxxxxxxxxxxxxxxxx."
  }
}

resource "null_resource" "aws_instance" {
  triggers = {
    ami = var.ami
  }
}

    Notes:

    • According to this post, the Amazon Machine Image (AMI) ID should start with ami- followed by a string of 17 alphanumeric characters.
    • The can function evaluates the given expression and returns a boolean value indicating whether the expression produced a result without any errors.

    Running terraform plan -var 'ami=foobar' (invalid value):

    │ Error: Invalid value for variable
│ 
│   on main.tf line 1:
│    1: variable "ami" {
│     ├────────────────
│     │ var.ami is "foobar"
│ 
│ The ami variable must be a valid AMI ID in the format ami-xxxxxxxxxxxxxxxxx.
│ 
│ This was checked by the validation rule at main.tf:5,3-13.

    See also custom conditions in order to produce custom error messages for several types of objects in a configuration.

    Login or Signup to reply.
  2. 0

    As far as I know you can’t raise a custom error from terraform, but you can achieve the desired result by capturing the error message from the terraform process and then handling it accordingly, e.g.:

    #!/bin/bash
TFOUT="$(terraform apply -auto-approve 2>&1)"
if [ "$?" -ne 0 ]; then
  if grep -q 'InvalidAMIID.NotFound' <<<"$TFOUT"; then
    echo "Custom message: Failed to create the AWS instance."
  fi
fi

    you can extend this to add handling for any error message.

    Login or Signup to reply.
Please signup or login to give your own answer.

Back To Top
Search