Monday, May 24, 2021

How to create Azure Container Registry using Terraform in Azure Cloud | Setup Azure Container Registry using Terraform

Hashicorp's Terraform is an open-source tool for provisioning and managing cloud infrastructure. Terraform can provision resources on any cloud platform. 

Terraform allows you to create infrastructure in configuration files(tf files) that describe the topology of cloud resources. These resources include virtual machines, storage accounts, and networking interfaces. The Terraform CLI provides a simple mechanism to deploy and version the configuration files to Azure.

Watch the steps in YouTube:

Advantages of using Terraform:

  • Reduce manual human errors while deploying and managing infrastructure.
  • Deploys the same template multiple times to create identical development, test, and production environments.
  • Reduces the cost of development and test environments by creating them on-demand.

How to Authenticate with Azure?

Terraform can authenticate with Azure in many ways, in this example we will use Azure CLI to authenticate with Azure and then we will create resources using Terraform.


Azure CLI needs to be installed.

Terraform needs to be installed.

Logging into the Azure CLI

Login to the Azure CLI using:

az login

The above command will open the browser and will ask your Microsoft account details. Once you logged in, you can see the account info by executing below command:

az account list

Now create a directory to store Terraform files.

mkdir tf-acr

cd tf-acr

Let's create a terraform file to use azure provider. To configure Terraform to use the Default Subscription defined in the Azure CLI, use the below cod.

Now initialize the working directory

sudo vi

provider "azurerm" {
  features {}
resource "azurerm_resource_group" "rg" {
  name     = "rg-tf-acr"
  location = "southcentralus"
resource "azurerm_container_registry" "acr" {
  name                     = "azcontainerregistry321"
  resource_group_name      =
  location                 = azurerm_resource_group.rg.location
  sku                      = "Basic"
  admin_enabled            = true
output "admin_password" {
  value       = azurerm_container_registry.acr.admin_password
  description = "The object ID of the user"
sensitive = true

Perform the below command to initialize the directory.

terraform init

Once directory is initialized, you can start writing code for setting up the infrastructure. Now Perform the below command to validate terraform files.

terraform validate

perform plan command to see how many resources will be created.

terraform plan

terraform apply

Do you want to perform these actions?

type yes 

Now Login into Azure Cloud to see the resources created.

How to destroy the resources ?
Execute terraform destroy

The above command to destroy both resource group and container registry created before.

Thursday, May 13, 2021

How to store Terraform state file in S3 Bucket | How to manage Terraform state in S3 Bucket?

One of the things that Terraform does (and does really well) is “tracks” your infrastructure that you provision. It does this through the means ofstate.

By default, Terraform stores state locally in a file named terraform.tfstate. This does not work well in a team environment where if any developer wants to make a change he needs to make sure nobody else is updating terraform in the same time.

Why state file should not be stored in your local machine?

  • Local state doesn't work well in a team or collaborative environment.
  • Terraform state can include sensitive information.
  • Storing state locally increases the chance of inadvertent deletion.
With remote state, Terraform writes the state data to a remote data store, which can then be shared between all members of a team. Terraform supports storing state in many ways including the below:

  • Terraform Cloud
  • HashiCorp Consul
  • Amazon S3
  • Azure Blob Storage
  • Google Cloud Storage
  • Alibaba Cloud OSS
  • Artifactory or Nexus 

We will learn how to store state file in AWS S3 bucket. We will be creating S3 bucket and also create Dynamo Table where we will be storing lock.

Watch the steps in YouTube Channel:



mkdir project-terraform

cd project-terraform

First let us create necessary terraform files.

Create tf files

sudo vi

variable "region" {
    default = "us-east-2"

variable "instance_type" {
    default = "t2.micro"

sudo vi
provider "aws" {
  region     = "${var.region}"

#1 -this will create a S3 bucket in AWS
resource "aws_s3_bucket" "terraform_state_s3" {
  bucket = "terraform-coachdevops-state"
  force_destroy = true
# Enable versioning to see full revision history of our state files
  versioning {
         enabled = true

# Enable server-side encryption by default
server_side_encryption_configuration {
    rule {
      apply_server_side_encryption_by_default {
        sse_algorithm = "AES256"

# 2 - this Creates Dynamo Table
resource "aws_dynamodb_table" "terraform_locks" {
  name         = "tf-up-and-run-locks"
  billing_mode = "PAY_PER_REQUEST"
  hash_key     = "LockID"
        attribute {
         name = "LockID"
         type = "S"

now let us initialize terraform.

terraform init

terraform apply


this will create two resources - S3 bucket and AWS Dynamo table. But still terraform state file is stored locally. if you type below command, you will see the tfstate file locally.

 ls -al

To store state file remotely, you need to add following code with terraform block. This will add backend configuration.

sudo vi

#Step 3 - Creates S3 backend
terraform {
  backend "s3" {
    #Replace this with your bucket name!
    bucket         = "terraform-coachdevops-state"
    key            = "dc/s3/terraform.tfstate"
    region         = "us-east-2"
    #Replace this with your DynamoDB table name!
    dynamodb_table = "tf-up-and-run-locks"
    encrypt        = true

terraform init

and then type yes and enter. Now you see local state file has 0 bytes(empty)

Now login to AWS console, Click on S3, Click on the bucket name

Now you should be able to see tfstate file in S3.

Click on the terraform.tfstate file, you can see multiple version of your state file. Terraform is automatically pushing and pulling state data to and from S3.

How to perform Destroy?

It is not that straight forward as back end is referencing S3 bucket, if we delete S3 bucket, back end will not where to reference. So we need to perform below steps to perform destroy:

1. remove back end reference in the by commenting out backend section of the code.

sudo vi

remove the below code or comment out:


terraform {
  backend "s3" {
    #Replace this with your bucket name!
    bucket         = "terraform-coachdevops-state"
    key            = "dc/s3/terraform.tfstate"
    region         = "us-east-2"
    #Replace this with your DynamoDB table name!
    dynamodb_table = "tf-up-and-run-locks"
    encrypt        = true


We need to initialize again. so type below command

terraform init -migrate-state

type yes

Now you will see the local state file have been updated.

Now perform you can delete all the resources created by Terraform including S3 bucket and Dynamo table.

terraform destroy