Terraform Azure Virtual Machine Automation

---

Provisioning Virtual Machine and related resources in Azure Platform using Terraform as IaC tool to deploy it.

• The terraform.yml automates Continuous Integration (CI) executing terraform initialize, validate, plan, and performs Continuous Dilivery (CD) that executes terraform apply to provision resources as stated in main.tf.

• The workflow_dispatch.yml allows manual trigger towards deprovisioning of the resources for cost effectiveness.

• The terraform.tfvars allows setup for multiple environments.

• The backend.tf configuration creates new Resource Group to avoid saving on runner's machine, enhancing security and effective team collaboration.

Important

Note: It is important to Create TFState Backup Resource Group by running backend.ps1 Prior to allow Terraform to initialize backend configuration.

Connecting new VM using ssh

Identifying new vm (ubuntu's) IP:

az vm list-ip-addresses --resource-group <RESOURCE_GROUP_NAME> --name <VM_NAME> --query "[].virtualMachine.network.publicIpAddresses[0].ipAddress" --output tsv

Just replace <RESOURCE_GROUP_NAME> with the name of your resource group and <VM_NAME> with the name of your VM. This command will return the public IP address of your VM.

ssh [email protected]

---

Architecture

---

Generating a Graph

Prerequisite

Install Graphviz locally via Homebrew:

brew install graphviz
dot -V # verify it by running

Generate a Terraform Plan Graph (Advanced):

• In your terminal, navigate to the directory containing your Terraform configuration files.
• Run terraform init to initialize the configuration.
• Run terraform plan -out=plan.out to create a plan.
• Run terraform graph -plan=plan.out | dot -Tsvg > graph.svg to generate an SVG file of your Terraform plan.

View the Generated Graph:

Open the graph.svg file in Visual Studio Code to view the visual representation of your Terraform resources and their dependencies.

---

Key Learnings

Please visit the WIKI page.

---

Acknowledgements

• Terraform Documentation
• Azure Documentation
• PowerShell: PowerShell file used to centralize terraform commands related to this project.
• GitHub: Platform for version control and collaboration.
• backend.ps1 - Reference: Microsoft Reactor Series

---

Improvement

adding random provider and using below

resource "random_string" "random" { length = 6 special = false upper = false }

name = "${lower(var.base_name)}${random_string.random.result}" # lower() function converts var.base_name if anything is captial. Azure does not allow any special or upper case characters.

for above use concept of output.tf, meaning anyrandom thing generated or unique based on the code, get the output, and use the output as input in main.tf pointing with .rg_name_out reference: https://www.youtube.com/watch?v=0YLPfSLbp9Y&t=1292s

Schedule VM Deallocation on Azure - NOT WORKING

resource "azurerm_virtual_machine_schedule" "shutdown" { virtual_machine_id = azurerm_virtual_machine.example.id location = azurerm_virtual_machine.example.location daily_recurring { time_zone = "UTC" time = "2200" } }

azurerm_virtual_machine is suspended with latest version, so when updating version, cgange to azurerm_linx_virtual_machine

In the Yamel file, try to make changes that it should not activate on the main branch, but on any other branch, so that can be tested before going to the main and try to configure something like rollback in case of any kind of a failure