Error in terraform module mainly to do with log analytics - terraform

Question

I am in the process of learning TF, and on the subject of modules, at the same time I have decided to only create resources on my Azure account using TF as a way to accelerate my learning. To this note, I found this gitbub repo https://github.com/kumarvna/terraform-azurerm-virtual-machine
I have been following the contents and trying to reproduce on my test system, I have tried to contact the author to no avail and felt I have already wasted 2 weeks trying to fix the problem, let me ask on here for help.
My setup.
Pulled the code from the repo onto my laptop.
logged onto my Azure account from a powershell console.
created a folder called create_vm
and in that folder
On my main.tf file, I have the following. This is a linux example, but I had the same issues with a windows example also.
# Azurerm provider configuration
provider "azurerm" {
features {}
}
# Creates a new resource group
resource "azurerm_resource_group" "test_build" {
name = "testBuild"
location = "West Europe"
}
# Creates a new network
resource "azurerm_virtual_network" "example" {
name = "example-network"
location = azurerm_resource_group.test_build.location
resource_group_name = azurerm_resource_group.test_build.name
address_space = ["10.0.0.0/16"]
dns_servers = ["10.0.0.4", "10.0.0.5"]
subnet {
name = "subnet1"
address_prefix = "10.0.1.0/24"
}
}
# Creates a new la workspace
resource "azurerm_log_analytics_workspace" "la" {
name = "loganalytics-we-sharedtest2"
resource_group_name = azurerm_resource_group.test_build.name
}
module "virtual-machine" {
source = "kumarvna/virtual-machine/azurerm"
version = "2.3.0"
# Resource Group, location, VNet and Subnet details
resource_group_name = azurerm_resource_group.test_build.name
location = "westeurope"
virtual_network_name = azurerm_virtual_network.example.name
subnet_name = "subnet1"
virtual_machine_name = "vm-linux"
# This module support multiple Pre-Defined Linux and Windows Distributions.
# Check the README.md file for more pre-defined images for Ubuntu, Centos, RedHat.
# Please make sure to use gen2 images supported VM sizes if you use gen2 distributions
# Specify `disable_password_authentication = false` to create random admin password
# Specify a valid password with `admin_password` argument to use your own password
# To generate SSH key pair, specify `generate_admin_ssh_key = true`
# To use existing key pair, specify `admin_ssh_key_data` to a valid SSH public key path.
os_flavor = "linux"
linux_distribution_name = "ubuntu2004"
virtual_machine_size = "Standard_B2s"
generate_admin_ssh_key = true
instances_count = 2
# Proxymity placement group, Availability Set and adding Public IP to VM's are optional.
# remove these argument from module if you dont want to use it.
enable_proximity_placement_group = true
enable_vm_availability_set = true
enable_public_ip_address = true
# Network Seurity group port allow definitions for each Virtual Machine
# NSG association to be added automatically for all network interfaces.
# Remove this NSG rules block, if `existing_network_security_group_id` is specified
nsg_inbound_rules = [
{
name = "ssh"
destination_port_range = "22"
source_address_prefix = "*"
},
{
name = "http"
destination_port_range = "80"
source_address_prefix = "*"
},
]
# Boot diagnostics to troubleshoot virtual machines, by default uses managed
# To use custom storage account, specify `storage_account_name` with a valid name
# Passing a `null` value will utilize a Managed Storage Account to store Boot Diagnostics
enable_boot_diagnostics = true
# Attach a managed data disk to a Windows/Linux VM's. Possible Storage account type are:
# `Standard_LRS`, `StandardSSD_ZRS`, `Premium_LRS`, `Premium_ZRS`, `StandardSSD_LRS`
# or `UltraSSD_LRS` (UltraSSD_LRS only available in a region that support availability zones)
# Initialize a new data disk - you need to connect to the VM and run diskmanagemnet or fdisk
data_disks = [
{
name = "disk1"
disk_size_gb = 100
storage_account_type = "StandardSSD_LRS"
},
{
name = "disk2"
disk_size_gb = 200
storage_account_type = "Standard_LRS"
}
]
# (Optional) To enable Azure Monitoring and install log analytics agents
# (Optional) Specify `storage_account_name` to save monitoring logs to storage.
log_analytics_workspace_id = azurerm_log_analytics_workspace.la.id
# Deploy log analytics agents to virtual machine.
# Log analytics workspace customer id and primary shared key required.
deploy_log_analytics_agent = true
log_analytics_customer_id = azurerm_log_analytics_workspace.la.workspace_id
log_analytics_workspace_primary_shared_key = azurerm_log_analytics_workspace.la.primary_shared_key
# Adding additional TAG's to your Azure resources
tags = {
ProjectName = "demo-project"
Env = "dev"
Owner = "user#example.com"
BusinessUnit = "CORP"
ServiceClass = "Gold"
}
}
on variables.tf.
variable "log_analytics_workspace_name" {
description = "The name of log analytics workspace name"
default = null
}
variable "storage_account_name" {
description = "The name of the hub storage account to store logs"
default = null
}
variable "create_resource_group" {
description = "Whether to create resource group and use it for all networking resources"
default = true
}
Please note that I added the create_resource_group variable to try to resolve my issue to no avail.
I then run
terraform init
terraform plan
I get the following error with terraform plan
│ Error: Error: Log Analytics workspaces "loganalytics-we-sharedtest2" (Resource Group "rg-shared-westeurope-01") was not found
│
│ with data.azurerm_log_analytics_workspace.example,
│ on main.tf line 6, in data "azurerm_log_analytics_workspace" "example":
│ 6: data "azurerm_log_analytics_workspace" "example" {
│
╵
╷
│ Error: Error: Resource Group "rg-shared-westeurope-01" was not found
│
│ with module.virtual-machine.data.azurerm_resource_group.rg,
│ on .terraform\modules\virtual-machine\main.tf line 27, in data "azurerm_resource_group" "rg":
│ 27: data "azurerm_resource_group" "rg" {
│
What have I done ?
Looked through the code to see what I am missing.
Added the variable at the top. Tried to contact the author to no
avail.
Tried to use an existing resource group, I feel this defeats the
purpose of having a variable that asks if a new resource group can
be created in case it doesn't already exist.
What else is confusing ?
I initially had another folder for modules, i later came to realise that the module is a public one being pulled down whenever I ran terraform init, now is there a way to have this as a localised module ?
I have made the changes recommended by the answer below, however in order not to turn the question into a long winded one, I have placed the error that I got below.
│ Error: Error: Subnet: (Name "subnet1" / Virtual Network Name "testBuild_vnet" / Resource Group "testBuild") was not found
│
│ with module.virtual-machine.data.azurerm_subnet.snet,
│ on .terraform\modules\virtual-machine\main.tf line 36, in data "azurerm_subnet" "snet":
│ 36: data "azurerm_subnet" "snet" {
│
╵
╷
│ Error: Invalid count argument
│
│ on .terraform\modules\virtual-machine\main.tf line 443, in resource "azurerm_monitor_diagnostic_setting" "nsg":
│ 443: count = var.existing_network_security_group_id == null && var.log_analytics_workspace_id != null ? 1 : 0
│
│ The "count" value depends on resource attributes that cannot be determined until apply, so Terraform cannot predict how many instances will be created. To work around this, use the -target argument to first apply only the resources that the count depends on.

Answer 1

I think the misunderstanding is that you think the module creates a resource group, but that is not the case. This module expects an already existing resource group as var.resource_group_name (same goes for the input variables virtual_network_name, subnet_name and log_analytics_workspace_id).
The main difference between the resource_ and data_ prefix is that data sources are read-only and "only" fetch already existing infrastructure for further use:
Data sources allow Terraform to use information defined outside of
Terraform, defined by another separate Terraform configuration, or
modified by functions.
https://www.terraform.io/language/data-sources
So in your case it should work like (not tested):
# Azurerm provider configuration
provider "azurerm" {
features {}
}
# Creates a new resource group
resource "azurerm_resource_group" "test_build" {
name = "testBuild"
location = "West Europe"
}
# Creates a new network
resource "azurerm_virtual_network" "example" {
name = "example-network"
location = azurerm_resource_group.test_build.location
resource_group_name = azurerm_resource_group.test_build.name
address_space = ["10.0.0.0/16"]
dns_servers = ["10.0.0.4", "10.0.0.5"]
subnet {
name = "subnet1"
address_prefix = "10.0.1.0/24"
}
}
# Creates a new la workspace
resource "azurerm_log_analytics_workspace" "la" {
name = "loganalytics-we-sharedtest2"
resource_group_name = azurerm_resource_group.test_build.name
}
module "virtual-machine" {
source = "kumarvna/virtual-machine/azurerm"
version = "2.3.0"
# Resource Group, location, VNet and Subnet details
resource_group_name = azurerm_resource_group.test_build.name
location = "westeurope"
virtual_network_name = azurerm_virtual_network.example.name
subnet_name = "subnet1"
virtual_machine_name = "vm-linux"
# This module support multiple Pre-Defined Linux and Windows Distributions.
# Check the README.md file for more pre-defined images for Ubuntu, Centos, RedHat.
# Please make sure to use gen2 images supported VM sizes if you use gen2 distributions
# Specify `disable_password_authentication = false` to create random admin password
# Specify a valid password with `admin_password` argument to use your own password
# To generate SSH key pair, specify `generate_admin_ssh_key = true`
# To use existing key pair, specify `admin_ssh_key_data` to a valid SSH public key path.
os_flavor = "linux"
linux_distribution_name = "ubuntu2004"
virtual_machine_size = "Standard_B2s"
generate_admin_ssh_key = true
instances_count = 2
# Proxymity placement group, Availability Set and adding Public IP to VM's are optional.
# remove these argument from module if you dont want to use it.
enable_proximity_placement_group = true
enable_vm_availability_set = true
enable_public_ip_address = true
# Network Seurity group port allow definitions for each Virtual Machine
# NSG association to be added automatically for all network interfaces.
# Remove this NSG rules block, if `existing_network_security_group_id` is specified
nsg_inbound_rules = [
{
name = "ssh"
destination_port_range = "22"
source_address_prefix = "*"
},
{
name = "http"
destination_port_range = "80"
source_address_prefix = "*"
},
]
# Boot diagnostics to troubleshoot virtual machines, by default uses managed
# To use custom storage account, specify `storage_account_name` with a valid name
# Passing a `null` value will utilize a Managed Storage Account to store Boot Diagnostics
enable_boot_diagnostics = true
# Attach a managed data disk to a Windows/Linux VM's. Possible Storage account type are:
# `Standard_LRS`, `StandardSSD_ZRS`, `Premium_LRS`, `Premium_ZRS`, `StandardSSD_LRS`
# or `UltraSSD_LRS` (UltraSSD_LRS only available in a region that support availability zones)
# Initialize a new data disk - you need to connect to the VM and run diskmanagemnet or fdisk
data_disks = [
{
name = "disk1"
disk_size_gb = 100
storage_account_type = "StandardSSD_LRS"
},
{
name = "disk2"
disk_size_gb = 200
storage_account_type = "Standard_LRS"
}
]
# (Optional) To enable Azure Monitoring and install log analytics agents
# (Optional) Specify `storage_account_name` to save monitoring logs to storage.
log_analytics_workspace_id = azurerm_log_analytics_workspace.la.id
# Deploy log analytics agents to virtual machine.
# Log analytics workspace customer id and primary shared key required.
deploy_log_analytics_agent = true
log_analytics_customer_id = azurerm_log_analytics_workspace.la.workspace_id
log_analytics_workspace_primary_shared_key = azurerm_log_analytics_workspace.la.primary_shared_key
# Adding additional TAG's to your Azure resources
tags = {
ProjectName = "demo-project"
Env = "dev"
Owner = "user#example.com"
BusinessUnit = "CORP"
ServiceClass = "Gold"
}
}
Just adding a new variable called create_resource_group will not do anything as long as there is no corresponding logic/code behind it.