terraform-provider-azurerm: azurerm_application_insights Unable to Deploy or Import App Insights

Is there an existing issue for this?

I have searched the existing issues

Community Note

Please vote on this issue by adding a 👍 reaction to the original issue to help the community and maintainers prioritize this request
Please do not leave “+1” or “me too” comments, they generate extra noise for issue followers and do not help prioritize the request
If you are interested in working on this issue or have submitted a pull request, please leave a comment

Terraform Version

1.1.2

AzureRM Provider Version

3.7.0

Affected Resource(s)/Data Source(s)

azurerm_application_insights

Terraform Configuration Files

app_insights.tf

module "application_insights_int_sbx_01" {
  source              = "../../modules/application_insights"
  name                = "appin-${lookup(local.short_region_names, var.location)}-${var.environment_code}-sh-int-01"
  location            = module.azurerm_resource_group_int_01.location
  resource_group_name = module.azurerm_resource_group_int_01.name
  workspace_id        = module.log_ana_wrksp_01.id

  tags = {
    System      = "App Insight"
    Environment = var.environment
    Billing     = "Integration"
  }
}

main.tf

resource "azurerm_application_insights" "appin_int_01" {
  name                = var.name
  location            = var.location
  resource_group_name = var.resource_group_name
  workspace_id        = var.workspace_id
  application_type    = var.application_type
  tags                = var.tags
}

variables.tf
variable "name" {
  description = "(Required) Specifies the name of the Application Insights component. Changing this forces a new resource to be created."
  type        = string
}

variable "location" {
  description = "(Required) Specifies the supported Azure location where the resource exists. Changing this forces a new resource to be created."
  type        = string
  default     = "uksouth"
}

variable "resource_group_name" {
  description = "(Required) The name of the resource group in which to create the Application Insights component."
  type        = string
}

variable "workspace_id" {
  description = "(Optional) Specifies the id of a log analytics workspace resource"
  type        = string
}

variable "application_type" {
  description = "(Optional) Specifies the id of a log analytics workspace resource"
  type        = string
  default     = "web"
}

variable "tags" {
  description = "Tags to add to the application insight."
}

Debug Output/Panic Output

│ Error: making Read request on AzureRM Application Insights Billing Feature 'appin-uks-pre-sh-int-01': insights.ComponentCurrentBillingFeaturesClient#Get: Failure responding to request: StatusCode=200 -- Original Error: Error occurred unmarshalling JSON - Error = 'json: cannot unmarshal array into Go value of type insights.ApplicationInsightsComponentBillingFeatures' JSON = '[]'

Expected Behaviour

Terraform apply should create the application Insights instance Terraform import should import the application Insights instance

Actual Behaviour

Terraform apply results in the above error, the app insights instance is created. On the next apply Terraform errors that the resource already exists. Terraform Import to try and workaround this fails with the same error so no workaround possible.

Steps to Reproduce

Terraform apply

Important Factoids

Azure UK South

References

No response

About this issue

Original URL
State: closed
Created 2 years ago
Reactions: 57
Comments: 30 (7 by maintainers)

Most upvoted comments

This must be Azure API problem, terraform is just victim here. Indeed resources get created at all times.

I have opened SevB ticket with Azure, I’m working with them, passing on the information gathered here.

Hope they will fix it soon!

+14

Tbohunek on Jun 15, 2022

I too have this issue. The same script works on another subscription with same RBAC access, same region. Only difference in subscription type - does NOT work on subscription part of Azure EA Enterprise agreement, but works fine on ‘pay-as-you-go’. Not sure if this is related, but the error message seems to related to billing (Insights Billing Feature)! Question - is everyone facing this issue on subscription part of EA?

+10

jain-ranjith on Jun 14, 2022

Hi all,

I can confirm this issue, I’m unable to create an AppInsights from TF correctly 😦 But the AppInsights seem to be created in Azure, so on next run it saying that AI already exists 😦

Regards, Alexandre

alex-3sr on Jun 15, 2022

There is a service heath issue for this: Application Insights - Limited resources in Application Insights - Extended Investigation

Status Active Health Event Type Service Issue Start time 8.6.2022 12.17.42 (for 8 dage siden) End Time

Last update Impact Statement: Starting at 10:17 UTC on 8 June 2022, you have been identified as a customer using Application Insights who may be experiencing issues with your subscriptions that don’t have any existing Application Insights resources, creating a new Application Insights resource via Terraform may fail with error. The “Usage and estimated costs” page of the Application Insights resource may not show the daily cap information.

Current Status: We have engaged additional teams and are actively working on rolling out the hotfix to mitigate the issue. The next update will be provided in 2 hours or as events warrant.

Workaround: Creating a sub-resource (e.g., web test) in the impacted Application Insights component might fix the issue for the whole subscription. If the issue persists, try refreshing the page or opening the impacted blade in a new browser. That sub-resource (e.g., web test) could be deleted immediately once the issue goes away.

https://app.azure.com/h/3_TK-3C0/0ef747

sorenhuus on Jun 17, 2022

Until then, I put in place workaround to convert to data after the initial run creates it but fails to put in state. Maybe it helps someone…

variable "override_appinsights_bug" {
  type        = bool
  description = "Set to true after initial run - github.com/hashicorp/terraform-provider-azurerm/issues/17229"
  default     = false
}

resource "azurerm_application_insights" "example" {
  count               = var.override_appinsights_bug ? 0 : 1
  ...
}

data "azurerm_application_insights" "example" {
  count               = var.override_appinsights_bug ? 1 : 0
  ...
}

resource "azurerm_example" "example" {
  ...
  application_insights_key = (var.override_appinsights_bug
    ? data.azurerm_application_insights.example[0].instrumentation_key
    : azurerm_application_insights.example[0].instrumentation_key
  )
}

Tbohunek on Jun 15, 2022

We are getting the same issue. We can also produce a similar error when running the following command az monitor app-insights component billing show --app xxx-appi -g xxx-rg This command succeed on other subscriptions where we didn’t get the terraforming issue.

chesnelg on Jun 14, 2022

Yawheeew, so the ticket worked! That was quick.

Indeed, creating classic test under availability section of the app insight fixes the issue for all insights within the subscription.

ServiceHealth event issued ✅
workaround available ✅
issue mitigated ✅

That gives me hope they will fix it very soon.

Tbohunek on Jun 25, 2022

Same issue in Terraform 1.2.2 and AzureRM version 3.9.0 and 3.10.0

asensionacher on Jun 14, 2022

@RanjithJain77 I have 2 subscriptions, both EA, offer MS-AZR-0017P. One of them works, the other fails.

Tbohunek on Jun 14, 2022

I can confirm that creating a ‘classic test’ for our App Insights resource in the affected subscription has resolved this same issue with our terraform deployment. Thanks @Tbohunek

mattwin1 on Jun 17, 2022

I found this issue to day.

Terraform Version

1.2.2

AzureRM Provider Version

3.10

RiddlerKnight on Jun 16, 2022

Hi, I’m facing same issue, but why it throwing error if code 200 - means success?

jul17 on Jun 14, 2022

It’s also now working for me without implementing the workaround in the same subscription I initially encountered the issue in. Guessing MS have fixed it and are rolling out the fix.

davidsampson-hv on Jun 19, 2022

For me it worked today even without the workaround, seems that the issue has been fixed overnight 😃

Negative. I just created in a new subscription and still missing cap there. @steindan1 I’d call you lucky. 😃 Servicehealth is also still open.

Tbohunek on Jun 17, 2022

For me it worked today even without the workaround, seems that the issue has been fixed overnight 😃

daniel-steinmann on Jun 17, 2022

Seems its been known for some time: Starting at 10:17 UTC on 8 June 2022 ?

sorenhuus on Jun 17, 2022

have same issue, i’ve got the error:

` Error: read Application Insights Billing Features “XXXX” (Resource Group “XXXXX”): insights.ComponentCurrentBillingFeaturesClient#Get: Failure responding to request: StatusCode=200 – Original Error: Error occurred unmarshalling JSON - Error = ‘json: cannot unmarshal array into Go value of type insights.ApplicationInsightsComponentBillingFeatures’ JSON = ‘[]’

i notice that the resource (app insight) is created despite the error. When running second time “terraform apply” i have error that say resource is already existing. i tried with these asurerm version: 2.99.0 ,3.9.0, 3.10.0 `

MOURIK on Jun 16, 2022

Just to confirm that this is on Azure’s end, I’ve tried to create an Application Insights instance manually, and I got the below error.

Screenshot 2022-06-15 at 16 55 18

emassimi on Jun 15, 2022