post upgrade to azurerm 3.26.0 (from 2.72.0), azurerm_monitor_diagnostic_setting getting forced replaced because of "resourceGroups" != "resourcegroups" in target_resource_id

[achupryn]

Is there an existing issue for this?

• [X] I have searched the existing issues

Community Note

• Please vote on this issue by adding a :thumbsup: 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

1.1.4

AzureRM Provider Version

3.26.0

Affected Resource(s)/Data Source(s)

azurerm_monitor_diagnostic_setting

Terraform Configuration Files

resource "azurerm_monitor_diagnostic_setting" "aks-logging" {
  count                      = var.enable_aks_logging ? 1 : 0
  name                       = "diagnostic_aksl"
  target_resource_id         = azurerm_kubernetes_cluster.k8s.id
  log_analytics_workspace_id = azurerm_log_analytics_workspace.k8s_logs.id
  ...
}

Debug Output/Panic Output

# plan
Terraform will perform the following actions:
  # azurerm_monitor_diagnostic_setting.aks-logging[0] must be replaced
-/+ resource "azurerm_monitor_diagnostic_setting" "aks-logging" {
      ~ id                         = "/subscriptions/<subscriptions_id>/resourcegroups/rgrp-aks-application-cluster/providers/Microsoft.ContainerService/managedClusters/aks-application-cluster|diagnostic_aksl" -> (known after apply)
        name                       = "diagnostic_aksl"
      ~ target_resource_id         = "/subscriptions/<subscriptions_id>/resourcegroups/rgrp-aks-application-cluster/providers/Microsoft.ContainerService/managedClusters/aks-application-cluster" -> "/subscriptions/<subscriptions_id>/resourceGroups/rgrp-aks-application-cluster/providers/Microsoft.ContainerService/managedClusters/aks-application-cluster" # forces replacement

# the difference is "resource**G**roups" != "resourcegroups" in target_resource_id

Expected Behaviour

azurerm_monitor_diagnostic_setting.aks-logging not changed

Actual Behaviour

azurerm_monitor_diagnostic_setting.aks-logging[0] must be replaced

The issue is ID casing - "resourceGroups" != "resourcegroups" in target_resource_id

Similar problem - #18467

Steps to Reproduce

No response

Important Factoids

No response

References

18467

[achupryn]

The same behavior in case update to v2.99

[teowa]

Hi @achupryn , thanks for submitting this, seems this issue is caused by PR https://github.com/hashicorp/terraform-provider-azurerm/pull/14241, which changes the resourcegroup case in the ID of azurerm_kubernetes_cluster resource (released in v2.88.0). I am working on a PR to fix this.

[Simondebb]

Hello @teowa,

I have a similaire issue with azurerm_api_management
azurerm version 2.99

azurerm_api_management.this will be updated in-place
  ~ resource "azurerm_api_management" "this" {
      ~ virtual_network_configuration {
          ~ subnet_id = "/subscriptions/***/resourcegroups/rg-myrg/providers/microsoft.network/virtualnetworks/vnet-myvnet/subnets/snet-mysnet" -> "/subscriptions/***/resourceGroups/rg-myrg/providers/Microsoft.Network/virtualNetworks/vnet-myvnet/subnets/snet-mysnet"
        }
        (8 unchanged blocks hidden)
}

resourcegroups != resourceGroups , microsoft.network != Microsoft.Network and virtualnetworks != virtualNetworks

Can you have look into this ? thanks a lot.

[teowa]

Hi @Simondebb , thanks for the reply, seems the issue is related to https://github.com/hashicorp/terraform-provider-azurerm/issues/18399. And the solution is in progress.

[achupryn]

v3.32.0 - issue is still present

[hypery2k]

@teowa any update on that one or workaround?

[teowa]

Hi @hypery2k, We are working on a more general solution, which can cope with more situations (instead of a specific resource), and it may take longer time.

And a workaround to avoid diff and force replacement is to manually change the resource ID reference using replace function, for example:

resource "azurerm_monitor_diagnostic_setting" "aks-logging" {
   target_resource_id         = replace(azurerm_kubernetes_cluster.k8s.id, "resourceGroups", "resourcegroups")
}

Hope this may help.

[hypery2k]

thx for the hint, that worked

[aadesola]

Hi @teowa , I have a similar issue and tried to use the replace function but didn't work. Although it's not an ID. resource_group_name = replace(var.someresourcegroup, "resourcegroups", "resourceGroups") any pointers on what I'm missing?

[teowa]

@aadesola , it is case by case, you should use the replace function to replace the ID segment where you want to avoid unintentional change. Could you please provide more details on your Terraform plan result? Maybe you should replace some property with ID format but not the resource_group_name.