Flaky creation of management groups: "The resource type 'operationResults/asyncOperation' could not be found in the namespace 'Microsoft.Management' for api version '2018-03-01-preview'

[Andrei-Predoiu]

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 (and AzureRM Provider) Version

terraform 1.1.0
azurerm 2.88.1
Bug is also observed on 2.87, untested on 2.80

Affected Resource(s)

azurerm_management_group

Terraform Configuration Files

resource "azurerm_management_group" "development_peered" {
  count                      = var.env == "Development" && var.peered == 1 ? 1 : 0
  display_name               = "${var.product}-${var.env}-${var.location}"
  parent_management_group_id = var.parent_management_group_id[2]
  subscription_ids           = [var.subscription_ids]
}

Debug Output

Panic Output

Expected Behaviour

The group should be created and the build should succeed.

Actual Behaviour

about 1 out of 5 times the plan application fails with the following error:

╷
│ Error: failed when waiting for creation of Management Group "4252ac91-509e-4f7d-9a90-507c11e7006a": Future#WaitForCompletion: the number of retries has been exceeded: StatusCode=404 -- Original Error: Code="InvalidResourceType" Message="The resource type 'operationResults/asyncOperation' could not be found in the namespace 'Microsoft.Management' for api version '2018-03-01-preview'. The supported api-versions are '2021-04-01,2020-10-01,2020-05-01,2020-02-01,2019-11-01,2018-03-01-beta'."
│ 
│   with module.management_group_for_spoke_and_rbac.azurerm_management_group.development_peered[0],
│   on ../modules/spoke_management_group/main.tf line 60, in resource "azurerm_management_group" "development_peered":
│   60: resource "azurerm_management_group" "development_peered" {
│ 
╵

Steps to Reproduce

1. terraform apply

Important Factoids

We are running in westeurope
skip_provider_registration is not enabled

References

[Andrei-Predoiu]

According to the error, the supported API versions are:
'2021-04-01,2020-10-01,2020-05-01,2020-02-01,2019-11-01,2018-03-01-beta'

[tombuildsstuff]

hi @Andrei-Predoiu

Thanks for opening this issue.

Digging into this, this appears to be an issue in the upstream API which should be resolved by #14635.

Thanks!

[github-actions]

This functionality has been released in v2.90.0 of the Terraform Provider. Please see the Terraform documentation on provider versioning or reach out if you need any assistance upgrading.

For further feature requests or bug reports with this functionality, please create a new GitHub issue following the template. Thank you!

[mdepedrof]

after upgrading to 2.90.0 it no longer crashes but the process of creating ManagementsGroups never ends. I am testing and it takes 22 minutes in "StillCreating" state although in azure it is already created.

[mdepedrof]

@Andrei-Predoiu it working for you?? do you have this behavior?

[BowBart]

@crower19 I had the behavior. Did an upgrade of the provider to 2.90.0. It works!

[zahaim]

Hello there, I encountered the same problem. After the upgrade to 2.90.0 it no longer crashes but the process of creating ManagementsGroups never ends.

module.management_groups["f92cfe88-1fd0-4353-afc2-21c423f63911"].azurerm_management_group.mg["corp"]: Still creating... [29m30s elapsed]
module.management_groups["f92cfe88-1fd0-4353-afc2-21c423f63911"].azurerm_management_group.mg["corp"]: Still creating... [29m40s elapsed]
module.management_groups["f92cfe88-1fd0-4353-afc2-21c423f63911"].azurerm_management_group.mg["preflight_corp"]: Still creating... [29m40s elapsed]
module.management_groups["f92cfe88-1fd0-4353-afc2-21c423f63911"].azurerm_management_group.mg["preflight_corp"]: Still creating... [29m50s elapsed]
module.management_groups["f92cfe88-1fd0-4353-afc2-21c423f63911"].azurerm_management_group.mg["corp"]: Still creating... [29m50s elapsed]
╷
│ Error: failed when waiting for creation of Management Group "preflight_corp": Future#WaitForCompletion: context has been cancelled: StatusCode=200 -- Original Error: context deadline exceeded
│ 
│   with module.management_groups["f92cfe88-1fd0-4353-afc2-21c423f63911"].azurerm_management_group.mg["preflight_corp"],
│   on modules/management_groups/main.tf line 3, in resource "azurerm_management_group" "mg":
│    3: resource "azurerm_management_group" "mg" {

[BowBart]

@zahaim have your tried applying giving it a different management group id? When I changed the id of the management group it ran trough.

[zahaim]

@BowBart I use name. Do you suggest using display_name and let Azure generate the id automatically? If possible I would prefer to avoid that, as simple id that equals the name is helpful to apply parent_management_group_id in next steps. But maybe this is an anti-pattern, not sure to be honest.

This is how I create the resource:

resource "azurerm_management_group" "mg" {
    for_each = var.name_subscription
    name               = each.key
    parent_management_group_id = "/providers/Microsoft.Management/managementGroups/${var.parent_management_group_id}"

    subscription_ids = each.value
}

[BowBart]

@zahaim I mean the name in this case, which is used as a unique in your tenant I believe. I also first had the long creation time, perhaps because continuous removing and creating. When I changed the group to a different name it worked!

[zahaim]

@BowBart ah, ok. Actually, something must have changed on Azure side as today I'm able to create / delete / create the same management groups without problems. I guess it depends on the load on Azure or something. Anyways, thanks for the tip!

[github-actions]

I'm going to lock this issue because it has been closed for 30 days ⏳. This helps our maintainers find and focus on the active issues.
If you have found a problem that seems similar to this, please open a new issue and complete the issue template so we can capture all the details necessary to investigate further.