azurerm_kusto_cluster_managed_private_endpoint crates an error when kusto cluster is in process of maintenance

[christiansalathe-art]

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

1.2.3

AzureRM Provider Version

3.58.0

Affected Resource(s)/Data Source(s)

azurerm_kusto_cluster_managed_private_endpoint

Terraform Configuration Files

#this code is just for showing and not able to run
resource "azurerm_kusto_cluster" "dataexplorer" {
  for_each                      = var.dec_multi
  provider                      = azurerm.multi
  name                          = each.value.adx_name
  resource_group_name           = each.value.adx_rg
  location                      = var.location
  engine                        = "V3"
  zones                         = [1,2]
  public_network_access_enabled = each.value.adx_public
  streaming_ingestion_enabled   = each.value.streaming_ingestion_enabled
  purge_enabled                 = each.value.adx_purge_enabled
  auto_stop_enabled             = each.value.adx_auto_stop_enabled
#  language_extensions           = [each.value.language_extensions] # doesn't work with cmk encryption
  disk_encryption_enabled       = true
  tags                          = var.tags_dec
  allowed_ip_ranges             = [each.value.iprange]

  sku {
    name     = each.value.adx_sku
    capacity = each.value.adx_capacity
  }

  identity {
    type = "SystemAssigned"
  }

}

resource "azurerm_kusto_cluster_managed_private_endpoint" "dec-managed-pe-blob" {
  for_each                     = { for k in compact([for k, v in var.dec_multi: v.managed_private_endpoint_blob ? k : ""]): k => var.dec_multi[k] } # k=key v=value
  provider                     = azurerm.multi
  name                         = "managed-pe-${azurerm_kusto_cluster.dataexplorer[each.key].name}-blob"
  resource_group_name          = each.value.adx_rg
  cluster_name                 = azurerm_kusto_cluster.dataexplorer[each.key].name
  private_link_resource_id     = each.value.managed_private_endpoint_resid
  group_id                     = "blob"
  request_message              = "Please approve"

    depends_on = [
      azurerm_private_endpoint.dec-pe,
      null_resource.rotation_policy
    ]
}


# time sleep because if issue with kusto cluster mainenance task
resource "time_sleep" "wait_seconds" {
  depends_on = [azurerm_kusto_cluster_managed_private_endpoint.dec-managed-pe-blob]
  create_duration = "600s"
}

resource "azurerm_kusto_cluster_managed_private_endpoint" "dec-managed-pe-dfs" {
  for_each                     = { for k in compact([for k, v in var.dec_multi: v.managed_private_endpoint_dfs ? k : ""]): k => var.dec_multi[k] } # k=key v=value
  provider                     = azurerm.multi
  name                         = "managed-pe-${azurerm_kusto_cluster.dataexplorer[each.key].name}-dfs"
  resource_group_name          = each.value.adx_rg
  cluster_name                 = azurerm_kusto_cluster.dataexplorer[each.key].name
  private_link_resource_id     = each.value.managed_private_endpoint_resid
  group_id                     = "dfs"
  request_message              = "Please approve"

    depends_on = [
      time_sleep.wait_seconds
    ]
}

Debug Output/Panic Output

module.modul-dataexplorer.azurerm_kusto_cluster_managed_private_endpoint.dec-managed-pe-dfs["adx2"]: Still creating... [2m0s elapsed]
module.modul-dataexplorer.azurerm_kusto_cluster_managed_private_endpoint.dec-managed-pe-dfs["adx2"]: Creation complete after 2m2s [id=/subscriptions/xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx/resourceGroups/rg-bat-1/providers/Microsoft.Kusto/clusters/iiiiiiiiii/managedPrivateEndpoints/managed-pe-iiiiiiiiii-dfs]
?
¦ Error: creating/updating Managed Private Endpoint (Subscription: "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
¦ Resource Group Name: "rg-bat-1"
¦ Cluster Name: "iiiiiiiiii"
¦ Managed Private Endpoint Name: "managed-pe-adxbat9t9hzj3c-blob"): polling after CreateOrUpdate: Code="ServiceIsInMaintenance" Message="[Conflict] Cluster 'adxbat9t9hzj3c' is in process of maintenance for a short period. You may retry to invoke the operation in a few minutes."
¦
¦   with module.modul-dataexplorer.azurerm_kusto_cluster_managed_private_endpoint.dec-managed-pe-blob["adx2"],
¦   on .terraform/modules/modul-dataexplorer/resources.tf line 88, in resource "azurerm_kusto_cluster_managed_private_endpoint" "dec-managed-pe-blob":
¦   88: resource "azurerm_kusto_cluster_managed_private_endpoint" "dec-managed-pe-blob" {
¦

Expected Behaviour

the terraform provider azurerm_kusto_cluster_managed_private_endpoint should not throw an error when the kust cluster s in process of maintenance

Actual Behaviour

the provider azurerm_kusto_cluster_managed_private_endpoint throw an error when the Cluster is in process of maintenance

the working workaround ist the time_sleep between creating the first and the second azurerm_kusto_cluster_managed_private_endpoint

Steps to Reproduce

create more than one azurerm_kusto_cluster_managed_private_endpoint for a azure data explorer kusto cluster

Important Factoids

no

References

no

[liuwuliuyun]

Hi @rcskosir thanks for raising this issue. I am able to reproduce your problem locally. I will try to work on a solution. Any suggestion from anyone on this issue is welcomed. Since this is related to Kusto API, currently the best workaround I could think of is the same proposed time_sleep mentioned in the description.

[Ben-Duf]

Hi, I have two azurerm_kusto_cluster_managed_private_endpoint also. I must wait the first one to finish before creating the second.

resource "azurerm_kusto_cluster_managed_private_endpoint" "dec-managed-pe-blob" {
...
}
resource "azurerm_kusto_cluster_managed_private_endpoint" "dec-managed-pe-dfs" {
   depends_on = [ 
      azurerm_kusto_cluster_managed_private_endpoint.dec-managed-pe-blob
   ]
}