Azure Container App error when using custom registry: operationexpired

Terraform Providers Azure
1

I am using this container app resource to create an Azure Container App using an image from an Azure Container Registry that I deployed.
The resource is declared as follows:

variable "bet-api-target-port" {
  description = "Port for the bet-api service"
  type        = number
}

resource "azurerm_container_app" "bet_api" {
  name                         = "bet-api"
  container_app_environment_id = data.azurerm_container_app_environment.app_env.id
  resource_group_name          = data.azurerm_resource_group.main_group.name
  revision_mode                = "Single"

  secret {
    name  = "registry-password"
    value = data.azurerm_container_registry.main_registry.admin_password
  }

  secret {
    name                = lower(data.azurerm_key_vault_secret.odds_api_key.name)
    key_vault_secret_id = data.azurerm_key_vault_secret.odds_api_key.id
    identity            = "System"
  }

  secret {
    name                = lower(data.azurerm_key_vault_secret.cosmos_db_connection_string.name)
    key_vault_secret_id = data.azurerm_key_vault_secret.cosmos_db_connection_string.id
    identity            = "System"
  }

  secret {
    name                = lower(data.azurerm_key_vault_secret.cosmos_db_database.name)
    key_vault_secret_id = data.azurerm_key_vault_secret.cosmos_db_database.id
    identity            = "System"
  }

  registry {
    server               = data.azurerm_container_registry.main_registry.login_server
    username             = data.azurerm_container_registry.main_registry.admin_username
    password_secret_name = "registry-password"
  }

  template {
    min_replicas = 1
    max_replicas = 5

    # Total CPU and memory for all containers defined in a Container App must add up to one of the following CPU - Memory combinations: [cpu: 0.25, memory: 0.5Gi]; [cpu: 0.5, memory: 1.0Gi]; [cpu: 0.75, memory: 1.5Gi]; [cpu: 1.0, memory: 2.0Gi]; [cpu: 1.25, memory: 2.5Gi]; [cpu: 1.5, memory: 3.0Gi]; [cpu: 1.75, memory: 3.5Gi]; [cpu: 2.0, memory: 4.0Gi]
    container {

      image  = "${data.azurerm_container_registry.main_registry.login_server}/fantanba/bet_api:latest"
      memory = "2Gi"
      cpu    = 1
      name   = "bet-api-image"

      env {
        name        = "ODDS_API_KEY"
        secret_name = lower(data.azurerm_key_vault_secret.odds_api_key.name)
      }

      env {
        name        = "COSMOSDB_CONNECTION_STRING"
        secret_name = lower(data.azurerm_key_vault_secret.cosmos_db_connection_string.name)
      }

      env {
        name        = "COSMOSDB_DATABASE"
        secret_name = lower(data.azurerm_key_vault_secret.cosmos_db_database.name)
      }
    }
  }

  ingress {
    external_enabled = true
    target_port      = var.bet-api-target-port
    transport        = "http"
    traffic_weight {
      percentage      = 100
    }
  }
}

Some secrets in the resource declaration use information from a separate key vault data source that is declared in another tf script but can be ignored for the purpose of this issue.

The issue that I am having is related to the usage of a personal azure container registry specified using the registry block, I am using password authentication due to some restrictions imposed on my azure account.

When I try to deploy this resource the run goes one for 10 minutes exactly then the following error is shown:

{"id":"/subscriptions/xxxxxxxxx/providers/Microsoft.App/locations/westeurope/containerappOperationStatuses/xxxxxxxx","name":"xxxxxxx","status":"Failed","error":{"code":"ContainerAppOperationError","message":"Failed to provision revision for container app 'bet-api'. Error details: Operation expired."},"startTime":"2024-07-12T18:28:08.0466874"}

When I try to deploy an example container app with hello world image it does not fail like this.

Sadly I do not have any additional informations about the issue, I can only confirm that:

  • all the data used by the secrets is being retrieved correctly (such as the registry password) because I checked the values in seperate runs.
  • the Container App resource is actually created as I can see from Azure Portal but it does not have any container
  • creating the same container app (with the same settings and the same image and registry) is successful when using Azure Portal

Does anyone know what could be the issue?

2

I found out the cause of my issue and it involves identities and key vaults.

First error: I did not declare an identity for this container, so I added

identity {
  type = "SystemAssigned"
}

and tested again and the issue changed, now the terraform deployment terminated correctly but the application was not working.

At this step I understood that the system assigned identity that I created did not have access to the azure key vault.
Unfortunately my azure account is heavily limited on identity managed (it is a university managed account) so I could not add correct permissions to the system identity for the azure key vault.

In the end I solved my issue by creating the secrets using the value of the secret pulled from the key vault directly at creation.
Example:

secret {
  name  = lower(data.azurerm_key_vault_secret.odds_api_key.name)
  value = data.azurerm_key_vault_secret.odds_api_key.value
}