Adding an additional geo_location to an azurerm_cosmosdb_account should not require replacement

[lukehoban]

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 -v
Terraform v0.12.0
+ provider.azurerm v1.28.0
+ provider.random v2.1.2

Affected Resource(s)

• azurerm_cosmosdb_account

Terraform Configuration Files

resource "azurerm_resource_group" "rg" {
  name     = "rg123"
  location = "WestUS"
}

resource "random_integer" "ri" {
  min = 10000
  max = 99999
}

resource "azurerm_cosmosdb_account" "db" {
  name                = "tfex-cosmos-db-${random_integer.ri.result}"
  kind                = "MongoDB"
  resource_group_name = "${azurerm_resource_group.rg.name}"
  location            = "${azurerm_resource_group.rg.location}"
  consistency_policy {
    consistency_level       = "BoundedStaleness"
    max_interval_in_seconds = 10
    max_staleness_prefix    = 200
  }
  offer_type          = "Standard"
  enable_automatic_failover = true
  geo_location {
    location          = "WestUS"
    failover_priority = 0
  }
#  geo_location {
#    location          = "EastUS"
#    failover_priority = 1
#  }
}

Expected Behavior

After deploying the code above, uncommenting the second geo_location and rerunning apply should have updated the existing CosmosDB database.

Actual Behavior

Instead, it requires a replacement.

  # azurerm_cosmosdb_account.db must be replaced
-/+ resource "azurerm_cosmosdb_account" "db" {
      ~ connection_strings                = (sensitive value)
        enable_automatic_failover         = true
        enable_multiple_write_locations   = false
      ~ endpoint                          = "https://tfex-cosmos-db-95504.documents.azure.com:443/" -> (known after apply)
      ~ id                                = "/subscriptions/8bafcca2-660a-4459-a503-b785cf317a3a/resourceGroups/rg123/providers/Microsoft.DocumentDB/databaseAccounts/tfex-cosmos-db-95504" -> (known after apply)
        is_virtual_network_filter_enabled = false
        kind                              = "MongoDB"
        location                          = "westus"
        name                              = "tfex-cosmos-db-95504"
        offer_type                        = "Standard"
      ~ primary_master_key                = (sensitive value)
      ~ primary_readonly_master_key       = (sensitive value)
      ~ read_endpoints                    = [
          - "https://tfex-cosmos-db-95504-westus.documents.azure.com:443/",
        ] -> (known after apply)
        resource_group_name               = "rg123"
      ~ secondary_master_key              = (sensitive value)
      ~ secondary_readonly_master_key     = (sensitive value)
      ~ tags                              = {} -> (known after apply)
      ~ write_endpoints                   = [
          - "https://tfex-cosmos-db-95504-westus.documents.azure.com:443/",
        ] -> (known after apply)

        consistency_policy {
            consistency_level       = "BoundedStaleness"
            max_interval_in_seconds = 10
            max_staleness_prefix    = 200
        }

      - geo_location { # forces replacement
          - failover_priority = 0 -> null
          - id                = "tfex-cosmos-db-95504-westus" -> null
          - location          = "westus" -> null
        }
      + geo_location { # forces replacement
          + failover_priority = 0
          + id                = (known after apply)
          + location          = "westus"
        }
      + geo_location { # forces replacement
          + failover_priority = 1
          + id                = (known after apply)
          + location          = "eastus"
        }
    }

Steps to Reproduce

1. terraform apply
2. Uncomment the second geo_location
3. terraform apply again

Notes

• Additional geo locations can be added to a CosmosDB account without downtime or replacement using the API or console.

[D4zedC0der]

So I have a similar scenario where I haven't added a new geo_location. We already had 2. Our plan output looks like more this:

      - geo_location { # forces replacement
          - failover_priority = 0 -> null
          - id                = "aca-pre-neu-csvi-geoloc2-1323" -> null
          - location          = "westeurope" -> null
          - prefix             = "aca-pre-neu-csvi-geoloc2-1323" -> null
        }
      + geo_location { # forces replacement
          + failover_priority = 0
          + id                = (known after apply)
          + location          = "westeurope"
          + prefix             = "aca-pre-neu-csvi-geoloc2-1323"
        }
      - geo_location { # forces replacement
          - failover_priority = 1 -> null
          - id                = "aca-pre-neu-csvi-1323-northeurope" -> null
          - location          = "northeurope" -> null
        }
      + geo_location { # forces replacement
          + failover_priority = 1
          + id                = (known after apply)
          + location          = "northeurope"
          + prefix             = "aca-pre-neu-csvi-1323-northeurope"
        }

Now the first geoLoc stanza to us looks identical so I don't understand why it needs to -/+ at all

The second one didn't have a "prefix" in it previously, but I believe in adding a prefix which matches - it should be the same right? Yet it is still trying to -/+. We have modules and multiple environments, so we are struggling to have a prefix for the first one but not for the second without at least destroying the CosmosDB in at least ONE of our environments.

Ultimately my question is - Would any of the referenced PR's resolve this?

If not - and if this is indeed caused by the prefix addition to the second geo_location stanza - is there any way to set the prefix on the second geo_location stanza without causing a -/+ ?

[eliasvakkuri]

I have a similar issue. In my case I tried using the dynamic functionality to allow a variable number of geo_location blocks e.g. per environment, but this appears to require a full replace of the CosmosDB account on each apply. So not usable currently.

[georgegil]

Having the same issue here.

Changing geo settings replaces the entire db

[matthawley]

Has anyone looked into this? This becomes a highly destructive action.

[jonhoare]

I am also having this issue today. I currently only have a single geo_location with priority 0. I am now trying to add a second region for geo_location, without updating the primary location which is already defined and it is wanting to completely destroy my primary location to recreate it, whilst adding the secondary location.

[ghost]

This has been released in version 2.14.0 of the provider. Please see the Terraform documentation on provider versioning or reach out if you need any assistance upgrading. As an example:

provider "azurerm" {
    version = "~> 2.14.0"
}
# ... other configuration ...

[ghost]

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 feel this issue should be reopened, we encourage creating a new issue linking back to this one for added context. If you feel I made an error 🤖 🙉 , please reach out to my human friends 👉 [email protected]. Thanks!