google_logging_metric should be recreated when renaming label

[Evesy]

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.
• If an issue is assigned to the modular-magician user, it is either in the process of being autogenerated, or is planned to be autogenerated soon. If an issue is assigned to a user, that user is claiming responsibility for the issue. If an issue is assigned to hashibot, a community member has claimed the issue already.

Terraform Version

0.14.6

Affected Resource(s)

• google_logging_metric

Terraform Configuration Files

resource "google_logging_metric" "host_error" {
  name        = "compute/host_errors"
  description = "Counter for  VM instances that have hostError's"
  filter      = "resource.type=gce_instance AND protoPayload.methodName=compute.instances.hostError"
  metric_descriptor {
    metric_kind = "DELTA"
    value_type  = "INT64"
    labels {
      key         = "instance"
      value_type  = "STRING"
      description = "Hostname of instance"
    }
    labels {
      key         = "zone"
      value_type  = "STRING"
      description = "Availability zone of instance"
    }
    display_name = "VM Host Errors"
  }
  label_extractors = {
    "instance" = "REGEXP_EXTRACT(protoPayload.resourceName, \"projects/.+/zones/.+/instances/(.+)\")"
    "zone"     = "EXTRACT(resource.labels.zone)"
  }
}

Expected Behavior

The resource should be wholly recreated if one of the label names is changed, due to the limitations outlined in https://cloud.google.com/logging/docs/logs-based-metrics/labels
A label can't be deleted after it is created

Actual Behavior

When I changed one of the label names (instance -> instance_name) the Terraform diff was as below:

# google_logging_metric.host_error will be updated in-place
  ~ resource "google_logging_metric" "host_error" {
        id               = "compute/host_errors"
      ~ label_extractors = {
          - "instance"      = "REGEXP_EXTRACT(protoPayload.resourceName, \"projects/.+/zones/.+/instances/(.+)\")" -> null
          + "instance_name" = "REGEXP_EXTRACT(protoPayload.resourceName, \"projects/.+/zones/.+/instances/(.+)\")"
            # (1 unchanged element hidden)
        }
        name             = "compute/host_errors"
        # (3 unchanged attributes hidden)

      ~ metric_descriptor {
            # (4 unchanged attributes hidden)

          - labels {
              - description = "Hostname of instance" -> null
              - key         = "instance" -> null
              - value_type  = "STRING" -> null
            }
          + labels {
              + description = "Hostname of instance"
              + key         = "instance_name"
              + value_type  = "STRING"
            }
            # (1 unchanged block hidden)
        }
    }

This resulted in the below error when applying the plan:

google_logging_metric.host_error: Modifying... [id=compute/host_errors]
Error: Error updating Metric "compute/host_errors": googleapi: Error 400: Cannot delete existing label: instance

Steps to Reproduce

1. Create resources as listed above
2. Rename one of the labels
3. Plan & apply

[edwardmedia]

It looks like api does not allow updating the label.keys

[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.