Updates to metric_query on aws_cloudwatch_metric_alarm fail if dimensions is set

[strongoose]

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

+ provider.aws v1.59.0```

### Affected Resource(s)

<!--- Please list the affected resources and data sources. --->
* aws_cloudwatch_metric_alarm

### Terraform Configuration Files

<!--- Information about code formatting: https://help.github.com/articles/basic-writing-and-formatting-syntax/#quoting-code --->

```hcl
resource aws_cloudwatch_metric_alarm test {
  alarm_name                = "TEST"
  alarm_description         = "Debugging metric update in terraform"
  comparison_operator       = "GreaterThanOrEqualToThreshold"
  evaluation_periods        = "1"
  threshold                 = 10
  insufficient_data_actions = []
  ok_actions                = []
  alarm_actions             = []

  metric_query = {
    id          = "e2"
    expression  = "m1*100/m2"
    label       = "5XX percentage"
    return_data = "true"
  }

  metric_query {
    id = "m1"

    metric {
      metric_name = "5XXError"
      namespace   = "AWS/ApiGateway"
      period      = "90"
      stat        = "Sum"
      unit        = "Count"

      dimensions {
        ApiName = "Test"
        Stage   = "Test"
      }
    }
  }

  metric_query {
    id = "m2"

    metric {
      metric_name = "Count"
      namespace   = "AWS/ApiGateway"
      period      = "90"
      stat        = "Sum"
      unit        = "Count"

      dimensions {
        ApiName = "Test"
        Stage   = "Test"
      }
    }
  }
}

To reproduce the error, make any change to the dimension metric queries, for example change metric_name on m1 to 4XXError.

Expected Behavior

The metric query is updates as expected to alert on percentage of 4XXs

Actual Behavior

Apply fails with:

1 error(s) occurred:

* aws_cloudwatch_metric_alarm.test: 1 error(s) occurred:

* aws_cloudwatch_metric_alarm.test: Updating metric alarm failed: InvalidParameter: 1 validation error(s) found.
- minimum field size of 1, PutMetricAlarmInput.Metrics[3].Id.


Terraform does not automatically rollback in the face of errors.
Instead, your Terraform state file has been partially updated with
any resources that successfully completed. Please address the error
above and apply again to incrementally change your infrastructure.

Steps to Reproduce

1. terraform apply the HCL above
2. change one of the metrics, e.g. change metric_name on line 22 to 4XXError
3. run terraform apply again

Important Factoids

This only seems to affect metrics with dimensions.

[strongoose]

Having looked at the state, it seems that the update operation creates a 4th empty metric with no ID:

"metric_query.3256485992.expression": "",
"metric_query.3256485992.id": "",
"metric_query.3256485992.label": "",
"metric_query.3256485992.metric.#": "0",
"metric_query.3256485992.return_data": "false",

this leaves terraform in an unrecoverable state until the resource is destroyed

[sjauld]

I can confirm that the address of PutMetricAlarmInput.Metrics[x].Id. matches the metric query that is being blanked by terraform

[luthor2016ad]

I confirm the issue
the only way to update the alarm is to destoy it before...

[bflad]

The fix for this has been merged and will release with version 2.4.0 of the Terraform AWS Provider, likely later today.

[bflad]

This has been released in version 2.4.0 of the Terraform AWS provider. Please see the Terraform documentation on provider versioning or reach out if you need any assistance upgrading.

[IslamAzab]

I can replicate the error by:

• creating a relative metric
• then trying to update it

Updating metric alarm failed: InvalidParameter: 1 validation error(s) found.
- minimum field size of 1, PutMetricAlarmInput.Metrics[2].Id.

$ terraform -v
Terraform v0.11.13
+ provider.aws v1.60.0

[sjauld]

This was fixed in 2.4 as per the above. Please try upgrading!

…

On Fri, 14 Jun 2019 at 07:59, Islam Azab ***@***.***> wrote: I can replicate the error by: - creating a relative metric - then trying to update it Updating metric alarm failed: InvalidParameter: 1 validation error(s) found. - minimum field size of 1, PutMetricAlarmInput.Metrics[2].Id. $ terraform -v Terraform v0.11.13 + provider.aws v1.60.0 — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#7593>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AB6Z4NYSRQCVZJXBNN6Z6PDP2K7NRANCNFSM4GYFSCPA> .

[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. Thanks!