Terraform-provider-aws: aws_cloudwatch_metric_alarm AlarmActions ARN incorrectly rejected in 1.41.0

[jeohist]

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

0.11.8

Affected Resource(s)

• aws_cloudwatch_metric_alarm

Terraform Configuration Files

data "aws_region" "current_region" {}

resource "aws_cloudwatch_metric_alarm" "foo" {
  count               = 1
  alarm_name          = "foo"
  alarm_description   = "Status checks have failed for system, recovering instance"
  namespace           = "AWS/EC2"
  statistic           = "Minimum"
  comparison_operator = "GreaterThanThreshold"
  threshold           = "0"
  unit                = "Count"
  evaluation_periods  = "2"
  period              = "60"
  metric_name         = "StatusCheckFailed_System"
  alarm_actions       = ["arn:aws:automate:${data.aws_region.current_region.name}:ec2:recover"]

  dimensions {
    InstanceId = "${var.instanceid}"
  }
}

Expected Behavior

aws_cloudwatch_metric_alarm.foo should have been created

Actual Behavior

• resource.aws_cloudwatch_metric_alarm.foo: "alarm_actions.0" doesn't look like a valid ARN ("^arn:[\w-]+:([a-zA-Z0-9\-])+:([a-z]{2}-(gov-)?[a-z]+-\d{1})?:(\d{12})?:(.*)$"): "arn:aws:automate:eu-west-1:ec2:recover"

Steps to Reproduce

1. terraform plan

Important Factoids

This broke since #6151 was merged

References

• AWS Documentation: https://docs.aws.amazon.com/AmazonCloudWatch/latest/APIReference/API_PutMetricAlarm.html
• #6151

[bflad]

Sorry for the trouble! Bug fix pull request submitted with acceptance testing to cover these specially formatted ARNs: #6206

[jeohist]

@bflad No problem! Appreciate the quick fix :-)

[Seraf]

Hello,

Thanks for the quick fix ! Is a hotfix release planned in order to quickly fix this bug upstream ?

[bflad]

This will likely be released on the normal weekly cadence. As a workaround, you can pin your configuration to the previous provider release, e.g.

provider "aws" {
  # ... potentially other configuration ...
  version = "1.40.0"
}

[prasanthsurapaneni]

Thank you.

[bflad]

Thanks for your patience -- the fix for this has been merged and will release with version 1.42.0 of the AWS provider, likely today or tomorrow.

[swoodford]

I am also having fun troubleshooting this bug today, trying to run a terraform plan against an existing infrastructure with a CloudWatch StatusCheckFailed alarm that was previously created and working fine in a terraform apply:

Error: aws_cloudwatch_metric_alarm.status-check: "alarm_actions.0" doesn't look like a valid ARN ("^arn:[\\w-]+:([a-zA-Z0-9\\-])+:([a-z]{2}-(gov-)?[a-z]+-\\d{1})?:(\\d{12})?:(.*)$"): "arn:aws:automate:us-east-1:ec2:recover"

[bflad]

This has been released in version 1.42.0 of the AWS provider. Please see the Terraform documentation on provider versioning or reach out if you need any assistance upgrading.

[hashibot[bot]]

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!