Update on dependencies causes aws_autoscaling_notification to error on update

[kdanga]

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 other comments that do not add relevant new information or questions, 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

0.12.18

• provider.aws v2.41.0

Affected Resource(s)

• aws_autoscaling_notification

Terraform Configuration Files

# Copy-paste your Terraform configurations here - for large Terraform configs,
# please use a service like Dropbox and share a link to the ZIP file. For
# security, you can also encrypt the files using our GPG public key: https://keybase.io/hashicorp

resource "aws_autoscaling_group" "test-asg" {

  launch_configuration      = "${aws_launch_configuration.test-lc.id}"
  max_size                  = 1
  min_size                  = 0
  desired_capacity          = 0

  availability_zones = ["${var.availability_zone}"]

  # interpolate the LC into the ASG name so it always forces an update
  name                = "${aws_launch_configuration.test-lc.name}-asg"
  vpc_zone_identifier = ["${var.subnet}"]
}

resource "aws_launch_configuration" "test-lc" {
  name_prefix = "terraform-test-lc-${var.build}"

  root_block_device {
    volume_type           = "gp2"
    delete_on_termination = true
  }

  image_id             = "${var.image_id}"
  instance_type        = "${var.instance_id}"

  ebs_optimized = true
}

resource "aws_autoscaling_notification" "test-asg-notification" {
  group_names   = ["${aws_autoscaling_group.test-asg.name}"]
  notifications = ["autoscaling:EC2_INSTANCE_TERMINATE"]
  topic_arn     = "${aws_sns_topic.topic.arn}"
}

variable "build" {}

Debug Output

Expected Behavior

When forcing an update on the aws_launch_configuration resource, terraform should update the dependencies without any errors.

Actual Behavior

The aws_autoscaling_notification tries to do an update after both dependencies on aws_autoscaling_group and aws_launch_configuration were replaced. Fails with the following error:

aws_autoscaling_group.test-asg: Destroying... [id=terraform-test-lc-320191213192730297300000001-asg]
aws_autoscaling_group.test-asg: Still destroying... [id=terraform-test-lc-320191213192730297300000001-asg, 10s elapsed]
aws_autoscaling_group.test-asg: Still destroying... [id=terraform-test-lc-320191213192730297300000001-asg, 20s elapsed]
aws_autoscaling_group.test-asg: Destruction complete after 22s
aws_launch_configuration.test-lc: Destroying... [id=terraform-test-lc-320191213192730297300000001]
aws_launch_configuration.test-lc: Destruction complete after 1s
aws_launch_configuration.test-lc: Creating...
aws_launch_configuration.test-lc: Creation complete after 5s [id=terraform-test-lc-220191213192847027200000001]
aws_autoscaling_group.test-asg: Creating...
aws_autoscaling_group.test-asg: Creation complete after 3s [id=terraform-test-lc-220191213192847027200000001-asg]
aws_autoscaling_notification.test-asg-notification: Modifying... [id=arn:aws:sns:us-east-1:494976196876:vrs-alertTermination]

Error: Error deleting notification configuration for ASG "terraform-test-lc-320191213192730297300000001-asg", Topic ARN "arn:aws:sns:us-east-1:494976196876:vrs-alertTermination"

Steps to Reproduce

1. terraform apply and provide build valueof 1
2. terraform apply again, but with build value of 2

Important Factoids

References

• #0000

[lystopad]

Hi,
Same for me with terraform v0.14.11 and aws provider v3.42.0.

[mehmetyazicioglu]

hello,
i am having the same issue with terraform v.0.14.11 and aws provider v3.42.0.

[lystopad]

@kdanga , hi
I solved it with the following explicit configuration for ASG and LC resources:

    lifecycle {
        create_before_destroy = true
    }

After that terraform is not complaining. Pls, try it.

[justinretzolk]

Hey y'all 👋 Thank you for taking the time to file this, and for the additional discussion around it! Given the workaround provided above, and that there's been a number of AWS provider releases since this was initially field, can anyone confirm whether they're still experiencing this behavior?