Re-apply of timed out ASG creation does not enable requested CloudWatch metrics

[mattzimmerman]

Terraform Version

Terraform 0.8.4
edit also confirmed that we still see this issue on 0.8.7

Affected Resource(s)

Please list the resources as a list, for example:

• aws_autoscaling_group

Debug Output

Run #1 (initial plan): https://gist.github.com/mattzimmerman/b301b0a230e51544b53b535392d577c3
Run #1 (apply): https://gist.github.com/mattzimmerman/08eefb57d56f49e3985ce8579ed18dd3
Run #2 (retry after a failure): https://gist.github.com/mattzimmerman/9d6cb45daede7378cfda77079c8094a5

Expected Behavior

Re-apply of failed ASG creations should detect that requested CloudWatch metrics were missing and enable them.

Actual Behavior

CloudWatch metrics are not enabled before the ASG has healthy instances. After the timeout failure, no diff was detected on a re-apply (so CloudWatch metrics are never enabled on the ASG resource during subsequent applies if it fails the first time).

Steps to Reproduce

1. Apply an aws_autoscaling_group with an ELB healthcheck and CloudWatch GroupInServiceInstances enabled via a module. None of the instances ever become healthy, so creation of the ASG should fail after the timeout.
2. Re-apply that same aws_autoscaling_group.
3. Notice that no diff is detected, but CloudWatch metrics are not enabled on that ASG.

[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 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.