Terraform bug impacting deployments

[dav3r]

🐛 Bug Report

There is currently a bug in Terraform (as of 0.12.7) that is preventing route53 (DNS) records from being deleted when a "fresh" DNS record needs to be created. This results in multiple route53 records (with different IP addresses) for the same domain.

The current workaround is to manually delete the outdated route53 records.

To Reproduce

Steps to reproduce the behavior:

• Any terraform apply that causes the replacement of a route53 record set

Expected behavior

The old route53 record should be deleted by terraform before it creates the new record.

Any helpful log output

❱ terraform apply
...
  # aws_route53_record.cyhy_rev_dashboard_PTR must be replaced
-/+ resource "aws_route53_record" "cyhy_rev_dashboard_PTR" {
...
}
...
aws_route53_record.cyhy_rev_dashboard_PTR: Destroying... [id=Z2AINF9RNTYH7J_101.10.10.10.in-addr.arpa._PTR]
...
aws_route53_record.cyhy_rev_dashboard_PTR: Destruction complete after 0s
# LIES!!!  The record above still exists! :(
...
aws_route53_record.cyhy_rev_dashboard_PTR: Creating...
aws_route53_record.cyhy_rev_dashboard_PTR: Creation complete after 46s [id=Z2AINF9RNTYH7J_38.10.10.10.in-addr.arpa._PTR]
# TRUTH!