Terraform is not removing tainted aws_resource and it's failing to remove aws_volume_attachment

[hgontijo]

Terraform Version

0.7.4

Affected Resource(s)

• aws_volume_attachment
• aws_instance

Terraform Configuration Files

resource "aws_instance" "teamcity_server" {
  ami = "${var.server_ami}"
  availability_zone = "${var.server_az}"
  instance_type = "${var.server_instance_type}"
  vpc_security_group_ids = [ "${split(",", var.security_groups)}" ]
  subnet_id = "${var.subnet_id}"

  key_name = "${var.key_name}"
  iam_instance_profile = "tc-server"

  root_block_device {
    volume_type = "gp2"
  }
}

resource "aws_ebs_volume" "data_volume" {
  availability_zone = "${var.server_az}"
  snapshot_id = "${var.data_volume_snapshot}"
  size = "${var.data_volume_size}"
  type = "gp2"
}

resource "aws_volume_attachment" "data_volume_attachment" {
  device_name = "/dev/sdf"
  instance_id = "${aws_instance.teamcity_server.id}"
  volume_id = "${aws_ebs_volume.data_volume.id}"
}

I ommited non-relevant parts.

Expected Behavior

Since the aws_instance has an aws_volume_attachment, we used to taint aws_instance before running terraform destroy to ensure the instance is marked for deletion so that aws_instance, aws_volume_attachment, aws_ebs_volume are properly destroyed regardless of the volume attachment dependency. This was working fine until v0.6.16.

Actual Behavior

$ terraform taint -no-color -allow-missing -module=teamcity aws_instance.teamcity_server
The resource aws_instance.teamcity_server in the module root.teamcity has been marked as tainted!
$ terraform destroy
...
Error applying plan:

1 error(s) occurred:

* aws_volume_attachment.data_volume_attachment: Error waiting for Volume (vol-af94311b) to detach from Instance: i-a6182dbe

Full logs

Based on the logs, aws_instance is not being removed as expected by running taint before destroy.

Steps to Reproduce

1. terraform apply
2. `terraform taint -no-color -allow-missing -module=teamcity aws_instance.teamcity_server``
3. terraform destroy

References

That maybe related to:

• EBS Volume Attachement Destroy/Recreate state issue in 0.7.4 #9000

[kshcherban]

Same issue with 0.7.10
Seems like before detaching instance should be stopped.

[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.