Race condition w/ subnet destroy in AWS.

[russroy]

I use terraform to create a single instance and subnet resource within AWS. Attempting to destroy though one encounters a failure due to race condition. Destruction needs to happen in a synchronous manner. As it is, terraform correctly destroys the instance before the subnet upon which it is dependent but, in AWS, instance destruction is not instantaneous and subnet deletion fails:

rroy:work-terraform rroy$ terraform show terraform.tfstate
aws_instance.rroy-intc-061600-enchantd-0-0:
  id = i-4c5033a2
  ami = ami-06d3766e
  availability_zone = us-east-1b
  instance_type = t2.micro
  key_name = rroy-aws
  private_dns = ip-100-xxx-xxx-90.ec2.internal
  private_ip = 100.xxx.xxx.90
  public_dns = 
  public_ip = 
  security_groups.# = 1
  security_groups.0 = default
  subnet_id = subnet-00bca728
aws_subnet.rroy-intc-061600-enchantd-0:
  id = subnet-00bca728
  availability_zone = us-east-1b
  cidr_block = 100.xxx.xxx.0/24
  vpc_id = vpc-b8cb36dd

rroy:work-terraform rroy$rroy:work-terraform rroy$ terraform destroy .
aws_subnet.rroy-intc-061600-enchantd-0: Refreshing state... (ID: subnet-00bca728)
aws_instance.rroy-intc-061600-enchantd-0-0: Refreshing state... (ID: i-4c5033a2)
aws_instance.rroy-intc-061600-enchantd-0-0: Destroying...
aws_instance.rroy-intc-061600-enchantd-0-0: Destruction complete
aws_subnet.rroy-intc-061600-enchantd-0: Destroying...
aws_subnet.rroy-intc-061600-enchantd-0: Error: Error deleting subnet: The subnet 'subnet-00bca728' has dependencies and cannot be deleted. (DependencyViolation)
Error applying plan:

1 error(s) occurred:

* 1 error(s) occurred:

* Error deleting subnet: The subnet 'subnet-00bca728' has dependencies and cannot be deleted. (DependencyViolation)

Terraform does not automatically rollback in the face of errors.
Instead, your Terraform state file has been partially updated with
any resources that successfully completed. Please address the error
above and apply again to incrementally change your infrastructure.

# A minute or so later (once the instance destruction is complete)
# we're able to re-run the same cmd, this time successfully.

rroy:work-terraform rroy$ terraform destroy .
aws_subnet.rroy-intc-061600-enchantd-0: Refreshing state... (ID: subnet-00bca728)
aws_subnet.rroy-intc-061600-enchantd-0: Destroying...
aws_subnet.rroy-intc-061600-enchantd-0: Destruction complete

Apply complete! Resources: 0 added, 0 changed, 1 destroyed.
rroy:work-terraform rroy$ 

[mitchellh]

Thanks! This is fixed in master.

[pmoust]

@mitchellh I am experiencing the same with terraform just built from master

[russroy]

Likewise.

On Fri, Oct 3, 2014 at 5:57 PM, Panagiotis Moustafellos <
[email protected]> wrote:

@mitchellh https://github.com/mitchellh I am experiencing the same with
terraform just built from master

—
Reply to this email directly or view it on GitHub
#357 (comment).

[mitchellh]

Fixed

[russroy]

This is broken again (or at least the symptom is the same) in v0.3.6.dev.

[hschaeidt]

I'm running into the same issue. 0.3.6
This issue can be bypassed in the meantime by calling the terraform destroy twice.

[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.