How to fix stack deletion failures caused by externally deleted resources

If you're unable to delete a stack because a resource was deleted outside of Terraform, the deletion process may fail during the planning refresh step. This happens because Terraform is still trying to read the missing resource from its state file.

Solution: Remove the stale resource from Terraform state

To resolve this issue, you need to detach the stale resource from the Terraform state using one-off Tasks:

1. Identify the resource address: Run a one-off Task with terraform state list to locate the exact address of the resource that's causing the failure.

2. Remove the resource from state: Run another one-off Task with terraform state rm [resource_address] to remove the problematic resource from the state file.

Why this happens

When a resource is deleted outside of Terraform (manually or through other tools), Terraform's state file still contains a reference to that resource. During the refresh step of any Terraform operation, Terraform attempts to read the current state of all resources, which fails for the deleted resource and causes the entire operation to fail.

Additional resources

For more detailed information about removing resources from Terraform state, including how to handle multiple resources, check out our comprehensive guide: Terraform State RM blog post.