random provider returns same result with each apply/destroy cycle

[pmoosh]

Terraform CLI and Provider Versions

Terraform v0.14.9
provider registry.terraform.io/hashicorp/random v3.1.1

Terraform Configuration

resource "random_pet" "hostname" {
  length    = 2
  separator = "-"
}


output "hostname" {
  value =  random_pet.hostname

}

terraform {
  required_providers {
    random = {
      source = "hashicorp/random"
      version = "3.1.1"
    }
  }
}

Expected Behavior

should return different hostname output with each apply/destroy cycle

Actual Behavior

always returns touched-walleye

Steps to Reproduce

1. ` terraform init'
2. terraform apply -auto-approve
3. `terraform destroy -auto-approve
4. terraform apply -auto-approve
5. terraform destroy -auto-approve

How much impact is this issue causing?

High

Logs

no logs

Additional Information

this issue was introduced with version change from 3.1.0 to 3.1.1

Code of Conduct

• I agree to follow this project's Code of Conduct

[jkerry]

Sifting through the release diff the upgrade on this dependency pulled in 2 years of change and might have been unintentional:

release diff
dependency change

[detro]

It looks like that in updating the dependency for generating petname, the developer has squeezed in a "reversal" of the default behaviour: dustinkirkland/golang-petname@8e5a1ed

The little library will generate the same name in order, unless non-determinism is desired (and that's what we want here).

I'll look into how easy it is to update our use of it VS merging a dependency reversal in @jkerry PR (thank you btw)

[github-actions]

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.