eks module tries to force cluster recreation if no cluster_security_group_id specified

[davidaah]

Description

There seems to be a regression of a very old issue #798 where if
cluster_security_group_id is not specified and creation is disabled, the EKS module will try and replace the cluster every time.

Technically, the EKS module documentation does indicate cluster_security_group_id is a required parameter if create_cluster_security_group is set to false, however it is an optional parameter in the dependent aws_eks_cluster module and it should not be required since EKS creates a baseline cluster security group on cluster creation no matter what.

I think this should be as simple as wrapping this line in compact() --
https://github.com/terraform-aws-modules/terraform-aws-eks/blob/master/main.tf#L16 to restore the previous behavior of removing any empty string values

Versions

• Terraform:
  • Terraform v1.1.4 on darwin_amd64
• Provider(s):
  • provider registry.terraform.io/hashicorp/aws v3.74.3
  • provider registry.terraform.io/hashicorp/cloudinit v2.2.0
  • provider registry.terraform.io/hashicorp/tls v3.1.0
• Module:
  • 18.2.7

Reproduction

Steps to reproduce the behavior:

• Run terraform plan/terraform apply
• After first apply, re-run terraform plan

      ~ vpc_config {
          ~ cluster_security_group_id = "sg-XXXXXX" -> (known after apply)
          ~ security_group_ids        = [ # forces replacement
              + "",
            ]
          ~ vpc_id                    = "vpc-YYYYY" -> (known after apply)
            # (4 unchanged attributes hidden)
        }

Code Snippet to Reproduce

module "eks" {
  source  = "terraform-aws-modules/eks/aws"
  version = "18.7.2"

  cluster_name                    = "test-cluster"
...

  create_cluster_security_group = false
  create_node_security_group    = false
...
}

Expected behavior

Modules should not detect any changes

Actual behavior

Module thinks something has changed and tries to destroy

[drunkirishcoder]

can confirm, ran into this bug too, and compact fixes it.

[antonbabenko]

This issue has been resolved in version 18.10.1 🎉

[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.