Terraform-aws-eks: Don't want to use a EKS cluster name for the name_prefix of Auto Scaling Group

[tsub]

I have issues

The Auto Scaling Group name_prefix uses the EKS cluster name.

https://github.com/terraform-aws-modules/terraform-aws-eks/blob/6ea3582940c0312b5031c8c9c8dafb1e86a98ceb/workers.tf#L5

In this case, there are the following problems.

For example, when replacing the Auto Scaling Group with Blue-Green Deployment, use the name parameter to specify blue orgreen.

module "eks" {
  source                 = "terraform-aws-modules/eks/aws"
  cluster_name           = "staging"
  ...

  worker_groups = [
    {
      name                          = "blue"
      ...
    },
    {
      name                          = "green"
      ...
    },
  ]
}

If you also want EKS clusters to be Blue-Green Deployment, add -blue or-green to cluster_name parameter.

module "eks" {
  source                 = "terraform-aws-modules/eks/aws"
  cluster_name           = "staging-blue"
  ...

  worker_groups = [
    {
      name                          = "blue"
      ...
    },
    {
      name                          = "green"
      ...
    },
  ]
}

module "eks" {
  source                 = "terraform-aws-modules/eks/aws"
  cluster_name           = "staging-green"
  ...

  worker_groups = [
    {
      name = "blue"
      ...
    },
    {
      name = "green"
      ...
    },
  ]
}

As a result, the Auto Scaling Group's name_prefix parameter will have a duplicate blue, such asstaging-blue-blue-.

I'm submitting a...

• [ ] bug report
• [x] feature request
• [x] support request
• [ ] kudos, thank you, warm fuzzy

What's the expected behavior?

For example, how about adding a parameter that can specify an arbitrary name prefix instead of ${aws_eks_cluster.this.name}.

diff --git a/workers.tf b/workers.tf
index c50e1bc..54967a2 100644
--- a/workers.tf
+++ b/workers.tf
@@ -2,7 +2,7 @@

 resource "aws_autoscaling_group" "workers" {
   count       = local.worker_group_count
-  name_prefix = "${aws_eks_cluster.this.name}-${lookup(var.worker_groups[count.index], "name", count.index)}"
+  name_prefix = "${lookup(var.worker_groups[count.index], "name_prefix", aws_eks_cluster.this.name)}-${lookup(var.worker_groups[count.index], "name", count.index)}"
   desired_capacity = lookup(
     var.worker_groups[count.index],
     "asg_desired_capacity",

However, I do not think that the variable name name_prefix is easy to understand.

I can not think of good parameter names.

Or is there any better way?

Are you able to fix this problem and submit a PR? Link here if you have already.

I have not made a PR yet.

Environment details

• Affected module version: v5.1.0
• OS: any
• Terraform version: any

Any other relevant info

none

[max-rocket-internet]

I don't really like the name_prefix solution either but changing is a large breaking change as it requires all resources to be recreated. This makes the upgrade path difficult.

Please add your voice to https://github.com/terraform-aws-modules/terraform-aws-eks/issues/88

[tsub]

@max-rocket-internet

This issue is not a problem using name_prefix. The problem is that $ {aws_eks_cluster.this.name} used in name_prefix cannot be changed.

Therefore, I think that the purpose is different from Issue https://github.com/terraform-aws-modules/terraform-aws-eks/issues/88.

[max-rocket-internet]

@tsub very sorry, I think wasn't paying close enough attention. Reopening!

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[stale[bot]]

This issue has been automatically closed because it has not had recent activity since being marked as stale.

[michcio1234]

Let's keep this issue open. I am going to get back to #709 this or next week and fix it.

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[stale[bot]]

This issue has been automatically closed because it has not had recent activity since being marked as stale.

[yamamushi]

Was there ever any follow up to this issue? It seems to have been closed without being solved :-)

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[stale[bot]]

This issue has been automatically closed because it has not had recent activity since being marked as stale.

[debu99]

nobody follow?