How to Create a Proper Reproduction

Create a minimal reproduction that is immediately deployable and demonstrates the behavior in question.

Getting Started

Fork and Clone this repo
Update these variables in main.tf: (last updated 2024-30-01)
- Anything in locals
- Module Version numbers, and
- eks_blueprints_addons in helm-test.tf if necessary
Run terraform init -get=true && terraform apply -auto-approve; takes ~12 minutes to build
Get the kubeconfig file for testing, et al.:

aws eks update-kubeconfig --name <cluster_name> --region <region>
Push to your forked repo
Collect relevant details: outputs, logging, etc.
Provide those details and a link to your forked repo in the issue

When done:

terraform apply -destroy -auto-approve

Its immediately deployable
There is no ambiguity hidden behind unknown variables
It is minimal yet has everything required to simply run terraform init && terraform apply; we don't need to bring all the bells and whistles, we just want to focus on the relevant bits necessary reproduce the behavior in question

It does not reproduce the error referenced, but that was the intent of this - to show others what a reproduction should look like so that they can modify to demonstrate the behavior that plagues them, or they deem undesirable. Once we have this, we can discuss whether this is by design or there is a flaw in the module that should be corrected/addressed.