[Bug] After deleting a cluster from the command line, creating a cluster with the same name from the command line will sometimes fail

[jwatte]

What were you trying to accomplish?

Try to create a cluster from the command line with the same name as a previously deleted cluster.

What happened?

After this, I'm trying to re-create a cluster with the same name, but get the error:

creating CloudFormation stack "eksctl-jwatte-cluster-cluster": operation error CloudFormation: CreateStack, https response error StatusCode: 400, RequestID: 2966a9ad-f4e4-4ab6-847f-2962e762e982, AlreadyExistsException: Stack [eksctl-jwatte-cluster-cluster] already exists

How to reproduce it?

I create a cluster.
I wait for it to come up.
I apply nginx-ingress manifests to set up an ingress controller and a NLB.
I wait for that to come up.
I delete the cluster.
The cluster goes away, and eksctl says the cluster is deleted.

Anything else we need to know?
Bug #2014 reports this same problem, and documents a work-around that involves the web console.
However, the whole point of eksctl is to drive this process from the command line, suitable for automated processes.
Thus, the work-around is not a solution to the problem, and an open bug needs to exist to document this problem.

Possible solutions would be:

1. make it possible for eksctl to re-use the existing cloudformation template from the command line
2. make it possible to force-delete (or diagnose-and-fix) the cloudformation deletion from the command line

Versions

$ eksctl info
eksctl version: 0.175.0
kubectl version: v1.30.0
OS: darwin
$

[github-actions]

Hello jwatte 👋 Thank you for opening an issue in eksctl project. The team will review the issue and aim to respond within 1-5 business days. Meanwhile, please read about the Contribution and Code of Conduct guidelines here. You can find out more information about eksctl on our website

[TiberiuGC]

Hi @jwatte , by default, eksctl will delete the cluster's cloudformation stack asynchronously. Running the eksctl delete cluster command with --wait flag should reveal if an error occurs during cluster stack deletion process.

Could you also please check AWS Console and share the actual error?