-
Notifications
You must be signed in to change notification settings - Fork 4k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
(eks): Use L1 construct (CfnCluster) for L2 eks.Cluster #18620
Comments
Hi, @literalice. If you look at the custom resource handler that is responsible for managing the lifecycle of EKS clusters, you'll see that there is a lot of custom logic involved. If we used the L1 directly, we would lose that. |
@otaviomacedo have we considered offering an alternate L2 construct, or an alternate EKS library for constructs which utilize the L1 EKS constructs? I agree with both of the use cases OP listed as valid reasons to want this. |
Thank you for your comments, I understand the EKS custom resource requires logics for managing the cluster's lifecycle. If it seems to be reasonable, the way to implementation would be following.
or
|
@peterwoodworth Yes, that's fair. We're looking into possible solutions for this. One of them is to write Amazon public extensions, which would remove the need for custom resources and provide all the features. It's still early stages, so I don't have any details at this point. But I'll keep the community updated. |
This issue has not received any attention in 1 year. If you want to keep this issue open, please leave a comment below and auto-close will be canceled. |
I'd rally like to see this re-opened |
Description
Currently L2 Cluster construct of eks module doesn't use L1 CfnCluster.
It's confusing, for example if users need to "escape hatches" adding overrides to L1 CfnCluster doen't work.
Use Case
Proposed Solution
Sample implementation is here.
Simply use the CfnCluster in the L2 construct:
Currently EKS cluster uses the principal that created the cluster as the administrator to the cluster so Kubectl provider should also use the principal. Specifying the CloudFormation execution role and using the role as kubectlRole would be the one of the solutions to the issue.
Other information
No response
Acknowledge
The text was updated successfully, but these errors were encountered: