-
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
[aws-eks] private endpoint doesn't work with looked up vpc #9542
Comments
After some investigation, it seems that the problem is: aws-cdk/packages/@aws-cdk/aws-eks/lib/cluster.ts Line 1052 in 9ff55ae
For looked up VPC's the type of the class is not |
Hi @iliapolo , thx a lot for raising this. really appreciate. Just my 2 cents on the subnet selections for the kubectl handler Lambda given the enterprise context: i. In most cases, we will configure vpcSubnets when calling eks.Cluster or FargateCluster to create eks clusters, so we can make sure the ENIs of the EKS control plane will be put into the subnets as required. But regardless the vpcSubnets is set or not, I guess by default the kubectl handler Lambda can just be associated with the subnets associated with the EKS Control plane. ii. However, that might not be enough. This kubectl handler Lambda fundamentally is applying kubectl/helm, so arguably it's mostly doing the tasks of the data plane. So it would be great if we have a separate subnet selection option in eks.Cluster & eks.FargateClusters specifically for the kubectl handler Lambda, sth like kubectlSubnets along with kubectlEnvironment. Having said that, I fully appreciate that you need to address this issue by considering the whole thing holistically. So please feel free to ignore this suggestion if it's contradicting to other factors you might need to consider. Thx a lot in advance. Jasper |
It's impossible to configure a private access endpoint when using a looked up vpc (i.e
Vpc.fromLookup
).Reproduction Steps
Create a VPC using the following guide: https://docs.aws.amazon.com/batch/latest/userguide/create-public-private-vpc.html
Deploy the following:
What did you expect to happen?
Expected for the cluster to created and for the config map to be correctly applied
What actually happened?
Environment
Other
Following is an excerpt from CloudFormation resource for the kubectl handler, we can see that the subnet id's are indeed empty:
This is 🐛 Bug Report
The text was updated successfully, but these errors were encountered: