You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
There seems to be an incompatibility with the latest Helm version and the AWS CLI version we ship in the workflow-runner image. (See also helm/helm#10975)
Current Behavior
When using the 0.12 garden container images starting with 0.12.57 together with AWS you get the error Error: Kubernetes cluster unreachable: exec plugin: invalid apiVersion "[client.authentication.k8s.io/v1alpha1](http://client.authentication.k8s.io/v1alpha1)"
Expected behavior
No error, it should work
Reproducible example
We can reproduce this internally
Workaround
Not really available for users; They can't easily upgrade the AWS CLI to v2 within the container, and they can't choose the helm version in Garden
Suggested solution(s)
Either we downgrade helm in 0.12 to 3.8 or we update the AWS CLI. I feel like preferring the latter even though it also has the potential to be breaking for some users. But the list of breaking changes in AWS CLI v1=>v2 is very small and we did not have trouble with it internally at all.
Additional context
Your environment
OS: macOS
How I'm running Kubernetes: AWS EKS
garden version 0.12.57
The text was updated successfully, but these errors were encountered:
… versions
This backports the new methods of installing the latest versions of AWS CLI, azure CLI and Google Cloud CLI from the main branch (Bonsai).
Fixes#4571 because there is an incompatibility between AWS CLI v1 and the latest helm versions.
… versions (#4574)
This backports the new methods of installing the latest versions of AWS CLI, azure CLI and Google Cloud CLI from the main branch (Bonsai).
Fixes#4571 because there is an incompatibility between AWS CLI v1 and the latest helm versions.
Bug
There seems to be an incompatibility with the latest Helm version and the AWS CLI version we ship in the workflow-runner image. (See also helm/helm#10975)
Current Behavior
When using the 0.12 garden container images starting with 0.12.57 together with AWS you get the error
Error: Kubernetes cluster unreachable: exec plugin: invalid apiVersion "[client.authentication.k8s.io/v1alpha1](http://client.authentication.k8s.io/v1alpha1)"
Expected behavior
No error, it should work
Reproducible example
We can reproduce this internally
Workaround
Not really available for users; They can't easily upgrade the AWS CLI to v2 within the container, and they can't choose the helm version in Garden
Suggested solution(s)
Either we downgrade helm in 0.12 to 3.8 or we update the AWS CLI. I feel like preferring the latter even though it also has the potential to be breaking for some users. But the list of breaking changes in AWS CLI v1=>v2 is very small and we did not have trouble with it internally at all.
Additional context
Your environment
garden version
0.12.57The text was updated successfully, but these errors were encountered: