AWS CDK Asset Kubectl v20 Layer

Warnings

• breaking This package is specifically for AWS CDK v2. If you are using AWS CDK v1, you should use the older `aws-cdk-asset-kubectl` package (without the `vXX` suffix), which is not compatible with CDK v2.
  Fix: Ensure your CDK application is using `aws-cdk.lib@^2.0.0` and that you install `aws-cdk-asset-kubectl-v20` for CDK v2. For CDK v1, use `aws-cdk-asset-kubectl`.
• gotcha This layer provides `kubectl v1.20` and `aws-iam-authenticator v0.5.2` ONLY. If your EKS cluster is running a significantly different Kubernetes version, you might encounter compatibility issues. Always try to match your `kubectl` client version with your cluster's server version.
  Fix: Verify your EKS cluster's Kubernetes version. If it requires a different `kubectl` client, look for an `aws-cdk-asset-kubectl-vXX` package that matches, or consider building your own custom layer.
• gotcha This specific `aws-cdk-asset-kubectl-v20` layer does NOT include the `helm` binary. Many users expect `kubectl` layers to also contain `helm`. If you need `helm` in your Lambda function, you must use a different layer such as `aws-cdk-asset-kubectl-helm-v20` (or a higher version depending on your `kubectl` requirements).
  Fix: If `helm` is required, install the appropriate `aws-cdk-asset-kubectl-helm-vXX` package and add `KubectlHelmVXXLayer` to your Lambda function's layers.
• gotcha The layer is compiled for the `x86_64` architecture. If your Lambda function is configured to use the `ARM64` architecture, this layer will not be compatible.
  Fix: Ensure your Lambda function is configured with `aws_lambda.Architecture.X86_64`. If `ARM64` is required, you will need to find or build a compatible layer for that architecture.

Install

• pip install aws-cdk-asset-kubectl-v20 Install package

Imports

• KubectlV20Layer

  from aws_cdk_asset_kubectl_v20 import KubectlV20Layer

  The v20 suffix is crucial for CDK v2 compatibility; the base package name (without suffix) typically refers to CDK v1 assets.

Quickstart

This quickstart demonstrates how to add the `KubectlV20Layer` to an AWS Lambda function within an AWS CDK stack. The example Lambda function simply runs `kubectl version --client` to confirm the binary's presence and version. Remember to adjust the Lambda runtime for compatibility with your function code and the layer.

import os
from aws_cdk import App, Stack
from aws_cdk.aws_lambda import Runtime, Function, Code
from aws_cdk_asset_kubectl_v20 import KubectlV20Layer

class MyKubectlStack(Stack):
    def __init__(self, scope, id, **kwargs):
        super().__init__(scope, id, **kwargs)

        # Instantiate the Kubectl v1.20 layer
        kubectl_layer = KubectlV20Layer(self, 'KubectlLayer')

        # Example Lambda function using the layer
        # Replace 'my_handler' with your actual handler code
        # and ensure the runtime is compatible.
        Function(self, 'MyKubectlFunction',
            runtime=Runtime.PYTHON_3_9, # Use a compatible Python runtime
            handler='index.handler',
            code=Code.from_inline(
                'import os\nimport subprocess\n\ndef handler(event, context):\n    # Kubectl is available in /opt/kubectl\n    kubectl_path = os.path.join(os.sep, 'opt', 'kubectl')\n    try:\n        result = subprocess.run([kubectl_path, 'version', '--client'], capture_output=True, text=True, check=True)\n        print(f"Kubectl client version: {result.stdout}")\n        return {'statusCode': 200, 'body': result.stdout}\n    except subprocess.CalledProcessError as e:\n        print(f"Error executing kubectl: {e.stderr}")\n        return {'statusCode': 500, 'body': e.stderr}\n'
            ),
            layers=[kubectl_layer]
        )

app = App()
MyKubectlStack(app, "MyKubectlStackExample")
app.synth()