Skip to content

CONTAINER.md

Latest commit

 

History

History
68 lines (50 loc) · 1.91 KB

CONTAINER.md

File metadata and controls

68 lines (50 loc) · 1.91 KB

Running kubectl-ai in a Docker Container

1. Build the Docker Image

First, clone the kubectl-ai repository and build the Docker image from the source code.

git clone https://github.com/GoogleCloudPlatform/kubectl-ai.git
cd kubectl-ai
docker build -t kubectl-ai:latest -f images/kubectl-ai/Dockerfile .

2. Running against a GKE cluster

To access a GKE cluster, kubectl-ai needs two configurations from your local machine: Google Cloud credentials and a Kubernetes config file.

Create Google Cloud Credentials

First, create Application Default Credentials (ADC). kubectl uses these credentials to authenticate with your GKE cluster.

gcloud auth application-default login

This command saves your credentials into the ~/.config/gcloud directory.

Configure kubectl

Next, generate the kubeconfig file. This file tells kubectl which cluster to connect to and to use your ADC credentials for authentication.

gcloud container clusters get-credentials <cluster-name> --location <location>

This updates the configuration file at ~/.kube/config.

3. Running the Container

Finally, mount both configuration directories into the kubectl-ai container when you run it. This example shows how to run kubectl-ai with a web interface, mounting all necessary credentials and providing a Gemini API key.

export GEMINI_API_KEY="your_api_key_here"
docker run --rm -it -p 8080:8080 \
  -v ~/.kube:/root/.kube \
  -v ~/.config/gcloud:/root/.config/gcloud \
  -e GEMINI_API_KEY \
  kubectl-ai:latest \
  --ui-listen-address 0.0.0.0:8080 \
  --ui-type web

Alternatively with the default terminal ui:

export GEMINI_API_KEY="your_api_key_here"
docker run --rm -it \
  -v ~/.kube:/root/.kube \
  -v ~/.config/gcloud:/root/.config/gcloud \
  -e GEMINI_API_KEY \
  kubectl-ai:latest