Examkin

Soru

Zorluk: ZorDeploying and Managing Google Kubernetes Engine (GKE) Clusters and Workloads

You are deploying a new stateless microservice to an existing Google Kubernetes Engine (GKE) cluster named `prod-app-cluster` located in zone `us-central1-a`. You have authorized the `gcloud` CLI on a new developer workstation with appropriate IAM permissions for the target Google Cloud project. However, when you attempt to execute `kubectl apply -f deployment.yaml`, the command fails with an error stating `The connection to the server localhost:8080 was refused`. Which command should you execute to enable `kubectl` to communicate with the GKE cluster?

  1. Run `gcloud container clusters get-credentials prod-app-cluster --zone us-central1-a` to update the local kubeconfig file with cluster authentication details and endpoint information.Cevap
  2. B
    Run `gcloud config set container/cluster prod-app-cluster` to bind the default cluster property in gcloud settings.
  3. C
    Recreate the cluster as a GKE Autopilot cluster, because GKE Standard clusters do not allow external `kubectl` access without setting up an SSH tunnel to a node.
  4. D
    Append the `--enable-spot` flag to the `kubectl apply -f deployment.yaml` command to automatically target available Spot VM nodes.

Cevap

Execute `gcloud container clusters get-credentials prod-app-cluster --zone us-central1-a` to retrieve cluster entry points and credentials into the local environment.
Executing `gcloud container clusters get-credentials prod-app-cluster --zone us-central1-a` writes the cluster endpoint and authentication credentials into the client's `kubeconfig` file (`~/.kube/config`). This step configures `kubectl` with the correct control plane address and token necessary to deploy workloads to GKE.

Adım Adım Çözüm

1
Identify the root cause of the `localhost:8080 connection refused` error.
Recognize that `kubectl` defaults to `localhost:8080` when no cluster context or credentials are configured in `~/.kube/config`.
`kubectl` requires cluster endpoint address and valid authentication tokens to communicate with the GKE control plane.
2
Select the Google Cloud command designed to generate `kubeconfig` authentication settings.
Use `gcloud container clusters get-credentials` with the cluster name and region/zone flags.
This command securely retrieves the cluster's CA certificate, API endpoint, and generates an authentication token for the active GCP account.
3
Verify execution workflow.
Once `kubeconfig` is updated, `kubectl apply -f deployment.yaml` successfully targets the control plane of `prod-app-cluster`.
The local context is set as current, enabling seamless Kubernetes API requests.

Anahtar Kavram

Fetching GKE Cluster Credentials for kubectl authentication
Bu soruyu puanla

Konular

Deploying and Managing Google Kubernetes Engine (GKE) Clusters and Workloads