Docs

Volcengine Kubernetes Engine

Getting started

Using VKE via kubectl

This topic describes how to deploy applications in a VKE cluster by using the kubectl command-line tool.

Step 1: Create a cluster

For the detailed procedure and parameter description, see Step 1: Create a cluster in the Using VKE via the console topic.

tip

To use an existing cluster, make sure that access to the API server over the Internet is enabled.

Step 2: Connect to the cluster

You can use the kubectl client to connect to the cluster you created.

Download and install the latest version of kubectl in your local environment.

We recommend downloading kubectl from the official Kubernetes website. For installation details, see Installing and configuring kubectl.

On the Cluster page of the VKE console, find the cluster demo-cluster you have created and click its name.

On the Overview page of the cluster, click the Connection tab.

Select the Public network tab, and click Generate Kubeconfig. In the pop-up window, set Valid period (365 days in this example) to obtain the cluster access credentials.

In your local environment, open the $HOME/.kube/config file and paste the obtained cluster access credentials into it.

warning

$HOME/.kube/config is the file generated after you install the kubectl client in your local environment. $HOME is a variable, and its value depends on your local operating system (such as macOS, Linux, or Windows) and the path where you install the kubectl client.

Run the following command in your kubectl client to query the status of nodes (including virtual nodes) in the cluster and check whether the connection is successful.

kubectl get node

If output similar to the following information is displayed, kubectl is connected to the cluster.

NAME STATUS ROLES AGE VERSION
192.168.0.6 Ready <none> 4h15m v1.26.10-vke.11

Step 3: Deploy an application

After connecting to the cluster from kubectl, use the kubectl client to create a Deployment for deploying an application.

In your local environment, create a YAML file named nginx.yaml and paste the following sample code into it.

tip

Modify the parameter values based on the comments in the sample code.

apiVersion: apps/v1
kind: Deployment
metadata:
name: doc-deployment # Deployment name
namespace: default # Namespace name
spec:
replicas: 1 # Number of pod instances
selector:
matchLabels:
app: doc-deployment
template:
metadata:
labels: # # Make sure this matches the selector value in the Service YAML file created later.
app: doc-deployment
spec:
containers:
- name: doc-deployment-container # Container name
image: doc-cn-beijing.cr.volces.com/vke/nginx-demo:v1.0 # You can replace it with your own image address, or use the address provided here
ports:
- containerPort: 80 # Exposed container port number
protocol: TCP # The network protocol

Run the following command to deploy the application.

kubectl apply -f nginx.yaml

The following output is expected:

deployment.apps/doc-deployment created

Run the following command to view the deployment status.

kubectl get deployment ${your_deployment_name}

tip

Replace ${your_deployment_name} with the name of the Deployment. For example, the complete command looks like the following if the name is doc-deployment:

If the output is similar to the following information, the application has been deployed.

NAME READY UP-TO-DATE AVAILABLE AGE
doc-deployment 1/1 1 1 85s

In the $home directory on your local computer, create an nginx-svc.yaml file and copy the following sample code into that file.

tip

Modify parameter values according to the comments in the sample code as needed. For more descriptions, see Creating Load Balancer services through kubectl.

apiVersion: v1
kind: Service
metadata:
name: nginx-svc # Service name
namespace: default # Namespace name
spec:
selector: # Make sure that this matches the value of the matchLabels parameter under the selector of the previously created application file
app: doc-deployment
ports:
- protocol: TCP # Port protocol
name: doc-port # The port mapping name
port: 80 # The Service port.
targetPort: 80 # Container port: the port number or port name through which the workload provides services. For example, the default port for Nginx is 80.
type: LoadBalancer # Service type. Valid values: Nodeport, ClusterIP, and LoadBalancer.

Run the following command to create a Service.

kubectl apply -f nginx-svc.yaml

The information similar to the following message is expected to be displayed.

service/nginx-svc created

Run the following command to view the Service status.

kubectl get svc ${your_service_name}

tip

Replace ${your_service_name} with the service name you specified. For example, if the name is nginx-svc, the complete command is as follows:

If the output is similar to the following information, the application is deployed.

NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
nginx-svc LoadBalancer 172.**.**.** 180.**.**.** 80:31709/TCP 65s

In your browser, enter the EXTERNAL-IP returned in the previous step to access the Nginx welcome page.

Step 4: Monitor the application

After the application is deployed, you can use cloud-native observability to monitor its status, including CPU and memory usage, to facilitate subsequent O&M work. For details, see Enabling observability.

Release resources

If you no longer need a cluster, delete it and its associated resources by following the instructions in Deleting Clusters. To keep the cluster running without interruptions, ensure that your Volcano Engine account balance is at least CNY 100 (including any vouchers) and greater than zero (excluding vouchers).

If you continue to use VKE clusters, you will be charged the following two types of fees. For details, see Product billing.

Cluster management fees charged by VKE.

Cloud infrastructure costs: Volcano Engine resources used in the cluster, such as NAT Gateway, CLB, EIP, and EBS, are billed separately by the corresponding cloud services.

Last updated: 2026.07.16 15:55:45