Tech Blog by vCluster Press and Media Resources

Day 2: Getting Started with vind: Your First Deployment with LoadBalancer

Mar 10, 2026

|

5

min Read

Day 2: Getting Started with vind: Your First Deployment with LoadBalancer

Yesterday I introduced vind and why I think it’s a better alternative to KinD. Today, let’s get hands-on. We’ll install vind, create a cluster, deploy an application, and see the built-in LoadBalancer in action.

Prerequisites

You need two things:

Docker installed and running
vCluster CLI v0.31.0 or later

# Check Docker

Command:

docker version

Output:

Client: Version: 28.5.1 API version: 1.51 Go version: go1.24.8 Git commit: e180ab8 Built: Wed Oct 8 12:16:17 2025 OS/Arch: darwin/arm64 Context: desktop-linux Server: Docker Desktop 4.48.0 (207573) Engine: Version: 28.5.1 API version: 1.51 (minimum version 1.24) Go version: go1.24.8 Git commit: f8215cc Built: Wed Oct 8 12:18:25 2025 OS/Arch: linux/arm64 Experimental: false containerd: Version: 1.7.27 GitCommit: 05044ec0a9a75232cad458027ca83437aae3f4da runc: Version: 1.2.5 GitCommit: v1.2.5-0-g59923ef docker-init: Version: 0.19.0 GitCommit: de40ad0

# Install vCluster CLI (macOS)

$ brew install loft-sh/tap/vcluster

# Or upgrade if you already have it

Command:

sudo vcluster upgrade --version v0.32.1

Output:

11:49:25 info Downloading version v0.32.1... 11:49:30 done Successfully updated vcluster to version v0.32.1

# Verify

Command:

vcluster --version

Output:

vcluster version 0.32.1

Set Docker as the Default Driver

This is the key step that tells vCluster to use Docker instead of an existing Kubernetes cluster:

Command:

vcluster use driver docker

Output:

11:50:00 done Successfully switched driver to docker

You only need to do this once. It persists across sessions.

Create Your First Cluster

Command:

vcluster create blog-demo

Output:

11:58:04 info Using vCluster driver 'docker' to create your virtual clusters, which means the CLI is managing Docker-based virtual clusters locally 11:58:04 info If you prefer to use helm or the vCluster platform API instead, use '--driver helm' or '--driver platform', or run 'vcluster use driver helm' or 'vcluster use driver platform' to change the default 11:58:04 info Ensuring environment for vCluster blog-demo... 11:58:05 done Created network vcluster.blog-demo 11:58:12 info Will connect vCluster blog-demo to platform... 11:58:13 info Starting vCluster standalone blog-demo 11:58:14 done Successfully created virtual cluster blog-demo 11:58:14 info Finding docker container vcluster.cp.blog-demo... 11:58:14 info Waiting for vCluster kubeconfig to be available... 11:58:17 info Waiting for vCluster to become ready... 11:58:29 done vCluster is ready 11:58:29 done Switched active kube context to vcluster-docker_blog-demo - Use `vcluster disconnect` to return to your previous kube context - Use `kubectl get namespaces` to access the vcluster

That took about 20 seconds. Let’s see what we got:

kubectl get nodes -o wide

Output:

NAME STATUS ROLES AGE VERSION INTERNAL-IP EXTERNAL-IP OS-IMAGE KERNEL-VERSION CONTAINER-RUNTIME blog-demo Ready control-plane,master 30s v1.35.0 172.23.0.2 <none> Ubuntu 24.04.3 LTS 6.10.14-linuxkit containerd://2.1.6

A single-node cluster running Kubernetes v1.35.0 with containerd. Let’s check the namespaces:

Command:

kubectl get namespaces

Output:

NAME STATUS AGE default Active 108s kube-flannel Active 99s kube-node-lease Active 108s kube-public Active 109s kube-system Active 109s local-path-storage Active 99s

Everything you’d expect from a real Kubernetes cluster, including Flannel for networking and local-path-storage for persistent volumes.

Deploy an Application

Let’s deploy nginx with 2 replicas and expose it via a LoadBalancer service:

$ kubectl create deployment nginx --image=nginx:latest --replicas=2 deployment.apps/nginx created

Create Service with type:LoadBalancer:

$ kubectl expose deployment nginx --type=LoadBalancer --port=80 --target-port=80 service/nginx exposed

Wait a few seconds for the pods to come up:

Get Pods:

$ kubectl get pods -o wide NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES nginx-b6485fcbb-6w6gm 1/1 Running 0 29s 10.244.0.4 blog-demo <none> <none> nginx-b6485fcbb-8dvvg 1/1 Running 0 29s 10.244.0.5 blog-demo <none> <none>

Get Service:

$ kubectl get svc NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE kubernetes ClusterIP 10.96.0.1 <none> 443/TCP 3m31s nginx LoadBalancer 10.111.30.236 172.23.255.254 80:31860/TCP 37s

Get Deployments:

$ kubectl get deployments NAME READY UP-TO-DATE AVAILABLE AGE nginx 2/2 2 2 69s

Both pods are running. The LoadBalancer service is created, on Linux it gets an external IP automatically. On macOS, you need to run the vCluster create command with sudo to get full LoadBalancer support with HAProxy.

Tip: On macOS, run sudo vCluster create my-cluster to enable LoadBalancer IP assignment via HAProxy. Without sudo, you can still access services via NodePort.

curl 172.23.255.254
<!DOCTYPE html>
<html>
<head>
<title>Welcome to nginx!</title>
<style>
html { color-scheme: light dark; }
body { width: 35em; margin: 0 auto;
font-family: Tahoma, Verdana, Arial, sans-serif; }
</style>
</head>
<body>
<h1>Welcome to nginx!</h1>
<p>If you see this page, the nginx web server is successfully installed and
working. Further configuration is required.</p>

<p>For online documentation and support please refer to
<a href="http://nginx.org/">nginx.org</a>.<br/>
Commercial support is available at
<a href="http://nginx.com/">nginx.com</a>.</p>

<p><em>Thank you for using nginx.</em></p>
</body>
</html>

Cluster Management

vind gives you simple commands to manage your cluster lifecycle:

# List all clusters

$ vcluster list

NAME | STATUS | CONNECTED | AGE

-----------+---------+-----------+--------

blog-demo | running | True | 102s

# Disconnect from a cluster (keeps it running)

$ vcluster disconnect

# Reconnect later

$ vcluster connect blog-demo

# Delete when done

$ vcluster delete blog-demo

What’s Under the Hood?

When you create a cluster, vind sets up Docker containers:

$ docker ps --format "table {{.Names}}\t{{.Status}}"

NAMES STATUS vcluster.lb.blog-demo.nginx.default Up 3 minutes vcluster.cp.blog-demo Up 6 minutes

vcluster.cp.blog-demo: The control plane container with K8s, etcd, and everything
vcluster.lb.blog-demo.nginx.default: The HAProxy load balancer for LoadBalancer services

You can even peek inside:

# View control plane logs

$ docker exec vcluster.cp.blog-demo journalctl -u vcluster --no-pager | tail -5

Mar 06 06:31:14 blog-demo vcluster[188]: 2026-03-06 06:31:14 INFO commandwriter/commandwriter.go:128 allocated clusterIPs {"component": "vcluster", "component": "apiserver", "location": "alloc.go:329", "service": "default/nginx", "clusterIPs": "{\"IPv4\":\"10.111.30.236\"}"} Mar 06 06:31:14 blog-demo vcluster[188]: 2026-03-06 06:31:14 INFO record/event.go:389 Event occurred {"component": "vcluster", "object": {"name":"nginx","namespace":"default"}, "fieldPath": "", "kind": "Service", "apiVersion": "v1", "type": "Normal", "reason": "EnsuringLoadBalancer", "message": "Ensuring load balancer"} Mar 06 06:31:14 blog-demo vcluster[188]: 2026-03-06 06:31:14 INFO cloudcontrollermanager/loadbalancer_docker.go:247 Image haproxy:3.3-alpine found locally, skipping pull. {"component": "vcluster"} Mar 06 06:31:14 blog-demo vcluster[188]: 2026-03-06 06:31:14 INFO cloudcontrollermanager/loadbalancer_docker.go:198 Started LoadBalancer container vcluster.lb.blog-demo.nginx.default with ip 172.23.255.254 and forward ports enabled {"component": "vcluster"} Mar 06 06:31:14 blog-demo vcluster[188]: 2026-03-06 06:31:14 INFO record/event.go:389 Event occurred {"component": "vcluster", "object": {"name":"nginx","namespace":"default"}, "fieldPath": "", "kind": "Service", "apiVersion": "v1", "type": "Normal", "reason": "EnsuredLoadBalancer", "message": "Ensured load balancer"}

# Check what's running inside the container

$ docker exec vcluster.cp.blog-demo crictl ps

CONTAINER IMAGE CREATED STATE NAME ATTEMPT POD ID POD NAMESPACE 0ce64fdd8ec9d 2af158aaca82b 5 minutes ago Running nginx 0 d6ced28cb4fe5 nginx-b6485fcbb-8dvvg default 56f9406394d3e 2af158aaca82b 5 minutes ago Running nginx 0 e5b6ab67d5930 nginx-b6485fcbb-6w6gm default 181ce2847ff1c b41714cf62496 7 minutes ago Running local-path-provisioner 0 536bc927ec4eb local-path-provisioner-5b9b9995f4-4d582 local-path-storage e7ab4ae96019d e3b7e577b2a83 7 minutes ago Running coredns 0 c765485409d89 coredns-79cf5f4c56-jh6jv kube-system d1865682d318c 253e2cac1f011 7 minutes ago Running kube-flannel 0 de40359cb6a81 kube-flannel-ds-5frp9 kube-flannel c9133c20d5937 de369f46c2ff5 7 minutes ago Running kube-proxy 0 52d2a265d3d3c kube-proxy-tvd85 kube-system

Custom Kubernetes Versions

Want a specific Kubernetes version? Easy:

Command:

vcluster create my-cluster \ --set controlPlane.distro.k8s.version=v1.32.0

Ouput:

NAME STATUS ROLES AGE VERSION my-cluster Ready control-plane,master 66s v1.32.0

No need to find and pull specific node images like with KinD. Just specify the version.

Cleanup

$ vcluster delete blog-demo info Removing vCluster container vcluster.cp.blog-demo... info Deleted kube context vcluster-docker_blog-demo done Successfully deleted virtual cluster blog-demo

Clean and fast.

Tomorrow: Multi-Node Clusters

Today we created a single-node cluster and deployed an app. But what if you need to test pod scheduling across nodes, node affinity, or node drains

Tomorrow, we’ll create a multi-node vind cluster with 3 worker nodes and see pods distributed across them, just like a real cluster.

All commands tested on macOS (Apple Silicon M1) with Docker Desktop and vCluster CLI v0.32.1.

vind is open source: github.com/loft-sh/vind , so do star the repo if you like vind

vind

vCluster

Platform Engineering

Open Source

Docker