Skip to content

configuration.md

Latest commit

 

History

History
362 lines (270 loc) · 7.28 KB

configuration.md

File metadata and controls

362 lines (270 loc) · 7.28 KB

vind Configuration Guide

vind supports extensive configuration options to customize your Kubernetes clusters. This guide covers all available configuration parameters.

Configuration Methods

You can configure vind in several ways:

  1. Command-line flags - Quick options
  2. Values file - YAML configuration file
  3. Helm values - Using --set flags

Basic Configuration

Using a Values File

Create a vcluster.yaml file:

experimental:
  docker:
    # Network configuration
    network: "my-custom-network"
    
    # Container image
    image: "ghcr.io/loft-sh/vm-container"
    
    # Load balancer and registry proxy are enabled by default
    # No need to explicitly enable them

Then create the cluster:

vcluster create my-cluster -f vcluster.yaml

Configuration Options

Network Configuration

experimental:
  docker:
    # Use a custom Docker network
    # If not specified, a network will be created automatically
    network: "my-custom-network"

Example:

vcluster create my-cluster \
  --set experimental.docker.network=my-network

Additional Nodes

Add worker nodes to your cluster:

experimental:
  docker:
    nodes:
      - name: worker-1
        env:
          - "NODE_LABEL=worker"
        ports:
          - "30080:80"
        volumes:
          - "/host/data:/container/data"
        args:
          - "--cap-add=SYS_ADMIN"
      
      - name: worker-2
        # ... more configuration

Example:

vcluster create my-cluster \
  --set experimental.docker.nodes[0].name=worker-1

Port Mappings

Expose additional ports:

experimental:
  docker:
    ports:
      - "8080:80"      # Host:Container
      - "8443:443"
      - "30000-30100:30000-30100"  # Port ranges

Example:

vcluster create my-cluster \
  --set experimental.docker.ports[0]="8080:80" \
  --set experimental.docker.ports[1]="8443:443"

Volume Mounts

Mount host directories or volumes:

experimental:
  docker:
    volumes:
      - "/host/path:/container/path:ro"  # Read-only
      - "/host/data:/container/data:rw"  # Read-write
      - "volume-name:/container/path"     # Named volume

Example:

vcluster create my-cluster \
  --set experimental.docker.volumes[0]="/host/data:/container/data"

Environment Variables

Set environment variables:

experimental:
  docker:
    env:
      - "KEY1=value1"
      - "KEY2=value2"
      - "NODE_NAME=my-node"

Example:

vcluster create my-cluster \
  --set experimental.docker.env[0]="CUSTOM_VAR=value"

Extra Docker Arguments

Pass additional arguments to docker run:

experimental:
  docker:
    args:
      - "--cap-add=SYS_ADMIN"
      - "--cap-add=NET_ADMIN"
      - "--device=/dev/fuse"
      - "--security-opt=apparmor=unconfined"

Example:

vcluster create my-cluster \
  --set experimental.docker.args[0]="--cap-add=SYS_ADMIN"

Load Balancer Configuration

Load balancer is enabled by default. LoadBalancer services work out of the box with automatic IP assignment.

Benefits:

  • LoadBalancer services work out of the box
  • No need for MetalLB or other load balancer solutions
  • Automatic IP assignment

Note: On macOS with Docker Desktop, privileged port binding may require sudo access.

Registry Proxy Configuration

Registry proxy is enabled by default. It provides pull-through caching via local Docker daemon.

Benefits:

  • Faster image pulls
  • Reduced bandwidth usage
  • Use purely local images

Requirements:


## Node-Specific Configuration

Configure individual nodes:

```yaml
experimental:
  docker:
    nodes:
      - name: worker-1
        # Node-specific ports
        ports:
          - "30080:80"
        
        # Node-specific volumes
        volumes:
          - "/host/data:/data"
        
        # Node-specific environment variables
        # Note: These are Docker container env vars, not Kubernetes node labels
        env:
          - "CUSTOM_VAR=value"
        
        # Node-specific Docker args
        args:
          - "--cap-add=SYS_ADMIN"

Complete Example

Here's a complete configuration example:

experimental:
  docker:
    # Network
    network: "vind-network"
    
    # Container image
    image: "ghcr.io/loft-sh/vm-container"
    
    # Port mappings
    ports:
      - "8080:80"
      - "8443:443"
      - "30000-30100:30000-30100"
    
    # Volume mounts
    volumes:
      - "/host/data:/container/data"
      - "vind-storage:/var/lib/data"
    
    # Environment variables
    env:
      - "CLUSTER_NAME=my-cluster"
      - "ENVIRONMENT=development"
    
    # Extra Docker args
    args:
      - "--cap-add=SYS_ADMIN"
    
    # Load balancer and registry proxy are enabled by default
    
    # Additional nodes
    nodes:
      - name: worker-1
        env:
          - "NODE_LABEL=worker"
        ports:
          - "30080:80"
      
      - name: worker-2
        env:
          - "NODE_LABEL=worker"
        ports:
          - "30081:80"

Create the cluster:

vcluster create my-cluster -f vcluster.yaml

Kubernetes Version

Specify the Kubernetes version (defaults to v1.34.0):

controlPlane:
  distro:
    k8s:
      version: "v1.34.0"  # or any supported Kubernetes version

Or use a custom image tag:

controlPlane:
  distro:
    k8s:
      image:
        tag: "v1.34.0"

Note: The default Kubernetes version is v1.34.0. You can specify any supported version.

Advanced Configuration

Custom CNI

vind natively supports Flannel CNI. To use other CNI plugins (Calico, Cilium, etc.):

  1. Disable Flannel in your configuration:
deploy:
  cni:
    flannel:
      enabled: false

  1. Create the cluster with Flannel disabled

  2. Manually install your preferred CNI plugin after cluster creation

For example, to use Calico:

# After creating cluster and connecting
kubectl apply -f https://raw.githubusercontent.com/projectcalico/calico/v3.26.1/manifests/tigera-operator.yaml
kubectl apply -f https://raw.githubusercontent.com/projectcalico/calico/v3.26.1/manifests/custom-resources.yaml

Custom CSI

Configure Container Storage Interface:

storage:
  persistence: true
  size: "20Gi"
  storageClassName: "local-path"

Configuration Reference

For the complete configuration reference, see the vCluster configuration documentation.

Best Practices

  1. Use values files for complex configurations
  2. Version control your configuration files
  3. Test configurations in a development cluster first
  4. Document custom configurations for your team
  5. Use environment variables for sensitive data

Troubleshooting Configuration

Invalid Configuration

If you get configuration errors:

  1. Validate your YAML syntax
  2. Check the configuration reference
  3. Use --debug flag for detailed errors: 
    vcluster create my-cluster -f config.yaml --debug

Configuration Not Applied

If changes aren't applied:

  1. Delete and recreate the cluster
  2. Check for typos in configuration keys
  3. Verify the configuration format

For more help, see the Troubleshooting Guide.