fabric-samples/full-stack-asset-transfer-guide/docs/CloudReady/11-kube-multipass.md

Deploy a Kubernetes Cluster

PREV: Setup <==> NEXT: Deploy a Fabric Network

---

Ready?

just check-setup

WINDOWS note, please create the multipass VM with the command below from an elevated command prompt. Then proceed from logged into this VM with the KIND instructions

Provision a Multipass Virtual Machine

multipass launch \
  --name        fabric-dev \
  --disk        80G \
  --cpus        8 \
  --mem         8G \
  --cloud-init  infrastructure/multipass-cloud-config.yaml

# todo: scp not volume mounts
multipass mount $PWD fabric-dev:/home/ubuntu/full-stack-asset-transfer-guide

export WORKSHOP_IP=$(multipass info fabric-dev --format json | jq -r .info.\"fabric-dev\"."ipv4[0]")
export WORKSHOP_INGRESS_DOMAIN=$(echo $WORKSHOP_IP | tr -s '.' '-').nip.io

echo "Multipass VM created with IP: $WORKSHOP_IP"
echo "WORKSHOP_DOMAIN=$WORKSHOP_INGRESS_DOMAIN"

Start a KIND Cluster

• Open a shell on the virtual machine:

# todo ssh authorized_keys -> ubuntu@${WORKSHOP_IP} not multipass shell
multipass shell fabric-dev

cd ~/full-stack-asset-transfer-guide

# Bind a docker container registry to the VM's external IP  
export CONTAINER_REGISTRY_ADDRESS=0.0.0.0
export CONTAINER_REGISTRY_PORT=5000

# Expose the Kube API controller on the VM's public interface
export KIND_API_SERVER_ADDRESS=$(hostname -I | cut -d ' ' -f 1)
export KIND_API_SERVER_PORT=8888

# Create a Kubernetes cluster in Docker, configure an Nginx ingress, and docker container registry
just kind

# KIND will set the current kube client context in ~/.kube/config
kubectl cluster-info

# Copy the kube config to the host OS volume share:
# todo: scp not volume share
cp ~/.kube/config ~/full-stack-asset-transfer-guide/config/multipass-kube-config.yaml

• Exit the multipass VM

• From the host OS:

# Connect the local kube client to the k8s API server running on the VM:
cp config/multipass-kube-config.yaml ~/.kube/config

# Display kube client connection to k8s running on the VM:
kubectl cluster-info

# Observe the target Kubernetes workspace:  
k9s -n test-network

Troubleshooting:

• Run KIND on your local system
• Run KIND on an EC2 instance at AWS
• ssh to a workshop EC2 instance (see the login information on the back of your Conga Trading Card)

Take it Further:

• Run k8s directly on your laptop with KIND (export WORKSHOP_DOMAIN=localho.st)
• Provision an EC2 instance on your AWS account with a #cloud-config
• Connect your kube client to a cloud k8s provider

---

PREV: Setup <==> NEXT: Deploy a Fabric Network