Skip to content

Home

Jump to bottom
Zespre Chang edited this page Jul 31, 2024 · 4 revisions

Welcome to the KubeVirtBMC wiki!

The Slow Start Guide

You need to have a Kubernetes cluster accessible and have KubeVirt installed on that cluster. Here, I'll walk you through using a single-node cluster.

Prerequisites

  1. Install Ubuntu 22.04 on a physical server or virtual machine
  2. Install Kubernetes on the newly provisioned machine (I choose RKE2 as my distribution of choice) 
    # Install RKE2
curl -sfL https://get.rke2.io | sudo INSTALL_RKE2_VERSION=v1.28.11+rke2r1 sh -

# Start RKE2 service
sudo systemctl enable rke2-server.service --now
  3. Prepare your environment for quick access. By default, kubectl binary is at /var/lib/rancher/rke2/bin/kubectl and the kubeconfig file is at /etc/rancher/rke2/rke2.yaml. 
    mkdir -p ~/.kube
sudo cp /etc/rancher/rke2/rke2.yaml ~/.kube/config
sudo chown $(id -u):$(id -g) ~/.kube/config
export PATH=$PATH:/var/lib/rancher/rke2/bin
  4. Check if the machine is ready for running VMs on it 
    sudo apt update
sudo apt install libvirt-clients

# Make sure there's no "FAIL" in the command output
sudo virt-host-validate qemu
  5. Install KubeVirt (here I use v1.1.0 for example) 
    # Deploy KubeVirt operator
kubectl apply -f https://github.com/kubevirt/kubevirt/releases/download/v1.1.0/kubevirt-operator.yaml

# Instruct the operator to start deploy KubeVirt component by creating the KubeVirt custom resource
kubectl apply -f https://github.com/kubevirt/kubevirt/releases/download/v1.1.0/kubevirt-cr.yaml

KubeVirtBMC Installation

While we're still actively working on a Helm chart for easy installation (progress), the only way to install KubeVirtBMC at the moment is to clone the project source code and use make to create all the Kubernetes resources. So you'll have to make sure make and go commands are at your disposal.

# Install make
sudo apt install make

# Install Go
wget https://go.dev/dl/go1.22.5.linux-amd64.tar.gz -O go.tar.gz
sudo tar -xzvf go.tar.gz -C /usr/local
export PATH=$HOME/go/bin:/usr/local/go/bin:$PATH

Until we find out a more convenient way to provision the certificates for the webhook service to work, deploying cert-manager to the cluster is still required:

kubectl apply -f https://github.com/cert-manager/cert-manager/releases/download/v1.14.2/cert-manager.yaml

At last, we can install KubeVirtBMC (v0.3.0 is the latest version at the time being):

git clone https://github.com/starbops/kubevirtbmc.git
cd kubevirtbmc/
git checkout tags/v0.3.0 -b v0.3.0
make deploy IMG=starbops/virtbmc-controller:v0.3.0

Create a Dead VM

Before we fix the issue, KubeVirtBMC only supports VM defined with .spec.runStrategy. Otherwise, using the following command to create a test virtual machine is much faster:

kubectl apply -f https://kubevirt.io/labs/manifests/vm.yaml

Forget about the above command; let's do it the other way:

cat <<EOF | kubectl apply -f -
apiVersion: kubevirt.io/v1
kind: VirtualMachine
metadata:
  name: testvm
spec:
  runStrategy: Halted
  template:
    metadata:
      labels:
        kubevirt.io/size: small
        kubevirt.io/domain: testvm
    spec:
      domain:
        devices:
          disks:
            - name: containerdisk
              disk:
                bus: virtio
            - name: cloudinitdisk
              disk:
                bus: virtio
          interfaces:
          - name: default
            masquerade: {}
        resources:
          requests:
            memory: 64M
      networks:
      - name: default
        pod: {}
      volumes:
        - name: containerdisk
          containerDisk:
            image: quay.io/kubevirt/cirros-container-disk-demo
        - name: cloudinitdisk
          cloudInitNoCloud:
            userDataBase64: SGkuXG4=
EOF

Check if the VM is created successfully and remain in the stopped state:

$ kubectl get vms
NAME     AGE    STATUS     READY
testvm   2m     Stopped    False

Using IPMI to Power On the VM

There should already be a VirtualMachineBMC custom resource created by virtbmc-controller.

$ kubectl -n kubevirtbmc-system get virtualmachinebmcs
NAME             AGE
default-testvm   2m

You should be able to power on the VM by the ipmitool power on command, using a ephemeral Pod:

$ kubectl run -it --rm ipmitool --image=mikeynap/ipmitool --command -- /bin/sh
If you don't see a command prompt, try pressing enter.
/ # ipmitool -I lan -U admin -P password -H default-testvm-virtbmc.kubevirtbmc-system.svc.cluster.local power status
Chassis Power is off
/ # ipmitool -I lan -U admin -P password -H default-testvm-virtbmc.kubevirtbmc-system.svc.cluster.local power on
Chassis Power Control: Up/On
/ # ipmitool -I lan -U admin -P password -H default-testvm-virtbmc.kubevirtbmc-system.svc.cluster.local power status
Chassis Power is on
/ # exit
pod "ipmitool" deleted

Before the feature is fully delivered, the only way to access the IPMI endpoint of each VM is through the cluster service network (or the pod network if you'd like.)

Check if the VM is actually start running:

$ kubectl get vms
NAME     AGE    STATUS     READY
testvm   2m39   Running    True
$ kubectl get vmis
NAME     AGE     PHASE      IP            NODENAME    READY
testvm   2m39    Running    10.42.0.38    kv          True

Clone this wiki locally