Skip to content

Latest commit

 

History

History
171 lines (135 loc) · 5.47 KB

getting-started.md

File metadata and controls

171 lines (135 loc) · 5.47 KB
Raw

Getting started

Make the cloud controller image

The cloud-controller-manager has to be deployed in a kubernetes cluster. To do so you need an image which can be deployed as a kubernetes object.

Using the Makefile compile the cloud-controller-manager and make an image out of it

make docker-build

It will :

  • Fetch the go dependencies
  • Build the cloud-controller-manager
  • Make a docker image out of it

(Optional) You can push own fresh image of the controller to a registry :

# docker tag scaleway/scaleway-cloud-controller-manager:d1e51ce3 rg.fr-par.scw.cloud/test/scaleway-cloud-controller-manager:dev
docker push rg.fr-par.scw.cloud/test/scaleway-cloud-controller-manager:dev
The push refers to repository [rg.fr-par.scw.cloud/test/scaleway-cloud-controller-manager]
c6ff308811cb: Pushed 
cd099fd7777d: Pushed 
8dfad2055603: Pushed 
dev: digest: sha256:13ab072a17eec9694b40bb978f4ef75cb964ae65938a8b1c8bd7d6a8f2c1bd58 size: 950

Create a Kubernetes cluster on Scaleway using kubeadm

Create a Kubernetes cluster using kubeadm.

For the purpose on this example, you will need to create 3 ubuntu bionic instances.

  • master1
  • node1
  • node2

Run the following commands on each instances :

apt-get update && apt-get install -y \
    iptables \
    arptables \
    ebtables \
    apt-transport-https \
    ca-certificates \
    curl \
    gnupg-agent \
    software-properties-common
curl -s https://packages.cloud.google.com/apt/doc/apt-key.gpg | apt-key add -
add-apt-repository \
   "deb [arch=amd64] https://apt.kubernetes.io kubernetes-xenial main"
curl -fsSL https://download.docker.com/linux/ubuntu/gpg | apt-key add -
add-apt-repository \
   "deb [arch=amd64] https://download.docker.com/linux/ubuntu \
   $(lsb_release -cs) \
   stable"
apt-get update && apt-get install -y \
    docker-ce docker-ce-cli containerd.io kubelet kubeadm kubectl
apt-mark hold \
    docker-ce docker-ce-cli containerd.io kubelet kubeadm kubectl
echo KUBELET_EXTRA_ARGS=\"--cloud-provider=external\" > /etc/default/kubelet

Intialize the master:

kubeadm init --control-plane-endpoint=$(scw-metadata PUBLIC_IP_ADDRESS) --apiserver-cert-extra-sans=$(scw-metadata PUBLIC_IP_ADDRESS)
mkdir -p ~/.kube
cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
chown $(id -u):$(id -g) $HOME/.kube/config
kubectl apply -f https://docs.projectcalico.org/v3.11/manifests/calico.yaml

Write the kubeadm join command, you will need it to join worker nodes. Also, you can also copy the kubeconfig file to your computer.

Execute the join command on your nodes, repeat this operation on each worker nodes:

kubeadm join 10.68.34.145:6443 --token itvo0b.kwoao79ptlj22gno \
    --discovery-token-ca-cert-hash sha256:07bc3f9601f1659771a7a6fd696c2969cbc757b088ec752ba95d5a42c06ed91f

And then, execute on the master (or your computer if you copied the kubeconfig):

master1# kubectl get nodes
NAME      STATUS     ROLES    AGE     VERSION
master1   NotReady   master   5m43s   v1.17.2
node1     NotReady   <none>   2m17s   v1.17.2
node2     NotReady   <none>   77s     v1.17.2

The cluster is ready, you now have a working cluster, you can now deploy the cloud-controller-manager

Deploy the cloud-controller-manager on this cluster

To deploy the cloud-controller-manager you will need :

  • Your access key.
  • Your secret key.
  • Your organization or project id.
  • The Scaleway region.

Create a k8s-scaleway-secret.yml file containing these informations.

Hint: You can find those information on https://console.scaleway.com/account/credentials

apiVersion: v1
kind: Secret
metadata:
  name: scaleway-secret
  namespace: kube-system
stringData:
  SCW_ACCESS_KEY: "xxxxxxxxxxxxxxxx"
  SCW_SECRET_KEY: "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
  SCW_DEFAULT_REGION: "fr-par"
  SCW_DEFAULT_PROJECT_ID: "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxx"

Create the secret and deploy the controller

kubectl create -f k8s-scaleway-secret.yml
kubectl apply -f https://raw.githubusercontent.com/scaleway/scaleway-cloud-controller-manager/master/examples/k8s-scaleway-ccm-latest.yml

Check the cloud-controller-manager is working

Check the cloud-controller-manager is running.

root@master1:~# kubectl get pods -n kube-system -l app=scaleway-cloud-controller-manager
NAME                                                 READY   STATUS    RESTARTS   AGE
scaleway-cloud-controller-manager-774f5487d5-7z5dd   1/1     Running   7          3m
root@master1:~# kubectl get nodes
NAME      STATUS     ROLES    AGE     VERSION
master1   NotReady   master   5m43s   v1.17.2
node1     NotReady   <none>   2m17s   v1.17.2
node2     NotReady   <none>   77s     v1.17.2

Deploy a LoadBalancer service and verify a public ip is assigned to this service. If it is the case you're all good, you have deployed a cluster with kubeadm and the `scaleway-cloud-controlle-manager

Create a lb.yaml that contain the following object:

apiVersion: v1
kind: Service
metadata:
  name: example-service
spec:
  selector:
    app: example
  ports:
    - port: 8765
      targetPort: 9376
  type: LoadBalancer
# kubectl create -f lb.yml
# kubectl get services
root@master1:~# kubectl get services
NAME              TYPE           CLUSTER-IP      EXTERNAL-IP    PORT(S)          AGE
example-service   LoadBalancer   10.110.17.115   51.159.26.63   8765:32370/TCP   6d23h
kubernetes        ClusterIP      10.96.0.1       <none>         443/TCP          7d