Setup kubernetes cluster using Talos (Lab)

k8s-using-talos-in-vms.md

Setup Kubernetes cluster in minutes using Talos

1. Get talosctl

Download and install talosctl binary

wget https://github.com/talos-systems/talos/releases/download/v0.8.1/talosctl-linux-amd64

sudo install talosctl-linux-amd64 /usr/local/bin/talosctl 

2. Setup Docker Registry (Optional)

Create a local registry mirror for images required by talos for offline/quick installation

docker run -d -p 6000:5000 --restart always --name registry-airgapped registry:2

Identify the images required by talos

talosctl images

First, we pull all the images to our local Docker daemon

for image in `talosctl images`; do docker pull $image; done

Verify by running docker images command

Now we need to re-tag them so that we can push them to our local registry. We are going to replace the first component of the image name (before the first slash) with our registry endpoint 127.0.0.1:6000

for image in `talosctl images`; do \
    docker tag $image `echo $image | sed -E 's#^[^/]+/#127.0.0.1:6000/#'` \
done

As the next step, we push images to the internal registry

for image in `talosctl images`; do \
    docker push `echo $image | sed -E 's#^[^/]+/#127.0.0.1:6000/#'` \
done

We can now verify that the images are pushed to the registry

curl http://127.0.0.1:6000/v2/_catalog

The output should be similar

{"repositories":["autonomy/kubelet","coredns","coreos/flannel","etcd-development/etcd",
"kube-apiserver-amd64","kube-controller-manager-amd64","kube-proxy-amd64",
"kube-scheduler-amd64","talos-systems/install-cni","talos-systems/installer"]}

3. Create cluster nodes (VMs or Baremetal)

For the demonstration purpose we will use libvirtd/kvm for creating nodes in our lab environmnet

Download the ISO Image

In order to install Talos in Virtual Machines, you will need the ISO image from the Talos release page. You can download talos-amd64.iso from github.com/talos-systems/talos/releases

Create the master node

VM="k8s-master"
IP_ADDRESS="192.168.122.20"
MAC_ADDRESS="52:54:00:f2:d3:20" 

virsh net-update --network default \
     --command add-last --section ip-dhcp-host \
     --xml "<host mac='${MAC_ADDRESS}' name='${VM}' ip='${IP_ADDRESS}'/>" \
     --live --config

virt-install --name ${VM} \
     --ram 2048 --vcpus 2 --os-type linux --os-variant ubuntu20.04 \
     --network bridge=virbr0,mac=${MAC_ADDRESS} \
     --disk path=~/libvirt/images/${VM}.qcow2,size=20,device=disk,bus=scsi \
     --cdrom ~/Downloads/talos-amd64.iso \
     --console=pty,target.type=serial --graphics none --noautoconsole   

Create worker nodes

k8s-node1:

VM="k8s-node1"
IP_ADDRESS="192.168.122.21"
MAC_ADDRESS="52:54:00:f2:d3:21" 

virsh net-update --network default \
     --command add-last --section ip-dhcp-host \
     --xml "<host mac='${MAC_ADDRESS}' name='${VM}' ip='${IP_ADDRESS}'/>" \
     --live --config

virt-install --name ${VM} \
     --ram 2048 --vcpus 2 --os-type linux --os-variant ubuntu20.04 \
     --network bridge=virbr0,mac=${MAC_ADDRESS} \
     --disk path=~/libvirt/images/${VM}.qcow2,size=20,device=disk,bus=scsi \
     --cdrom ~/Downloads/talos-amd64.iso \
     --console=pty,target.type=serial --graphics none --noautoconsole   

k8s-node2:

VM="k8s-node2"
IP_ADDRESS="192.168.122.22"
MAC_ADDRESS="52:54:00:f2:d3:22" 

virsh net-update --network default \
    --command add-last --section ip-dhcp-host \
    --xml "<host mac='${MAC_ADDRESS}' name='${VM}' ip='${IP_ADDRESS}'/>" \
    --live --config

virt-install --name ${VM} \
     --ram 2048 --vcpus 2 --os-type linux --os-variant ubuntu20.04 \
     --network bridge=virbr0,mac=${MAC_ADDRESS} \
     --disk path=~/libvirt/images/${VM}.qcow2,size=20,device=disk,bus=scsi \
     --cdrom ~/Downloads/talos-amd64.iso \
     --console=pty,target.type=serial --graphics none --noautoconsole   

4. Form the cluster using talosctl

Generate the machine configurations to use for installing Talos and Kubernete. Using the DNS name of the load balancer, generate the base configuration files for the Talos machines

NOTE: We are using IP (192.168.122.20) of the master node in our example, since this is just a proof of concept type lab work In production environment this will be a DNS name of the load balancer

talosctl gen config talos-virsh-cluster https://192.168.122.20:6443 \
    --output-dir _out \
    --registry-mirror '*'=http://192.168.122.1:6000

This will create several files in the _out directory: init.yaml, controlplane.yaml, join.yaml, and talosconfig.

NOTE: If you have setup offline registry mirror, you can provide the local registry mirror using --registry-mirror option, by default talos will pull images from docker hub

Apply init.yaml to control-plane/master node

talosctl apply-config --insecure --nodes 192.168.122.20 --file _out/init.yaml

Note: This process can be repeated multiple times to create an HA control plane. Simply apply controlplane.yaml instead of init.yaml for subsequent nodes.

Create worker nodes using a process similar to the control plane creation above. Start the worker node VMs and wait for them to enter "maintenance mode".

talosctl apply-config --insecure --nodes 192.168.122.21 --file _out/join.yaml
talosctl apply-config --insecure --nodes 192.168.122.22 --file _out/join.yaml

5. Using the cluster

Once the cluster is available, you can make use of talosctl and kubectl to interact with the cluster.

First, configure talosctl to talk to your control plane node by issuing the following, updating paths and IPs as necessary:

export CONTROL_PLANE_IP=192.168.122.20
talosctl config endpoint $CONTROL_PLANE_IP
talosctl config node $CONTROL_PLANE_IP

Fetch the kubeconfig file from the control plane node by issuing:

talosctl kubeconfig

You can then use kubectl

kubectl get nodes

6. Deploy first app

Create a deployment

kubectl create deployment hello-world --image=k8s.gcr.io/echoserver:1.4

Expose the service to the world

kubectl expose deployment hello-world --type=NodePort --port=8080

7. Destroy cluster

Copy and pase following in your terminal

virsh shutdown k8s-master
virsh shutdown k8s-node1
virsh shutdown k8s-node2

virsh destroy k8s-master
virsh destroy k8s-node1
virsh destroy k8s-node2

virsh undefine k8s-master
virsh undefine k8s-node1
virsh undefine k8s-node2

virsh net-update default delete ip-dhcp-host "<host name='k8s-master'/>" --live --config
virsh net-update default delete ip-dhcp-host "<host name='k8s-node1'/>" --live --config
virsh net-update default delete ip-dhcp-host "<host name='k8s-node2'/>" --live --config


rm -rf ~/libvirt/images/* 

How do you handle the requests to https://discovery.talos.dev/ ?

nevermind, I figured it out. There is a spot in the config file to disable KubeSpan and enable kubernetes discovery.