Training Guide

Deploy K3s

The install.sh script provides a convenient way to download K3s and add a service to systemd or openrc. To install k3s as a service just run:

curl -sfL https://get.k3s.io | sh -

A kubeconfig file is written to /etc/rancher/k3s/k3s.yaml and the service is automatically started or restarted. The install script will install K3s and additional utilities, such as kubectl, crictl, k3s-killall.sh, and k3s-uninstall.sh, for example:

sudo kubectl get nodes

K3S_TOKEN is created at /var/lib/rancher/k3s/server/node-token on your server. To install on worker nodes we should pass K3S_URL along with K3S_TOKEN or K3S_CLUSTER_SECRET environment variables, for example:

curl -sfL https://get.k3s.io | K3S_URL=https://myserver:6443 K3S_TOKEN=XXX sh -

Kubernetes

Namespace

kubectl create namespace intro-k8s

Pod

kubectl create --namespace intro-k8s -f pod.yaml
kubectl logs myapp-pod
kubectl delete pod -n intro-k8s myapp-pod

Deployment

• we can launch random stuff, but this isn't repeatable

kubectl create -f deployment.yaml -n intro-k8s
kubectl get deployments -n intro-k8s
kubectl get pods -n intro-k8s

• describe pod, look at the image
• scale deployment manually

kubectl scale deploy/nginx --replicas=3
kubectl rollout status deploy/nginx
kubectl get deploy
kubectl get po

• upgrade with bad image

kubectl set image deploy/nginx nginx=nginx:1.17-alpne --record
kubectl rollout status deploy/nginx
kubectl get po
kubectl rollout undo deploy/nginx

• redo upgrade from manifest

kustomize build deployment/base

• edit base to change image and then apply

kubectl apply -k deployment/base

• how can we use this for different environments?

kustomize build deployment/overlay/staging
kustomize build deployment/overlay/production
kubectl apply -k deployment/overlay/staging
kubectl apply -k deployment/overlay/production
kubectl get deploy
kubectl get pods

ConfigMaps

• show ConfigMaps
• explain what they're for
• explain how they're generated by Kustomize
• they'll show up later

Services

• show services listening as NodePort
• go look at them

curl -I training-a:<port>

Ingress

• show deployment/overlay/ingress/single/ingress.yaml

kubectl apply -k deployment/overlay/ingress/single
kubectl get ingress

• visit https://training-a.cl.monach.us
• what about multiple apps?

kustomize build deployment/overlay/ingress/fanout
kubectl apply -k deployment/overlay/ingress/fanout 

• visit https://training-a.cl.monach.us (fail)
• visit https://training.cl.monach.us/nginx (works)
• deploy rancher-demo application

kustomize build rancher-demo/base
kubectl apply -k rancher-demo/base

• visit https://training-a.cl.monach.us/

Load Balancer (metalLB)

kubectl apply -f https://raw.githubusercontent.com/metallb/metallb/v0.10.2/manifests/namespace.yaml
kubectl apply -f https://raw.githubusercontent.com/metallb/metallb/v0.10.2/manifests/metallb.yaml

Rancher

Server Deploy

docker run -d --restart=unless-stopped -p 80:80 -p 443:443 -v /opt/rancher:/var/lib/rancher rancher/rancher:v2.4.5

Node Deploy

• Show how we would deploy an RKE cluster
• Import the training-a k3s cluster

Rancher Server Walkthrough

• Clusters
• Authentication & Security
• Storage
• Projects
• Namespaces
• Catalogs
• CLI/API/Kubectl

Application Deployment

• show workloads on running cluster
• edit them / delete them
• redeploy monachus/rancher-demo as a workload
  • expose port 8080
• put an Ingress in front of it
  • use training-a.cl.monach.us
• scale it

Rancher Application Catalog

• use the Hadoop example