12 week routine of practice and implementation details

• Kubernetes the hard way
• Create local cluster with kind
• create local cluster with minikube
• Use kubeadm to create a K8s cluster on Ubuntu

• Free courses on ACG Oct 2022
• Creating user and role in K8s
• How HTTPS works
• PKI - part 1
• PKI - part 2
• Chapter 1 - Section 6

Put all YAML files in WEEK2 directory

1. Create the YAML for a certificate signing request.
2. Create the YAML for a role named "pod-reader" that allows users to perform get, watch and list on pods.
3. Create the YAML for a roleBinding named "admin-binding", in the namespace "acme", granting permissions for the "admin" ClusterRole to a user named "bob".

• VIDEO: Create custom image and push to DockerHub
• VIDEO: Deployment vs. StatefulSet
• VIDEO: Pods and Containers
• VIDEO: Pod Lifecycle
• VIDEO: Create Pods & Deployment

Put all YAML files in WEEK3 directory

1. Take the image that you built in the first video above and deploy it as a pod in Kubernetes
2. Create a deployment with the image used in the previous challenge and scale it to 5 replicas
3. Create a second deployment that uses the image "mysql:8" and include an environment variable that is the MYSQL default password

• VIDEO: What is Helm and Helm Charts
• VIDEO: Introduction to Helm
• VIDEO: Hands on w/Helm
• Helm Quickstart
• The Big Three Concepts
• Helm Tutorial
• Helm vs. Kustomize

Put all YAML files in WEEK4 Directory

1. Install the nginx ingress controller nginx-stable via https://helm.nginx.com/stable
2. Install Hashicorp Vault via helm chart hashicorp/vault at https://helm.releases.hashicorp.com
3. Override the values file for vault using the default here: https://github.com/hashicorp/vault-helm/blob/main/values.yaml
4. Create a helm chart from scratch and deploy it to Kubernetes

• VIDEO: Types of Services in K8s
• Services - K8s Documentation
• EXERCISE: Use a service to expose your app
• EXERCISE: Using Source IP
• EXERCISE: Access services running in K8s

Put all the files for these exercises in the WEEK5 directory

1. Create a NodePort service in K8s that uses port 30000 on the node and port 8080 on the container
2. Only use the kubectl command-line to create a deployment named 'nginx-deploy' that uses the nginx image, exposed on port 80. Only 1 replica will be used.
3. Again, only use the kubectl command-line to expose the deployment 'nginx-deploy' and use the target port of 8080.
4. Change the service named 'kubernetes' already running in the default namespace to a NodePort type service using the kubectl command line

• VIDEO: Rolling Updates
• TUTORIAL: Scaling an App
• VIDEO: Using Set Image

Put all files in WEEK6 Directory

1. Create a deployment named “apache” that uses the image httpd:2.4.54 and contains three pod replicas.
2. After the deployment has been created, scale the deployment to five replicas
3. Change the image for the "apache" deployment to httpd:alpine
4. Look at the rollout history, then go back to the previous rollout (roll back)

• VIDEO: Network Policies
• VIDEO: Network Policy & Service Mesh Differences
• VIDEO + GITHUB REPO: Network Policies
• DOCS: Network Policies
• Network Policy Tutorial

1. Deploy the following resources in Kubernetes: https://gist.github.com/chadmcrowell/7e5cbb782bf2b9e90c6abe8cd8cd39b0
2. Start a new pod with the image radial/busyboxplus:curl, get a shell to it, and run curl http://db:15984 to try to communicate from pod to pod
3. Create a network policy that will deny all traffic from pod to pod
4. Create another network policy that applies to pods that have the label app=db. Allow traffic from pods that have the label app=api and only over port 5984.
5. Create another network policy that applies to pods that have the label app=web. Allow traffic from pods that have the label app=web on port 3000. Allow traffic out to pods that have the label app=db over port 5984. Also allow traffic to pods in the kube-system namespace that match label k8s-app=kube-dns on port 53.
6. Start a new pod with the image radial/busyboxplus:curl and label app=api, get a shell to it, and run curl http://db:15984 to try to communicate from pod to pod