K3S: GitOps & CI/CD

Introduction to GitOps and Continuous Deployment#

GitOps is the answer to that chaotic, last-minute “did anyone deploy the latest version?” question. It treats Git as the single source of truth, ensuring deployments are automated, consistent, and—most importantly—not reliant on someone’s memory.

What is GitOps, and why use it for Kubernetes deployments?#

GitOps automates Kubernetes deployments by continuously syncing cluster configurations from a Git repository. This means fewer manual kubectl commands and more predictable deployments.

Difference between CI/CD and GitOps#

Popular GitOps tools: ArgoCD, FluxCD#

• ArgoCD: User-friendly, UI-based GitOps controller.
• FluxCD: Lightweight and native to Kubernetes.

Using ArgoCD for Continuous Deployment#

ArgoCD lets you deploy, sync, and rollback applications with just a Git commit. Let’s get it up and running in K3s.

Installing ArgoCD in K3s#

kubectl create namespace argocd
kubectl apply -n argocd -f https://raw.githubusercontent.com/argoproj/argo-cd/stable/manifests/install.yaml

Accessing the ArgoCD UI#

kubectl port-forward svc/argocd-server -n argocd 8080:443

Now, visit https://localhost:8080 and log in.

Deploying an application using ArgoCD#

argocd app create myapp --repo https://github.com/myrepo.git --path k8s-manifests --dest-server https://kubernetes.default.svc --dest-namespace default

ArgoCD will now sync your app automatically with the Git repo!

Integrating K3s with GitHub Actions and Jenkins#

Automation is key. Let’s get GitHub Actions and Jenkins to deploy our applications on K3s.

Using GitHub Actions for Kubernetes Deployment#

name: Deploy to K3s
on: push
jobs:
  deploy:
    runs-on: ubuntu-latest
    steps:
      - name: Checkout repository
        uses: actions/checkout@v2
      - name: Set up kubectl
        run: |
          echo "${{ secrets.KUBECONFIG }}" | base64 --decode > kubeconfig.yaml
          export KUBECONFIG=kubeconfig.yaml
      - name: Apply Kubernetes manifests
        run: kubectl apply -f k8s/

This workflow triggers a deployment whenever changes are pushed.

Using Jenkins for Kubernetes Deployment#

1. Install the Kubernetes plugin in Jenkins.
2. Use this Jenkinsfile:

pipeline {
  agent any
  stages {
    stage('Checkout') {
      steps {
        git 'https://github.com/myrepo.git'
      }
    }
    stage('Deploy to K3s') {
      steps {
        sh 'kubectl apply -f k8s/'
      }
    }
  }
}

Now Jenkins will deploy changes whenever the pipeline runs.

Automating Deployments using Helm and Kustomize#

Using Helm for Package Management#

Helm makes Kubernetes deployments repeatable and maintainable.

Installing Helm#

curl https://raw.githubusercontent.com/helm/helm/main/scripts/get-helm-3 | bash

Deploying an application using Helm#

helm repo add myrepo https://charts.example.com/
helm install myapp myrepo/mychart

Using Kustomize for Kubernetes Configuration Management#

Kustomize lets you manage configurations without duplicate YAML files.

Writing a kustomization.yaml file#

resources:
  - deployment.yaml
  - service.yaml
patches:
  - target:
      kind: Deployment
      name: myapp
    patch: |-
      - op: replace
        path: /spec/replicas
        value: 5

Applying Kustomize#

kubectl apply -k .

Implementing Rollback and Version Control#

Mistakes happen—let’s make sure we can roll back quickly.

Rolling back a deployment in Kubernetes#

kubectl rollout undo deployment myapp-deployment

Managing versions in GitOps workflows#

• Use branches for testing and approval before merging to production.
• Maintain tags and release versions to track deployments.

Best practices for handling failed deployments#

• Always enable rollback policies in CI/CD pipelines.
• Implement health checks to detect issues early.
• Use progressive rollouts instead of deploying everything at once.

Hands-On Exercise#

Time to put all this into action:

• Deploy an application using ArgoCD and sync it with Git.
• Set up GitHub Actions or Jenkins for CI/CD with K3s.
• Package an application using Helm and deploy it to K3s.
• Implement Kustomize for managing Kubernetes configurations.

Master this, and your deployments will be faster, safer, and way less stressful. 🚀