3. Starting a Jenkins Service in MicroK8s

Jenkins is an open-source automation server that helps streamline CI/CD workflows by automating the build, test, and deployment processes. Deploying Jenkins on MicroK8s provides a lightweight, scalable, and containerized CI/CD environment. This guide walks you through the step-by-step installation and configuration of Jenkins on MicroK8s.

Why Jenkins?

Jenkins is widely used for:

• Automating repetitive tasks in software development.

• Integrating with various tools and plugins for seamless DevOps workflows.

• Managing deployments efficiently in containerized environments like Kubernetes.

Configuring MicroK8s on the Node Server

Step 1: Enable Required MicroK8s Add-ons

Before installing Jenkins, enable necessary MicroK8s services:

microk8s enable hostpath-storage dns

Set up Helm alias for easier package management. So that instead of using “microk8s.helm” commands everytime this will be reduced to only “helm”:

sudo snap alias microk8s.helm helm

Step 2: Add and Update Jenkins Helm Repository

Helm simplifies application deployment in Kubernetes. Add the Jenkins Helm repository and update it:

helm repo add jenkins https://charts.jenkins.io
helm repo update

Step 3: Create Jenkins Namespace

Create a dedicated namespace for Jenkins:

kubectl create namespace jenkins

Install Jenkins using Helm:

helm install jenkins jenkins/jenkins --namespace jenkins

Step 4: Retrieve Admin Password

To access Jenkins, check if the respective pods are up and running with (It might take atleast around 2minutes and 30 seconds for them to start running):

kubectl get pods --namespace jenkins

Ensure that all the pods are in running state. If not wait for a few minutes and try again. A typical debug method is to describe the pod that is in pending state to check why it is in pending state. Since you would typically find the jenkins pod name to be jenkins-0, (For helm installations) here’s how you would use the describe command:

kubectl describe pods jenkins-0 --namespace jenkins

Then once the pods are up and running we can extract the admin password (for the username “admin“) using:

kubectl exec --namespace jenkins -it svc/jenkins -c jenkins -- /bin/cat /run/secrets/additional/chart-admin-password && echo

Step 5: Change Service Type to NodePort

By default, Jenkins uses ClusterIP, making it accessible only within the cluster. To expose it externally, modify the service type:

If vim editor’s default commands are not working then export the default editor to nano and then try to edit the files you want using nano commands:

export EDITOR=nano

then, to edit the jenkins kubernetes service run:

kubectl edit svc jenkins --namespace jenkins

Change this section:

spec:
  type: NodePort

Alternatively, although this is the best practice, you can specify NodePort during installation:

helm install jenkins jenkins/jenkins --namespace jenkins --set controller.service.type=NodePort

Step 6: Configure Firewall

Allow traffic to the Jenkins port:

sudo ufw enable
sudo ufw allow 8080
sudo ufw status

You can also first check if the firewall is even active using:

sudo ufw status

Alternatively disable the firewall if you are in testing environment only.

sudo ufw disable

Step 7: Access Jenkins

Find your host machine’s IP. This should work when you have configured a static IP only (Typically done using router’s local host web sites to make sure your devices are discoverable with a single IP every time):

hostname -I

To ensure that jenkins is up and running, you can port forward and access it via localhost. Forward the Jenkins port:

kubectl --namespace jenkins port-forward svc/jenkins 8080:8080

Step 8: Initializing Jenkins Setup

Access Jenkins in your browser:

http://YOUR_STATIC_IP:31822

Follow the on-screen instructions to complete the setup.

Conclusion

With Jenkins running on MicroK8s, you now have a powerful CI/CD pipeline setup in a lightweight Kubernetes environment. You can further enhance it by integrating plugins for source control, automated testing, and deployment strategies.