Local Netflix Application Deployment with ArgoCD on Kubernetes [Part Two]

Pre-requistes

• One Master (Controller)

• Two Worker Nodes

How to install Kubernetes locally check out my previous blog:

How to a Bootstrap Kubernetes Cluster

Install Node Exporter using Helm:

Efficiently monitor your Kubernetes cluster by deploying the Prometheus Node Exporter using Helm. Helm streamlines the process, providing a standardized and customizable approach to set up and manage this essential monitoring tool.

helm repo add prometheus-community https://prometheus-community.github.io/helm-charts

helm repo update

kubectl create namespace prometheus-node-exporter

helm install prometheus-node-exporter prometheus-community/prometheus-node-exporter --namespace=prometheus-node-exporter

kato@master1:~$ kubectl get pods -n prometheus-node-exporter
NAME                             READY   STATUS    RESTARTS   AGE
prometheus-node-exporter-ngggc   1/1     Running   0          84s
prometheus-node-exporter-qf9wt   1/1     Running   0          84s
prometheus-node-exporter-sdxl2   1/1     Running   0          84s

Add a Job to Scrape Metrics on nodeip:9001/metrics in prometheus.yml:
Remember we have a prometheus server running in our environment we used in part one

Install Argocd
ArgoCD is a declarative, GitOps continuous delivery tool for Kubernetes applications. It allows you to maintain and manage your Kubernetes applications using Git repositories as the source of truth for declarative infrastructure manifests

Installation steps :

helm repo add argo https://argoproj.github.io/argo-helm
helm repo update
kubectl create namespace argocd
helm install my-argo-cd argo/argo-cd --version 5.0.0 -n argocd
watch  kubectl get all -n argocd

Expose argocd-server

By default argocd-server is not publicaly exposed. For the purpose of this workshop, we will use a NodePort to make it usable:

kato@master1:~$ kubectl edit svc my-argo-cd-argocd-server -n argocd

For login purposes
username: admin and for password

kato@master1:~$ export ARGO_PWD=$(kubectl -n argocd get secret argocd-initial-admin-secret -o jsonpath="{.data.password}" | base64 -d)
kato@master1:~$ echo $ARGO_PWD
y9o7F4hWV-TDhtif
kato@master1:~$

To acces the argocd UI :

We shall access one of the nodeportIPs in this case we shall access

Deploy Netflix Application with ArgoCD

1. Configure ArgoCD Application:

   • In the ArgoCD web interface or using the ArgoCD CLI, create a new application.

     

     • Provide a name for your application.

     • Specify the source type as Git.

     • Set the repository URL to your Git repository's URL (e.g., https://github.com/Gatete-Bruno/Netflix-Local).

     • Choose the target namespace where your application will be deployed. in my case i chose default

       

       1. Set Application Path:

          • If your application is in a subdirectory within your Git repository, set the path field to the relative path of your application within the repository.

          • The path that contains the kubernetes yaml files is Kubernetes

       2. Configure Sync Policy:

          • Configure how often ArgoCD should check for updates and automatically sync your application. This is typically done in the "Sync Policy" section.

          • I used Manual option

       3. Define the Cluster:

          • Specify the cluster where you want to deploy your application. This is usually done in the "Cluster" section of the application configuration.

     • 

   • 4. Sync the Application:
        Trigger an initial sync of your application either through the ArgoCD web interface or using the CLI. This will deploy your application to the specified target cluster.

        

        The Application is accessible on one of the nodeIPs

        5. kato@master1:~$ kubectl get nodes -o wide
           NAME      STATUS   ROLES                  AGE     VERSION    INTERNAL-IP    EXTERNAL-IP   OS-IMAGE             KERNEL-VERSION      CONTAINER-RUNTIME
           master1   Ready    control-plane,master   3h10m   v1.23.10   192.168.1.60   <none>        Ubuntu 20.04.5 LTS   5.4.0-169-generic   docker://20.10.8
           node1     Ready    worker                 3h10m   v1.23.10   192.168.1.61   <none>        Ubuntu 20.04.5 LTS   5.4.0-169-generic   docker://20.10.8
           node2     Ready    worker                 3h10m   v1.23.10   192.168.1.62   <none>        Ubuntu 20.04.5 LTS   5.4.0-169-generic   docker://20.10.8
           kato@master1:~$