nd064_C1

Cloud Native Fundamentals Scholarship Program Nanodegree Program

Course Homepage: https://sites.google.com/udacity.com/suse-cloud-native-foundations/home

Instructor: https://github.com/kgamanji

Student: https://www.linkedin.com/in/kelvin-tan-2086011a3/

This is the final project for the programme that I enrolled in March 2023.

At the time of this writing, I work as a devops engineer in the mobile games industry. I felt a need to upskill myself and came across a Facebook ad about a devops programme sponsored by SWIFT.

Project Steps

In this project, I forked the original repo from my instructor. The changes made were:

1. Best Practices For Application Deployment
2. Docker for Application Packaging
3. Continuous Integration with GitHub Actions
4. Kubernetes Declarative Manifests
5. Helm Charts
6. Continuous Delivery with ArgoCD
7. Standout Suggestions (Optional)
   1. Dynamic Healthcheck endpoint
   2. Push Docker images with custom tags
   3. Helm chart conditionals

Challenges Faced

I encountered some additional challenges outside of the project steps.

1. Vagrantfile private_network ip conflict
   Somehow the original 192.168.50.4 has a conflict on my Linux Mint, so I modified it to 192.168.56.4 in order for vagrant to build the vm.
2. Github action pytest: "Error: Version X with arch x64 not found"
   
   My pytest kept failing in github action.
   
   I fixed the issue by referring to [the list of available python versions in git action](https://raw.githubusercontent.com/actions/python-versions/main/versions-manifest.json)
   I found out the cause was in the dockerfile, ubuntu latest is used (which is 22.04), so I replaced python-version: [2.7, 3.5, 3.6, 3.7, 3.8] with ppython-version: ["2.7x", "3.7x", "3.8x"] because there is no 3.5 and 3.6 for 22.04 and 'x' basically means let the git action use any compatible minor python version.
   

Installation

# Clone this repo
git clone https://github.com/nopynospy/nd064_course_1.git

# cd into project directory
cd nd064_course_1/project

# Create vm
vagrant up

# Ssh into vm
vagrant ssh

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

# Wait for a few minutes for k3s to get started
sudo su
k3s kubectl get node

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

# Set up argocd
curl -o argocd-server-nodeport.yaml https://github.com/udacity/nd064_course_1/blob/main/solutions/argocd/argocd-server-nodeport.yaml
kubectl apply -f argocd-techtrends-staging.yaml

# Get encoded password
kubectl get secret argocd-initial-admin-secret -n argocd -o yaml

# Get password and login at https://192.168.56.4:30008/
# Login id is admin
# Decrypt password with the command below and replace encoded_password
echo encoded_password | base64 --decode

# Set up staging and prod applications
curl -o argocd-techtrends-staging.yaml https://github.com/nopynospy/nd064_course_1/blob/main/project/argocd/argocd-techtrends-staging.yaml
kubectl apply -f argocd-techtrends-staging.yaml
curl -o argocd-techtrends-prod.yaml https://github.com/nopynospy/nd064_course_1/blob/main/project/argocd/argocd-techtrends-prod.yaml
kubectl apply -f argocd-techtrends-prod.yaml

References

1. [How to structure, package, and release an application to a Kubernetes cluster, while using an automated CI/CD pipeline] (https://medium.com/@nisalikularatne/how-to-structure-package-and-release-an-application-to-a-kubernetes-cluster-while-using-an-b1f597289e93)