NextJs has been all the hype as of late - Most folks set up environment-specific pipelines as well as store K8s Configuration under parent environment folders, which leads to bloated code processes that do not conform to best practices.

A repository, Helm Charts, Kubernetes cluster and multi-branch pipelines are an excellent way to tackle this problem statement - allowing you to work with a minified approach to fast-paced deployment strategies.

• Jenkins: This OSS CI/CD tool helps reducing dependence on a cloud provider's Build and Release tools (eg: AWS CodeDeploy, Azure DevOps), and lets you migrate providers if needed with very less overhead work required.
• Node: (LTS), NextJS
• Docker: Image Build tool.
• Helm: This templating tool helps managing releases, and negates the need for Environment-specific folders with Kubernetes configuration.
• A Kubernetes cluster: Feel free to use kind/minikube to provision one locally, or use provision one on the respective cloud provider.

• Cluster NA/NS Creation
  • Setup Namespaces and Service accounts for your environments.
  • Attach correct RBAC policies to the service accounts
• EFK setup (Logging and Monitoring)
  • Create n Persistent Volumes for Elasticsearch to use as part of it's stateful volume claim.
  • Apply the files under ./pipelines/Kubernetes/Normal/EFK/elastic to set up ES.
  • Setup fluentd as a daemonset to forward container logs to elasticsearch.
  • Install Kibana to visualize logs - hit 'Discover' and add logstash* as an index.
• Ingress Nginx
  • Apply ./pipelines/Kubernetes/Normal/Ingress-Nginx/deploy-tls-termination.yaml if you're using AWS, otherwise install it via helm and make changes for your provider - you need to expose the service using an external loadbalancer.
  • Create ExternalNames for your frontend services.
  • Apply the ingress to take care of DNS/Routing to services in different namespaces.
• Jenkins
  • Create a multibranch pipeline with the repository details, and include the path to the JENKINSFILE.
  • Setup any other credentials that might be needed (eg: Artifactory/Dockerhub/ECR etc)
  • Wait for the branches to be registered post-scan, and trigger a build on the required branch.

• Pre and Post Deployment Checks
• Docker-compose file with mounts for easy local development (supports hot reloads)
• Include Dynamic Sonarqube runs/Quality Gate checks
• Impliment a SAST tool to scan Static code
• Use a SCA tool to scan OSS dependencies
• Include Image scanning for vulnerabilities
• Impliment a DAST tool to run on the deployed Front-end.