Download to read offline
Continuous Deployment using Kubernetes
- 1. Continuous Deployment usingKubernetes by Arun Veeramani DevOps & AI - Product Engineer Cognizant Accelerator – Cognizant Product Division
- 2. Few JARGONS tobe familiar with Definition Tools popularly used Deployment • It is the process of putting a new application, or new version of an application, onto a prepared application server Containerization • It is the process of bundling up the application code along with the environment needed for it to execute the code • It is a one time Job Docker, Vagrant Provisioning • It is the process of creating the infrastructure or the servers themselves, usually called ‘Infrastructure as Code’ • It is a one time Job Terraform, CloudFormation Container Orchestration • It is the process of arranging and coordinating multiple containers across a cluster of servers as per the resource crunch • It is a one time Job to Install the application and are then managed automatically by the cluster Kubernetes, Docker Swarm Configuration Management • It is a process of installing and managing software on the existing servers • It is a one time job to install the application Chef, Puppet, Ansible, and SaltStack Package Manager • A package manager or package management system is a collection of software tools that automate the process of installing, upgrading, configuring, and removing computer programs for a computer's operating system in a consistent manner Helm, Ksonnet
- 3.
- 4. Infrastructure Provisioning •Terraform Application Provisioning & Configuration Management •Ansible •Chef •Puppet ApplicationDeployment & Orchestration •Ansible •Chef •Puppet Infrastructure Provisioning •Terraform Application Provisioning & Configuration Management using CONTAINERS & VMs •Docker •Vagrant + Packer Application Deployment & Orchestration •Kubernetes •Docker Swarm Note: • Vagrant: Tool for building VMs for Development Environment • Vagrant Box: Base Images for Vagrant • Packer: Tool for building Machine Images, can create Vagrant Boxes • Docker: Tool for building Containers for Applications & Environments • Terraform: Infrastructure as Code Why Docker & Kubernetes? - The Workflow
- 5. Kubernetes is beyondComparison Support for Microservices Framework – Service Discover & Load Balancing Horizontal Scaling – Automatically based on CPU Self-Healing – Restarting or Rescheduling Containers that don’t respond Batch Job Execution Secret and Configuration Management Continuous Delivery – Automated Rollouts and Rollbacks Ingress – Manage external access to the Services in the Cluster (API Gateway) Kubernetes is more of a Framework to build any Platform, rather than just a tool for Continuous Deployment
- 6. Building a CDpipeline for Kubernetes A typical flow for Deploying any new Application into Kubernetes involves • Package your application into a Docker Image • Upload the Docker Image to a Container Registry (Harbor) • Deploy your Application (YAML) to the Cluster • Expose your Application as a Service The problems that arises with the above flow includes • Deploying Large Scale Applications with Release Management • Managing Deployment strategy across Team Members working on the same Deployment • Managing Rollbacks of new and old Deployments • Maintaining different versions of YAMLs for the same Application The solution to such problems involves “Automated Management of the Deployment Workflow” • Using Package Manger like HELM to deploy Applications using Charts • Using Workflow Engine like ARGO to deploy Applications as a Sequence of tasks • Using Package Manager like KSONNET to deploy Applications as a JSON Manifests
- 7. Continuous Deployment usingHELM Install Helm Define a Helm Chart Package the Chart Deploy the Release Upgrade / Rollback / Delete a Release https://www.slideshare.net/AdnanAbdulhussein/kubernetes-cicd-with-helm
- 8. Continuous Deployment usingARGO CI/CD Install ArgoProject Specify a Git Repository as a SOT for defining Desired Application State (GitOps) Commit your Kubernetes Manifest using Helm Charts / YAML in your Git Repository Run a CI Pipeline to build and push you Image into your Image Repo using ARGO CI / Jenkins Run a CD Pipeline to deploy your Manifests into the Environment specified using ARGO CD https://argoproj.github.io/docs/argo-cd/readme.html