GitOps: Git based application deployment patterns for Kubernetes

GitOps practices for Kubernetes. Presented at Chennai Kubernetes Meetup

Transcript

1. Git based application deployment patterns for Kubernetes Git ❤ K8s
2. Shahidh K Muhammed @shahidh_k
3. Git • Distributed version control system • Manage source code • Any text files can be tracked • Collaboration
4. Git - Source of Truth • Infrastructure as Code ◦ Terraform, CloudFormation • System Configuration as Code ◦ Ansible ◦ Chef ◦ Puppet • Application Configuration as Code ◦ Kubernetes ◦ Hasura • Application Source Code
5. Ops • Configure, build, test, deploy • CI/CD • Monitor, fix • Upgrades • Rollbacks • Automation
6. GitOps • Stable surface (tooling abstraction) for developers • Things change under this surface • Git as the boundary • Everything should be declarative • Driving DevOps through Git • Reduced toolset, Complete extensibility • Everything as Code - Kuberentes
7. Approach • Stable surface (tooling abstraction) for developers • Without restricting the ability of the operators to interface with infrastructure or to enforce controls and release processes. • "Below" this stable surface operators change plumbing • "Above" this surface developers use consistent tooling
8. Git as the boundary • For a developer, Git is the perfect tool for creating a "stable" boundary. • Git-hooks on the client-side/server-side • Webhook integrations offered by centralised git-hosting providers (github, bitbucket etc) • Constraint: ◦ git-hooks and webhooks can only make use of information inside the git repository ◦ or the environment that the git-hook/webhook runs in • Key constraint: Everything must become declarative.
9. Everything must become declarative Declarative tools ❤ Git
10. Git push to deploy • Comfortable and known workflow for developers • Given everything is declarative, Git is the only tool a developer need to drive DevOps tasks
11. GitOps - particular way of driving DevOps with Git
12. GitOps - Benefits • Reduced toolset ◦ Developers only need to use git for all DevOps tasks • Complete extensibility ◦ git-hooks and webhooks allow for infinite customisation and tooling to be implemented by operators without affecting the developer workflow in the slightest
13. GitOps - a checklist Developers • Declarative microservice runtime: Dockerfile • Declarative environment configuration: Kubernetes PodSpec • Declarative stateful actions: Jobs as containers, migration files
14. GitOps - a checklist Operators • Programmable git-remote/webhook listener on the cluster • Declarative infrastructure requirements: CloudFormation, Terraform • Declarative deployment: Kubernetes DeploymentSpec • Declarative tests - run as jobs in a pipeline triggered by a webhook or git-hooks • Implement CI/CD tasks automation in git-hooks • Implement deployment of stateful tasks like database migrations • Monorepo or a config repo - configuration, integration tests for the application are stored • Store secrets/tokens as references to actual values on the Kubernetes cluster, or a vault
15. Everything as Code
16. GitOps on Kubernetes • Setting up a git-remote that can receive git events or webhooks: Gitkube.sh: git push to your kubernetes cluster to build and deploy docker images • Synchronising state in git repo with state on a kubernetes cluster: Flux: A controller running on a kubernetes cluster that synchronises the declarative spec in the git repo with the state of the kubernetes cluster • Setting up git-hooks for running stateful migrations: Hasura: A CLI that sets up git-hooks on the local and remote git that can apply configuration, database migrations and can build/deploy docker images via git events.
18. Gitkube - Architecture
19. Demo • Gitkube • GitOps at Hasura: ◦ Deploying applications ◦ Changing subdomain ◦ Add a new domain ◦ Database migrations
20. Shahidh K Muhammed @shahidh_k Thanks for listening! Questions?