Build Your Own GitOps Operator for Kubernetes & Beyond
summary
Two concrete operators are built and proven: an Ansible Playbook operator for Docker Swarm running as a GitLab CI job every 15 minutes that clones the repo, decrypts secrets, renders manifests and deploys stacks and services (service pruning automatic, stack pruning manual), and a Helmfile cronjob operator for Kubernetes running in-cluster every 5 minutes that applies Helmfile releases with SOPS support. Alerting uses GitLab CI emails and Prometheus, both increase deployment stability and tenant delivery speed, and artifacts (Playbook and Helm chart) are published.
AI-written summary · about
chapters (14)
- 0:00Introduction
- 0:48About Mimacom
- 2:17Contents
- 3:05Principles
- 4:32File formats
- 5:27Features
- 7:37Story
- 9:16Technology
- 12:02Outcomes
- 13:07Project Context
- 15:12Pruning
- 16:37Conclusion
- 17:28Lessons Learned
- 19:04Announcement
related talks
45mGitops: The Kubernetes way
49mGitOps, a somewhat realistic scenario on Kubernetes with Flux
39mGitOps — How to Marry Git and Kubernetes for Deployment and Operations
47mGitOps - Using Git as your source of truth for build, deploy and observability
02h 47mA GitOps Architecture from Scratch: Gitlab, Ansible, Terraform, Kub and AWS
29mGitOps on all layers – maintain your Kubernetes cluster! — Hafid Haddouti
33mKubernetes As A Service with GitOps
34mGitOps: Git Based Application Deployment Patterns for Kubernetes