Skip to content
_CORE
AI & Agentic Systems Core Information Systems Cloud & Platform Engineering Data Platform & Integration Security & Compliance QA, Testing & Observability IoT, Automation & Robotics Mobile & Digital Banking & Finance Insurance Public Administration Defense & Security Healthcare Energy & Utilities Telco & Media Manufacturing Logistics & E-commerce Retail & Loyalty
References Technologies Blog Know-how Tools
About Collaboration Careers
CS EN DE
Let's talk

GitOps with Flux — Git as the Single Source of Truth for Infrastructure

09. 09. 2019 1 min read CORE SYSTEMSai
GitOps with Flux — Git as the Single Source of Truth for Infrastructure

Our workflow: Jenkins builds image → pushes to registry → operator runs helm upgrade → prays. GitOps flips the model: the Git repository describes the cluster state, Flux ensures the cluster matches. No imperative helm upgrade. Just git push.

Why GitOps

The problem: nobody knows what’s running on the cluster. Kubectl apply here, helm upgrade there, manual configmap edits. Configuration drift. The GitOps principle: what’s in git, runs. Always. Automatically. The agent reverts any manual change.

Bonuses: audit trail (git log), rollback = git revert, code review for infra changes (pull request), reproducibility (new cluster = git clone).

Flux vs. Argo CD

Argo CD: nicer UI, more features. Flux: lightweight, composable, better Helm integration. For 12 services with Helm charts, Flux won on simplicity.

Repository Structure

infrastructure/
├── base/           # Shared resources
├── environments/
│   ├── staging/    # Kustomize patches
│   └── production/
└── apps/
    ├── notification-service/
    ├── user-service/
    └── ...

Workflow in Practice

  1. Developer pushes code → Jenkins builds image myapp:1.23.5
  2. Flux detects new tag in registry
  3. Flux updates manifest in git (image tag)
  4. Flux applies change to cluster
  5. Rollback? git revert commit → Flux applies previous state
gitopsfluxkubernetesci/cd
Share:

CORE SYSTEMS

We build core systems and AI agents that keep operations running. 15 years of experience with enterprise IT.

Need help with implementation?

Our experts can help with design, implementation, and operations. From architecture to production.

Contact us