CI/CD

Status

Planned

Goal

Build a delivery pipeline that turns a version-controlled change into a tested, traceable, and reversible deployment.

What I want to practise

• repository and branching conventions
• automated linting and tests
• build artifacts and container images
• self-hosted versus managed runners
• secret handling
• environment promotion and approvals
• deployment health checks
• rollback after a failed release

Intended outcome

A small application should move from commit to a lab environment without manual file copying. The pipeline should make failure visible, preserve useful logs, and provide a documented recovery path.

Questions to answer

• Which work belongs in CI and which belongs in deployment automation?
• How should credentials reach a runner without entering the repository?
• What evidence proves a deployment is healthy?
• How quickly can the previous working version be restored?