Skip to main content

View Markdown · Documentation index (llms.txt)

Learning Guide — Containerized Compose Runtime (NGINX Ingress): purpose, walkthrough, and related learning states.

Convergence Rationale (C0)

State 004-containerized-compose-runtime is the C0 convergence baseline.

Rationale:

  • It is the first containerized, reproducible runtime that all later tracks can inherit from.

  • It preserves the baseline functional behavior while introducing a stable deployment substrate.

  • It provides the cleanest handoff point for architecture, functional, and platform deltas.

  • As the convergence baseline, it establishes the C0+ CI policy: convergence states publish container images via build/publish workflow plus GHCR run-bundle artifacts.

  • It keeps generated patchsets aligned with current dependency security targets so descendant convergence states inherit a clean runtime baseline.

  • Issue #465 upgrades the inherited Java baseline to public Spring Boot 4.0.8, Spring Framework 7.0.9, Data JPA 4.0.7, and Tomcat 11.0.26. Jackson 3 and Springdoc 3 compatibility changes preserve the existing HTTP and messaging contracts without adding security suppressions.

  • Issue #466 makes the C0 deployment boundary require exact-snapshot build/test/security evidence and scanned image digests. Descendant demo states inherit these checks through canonical generators while retaining one snapshot commit per generated branch.

  • Generated CI workflows come from the canonical CI asset installer after state overlays; historical workflow patches in states 004–007 are omitted so dependency remediation and deployment hardening compose without restoring stale checks.

Learning evidence at this checkpoint​

At C0, the learning contract combines the inherited edge and agent-harness knowledge with application containerization and the Compose runtime. The learner traces backend and frontend Dockerfiles into images and running containers before examining how Compose connects them. A learning partner asks the learner to distinguish container startup, service readiness and the retained trade result, using separate observations for each boundary. The canonical learning-contract.json supplies objective mappings, diagnostic questions, graduated hints and assessment evidence for both human and agent-assisted study.