Skip to content

Orchestrator

Orchestrator is the execution engine in the Oryvin architecture.

Its job is to run controlled work against real systems: apply overlays, build, deploy, verify runtime health, and report success or failure.

Orchestrator does not define system truth. It executes against source-derived work and deployed services.

Role in the architecture

Orchestrator sits between source-derived work and runtime verification.

A practical flow is:

  1. consume a patch or prepared source-derived input
  2. apply the overlay
  3. build the target system
  4. deploy the result
  5. verify /health
  6. capture or reference evidence when something fails

That gives Oryvin a repeatable execution loop rather than an ad hoc collection of manual steps.

Design goals

Controlled execution

Orchestrator should run clear, inspectable steps rather than hide work behind opaque abstractions.

Fail-fast behavior

Build, deployment, and verification failures should stop the run and make the failure explicit.

Runtime verification

Deployment is not considered complete until the target system exposes the expected runtime identity through /health.

Evidence-aware operation

Orchestrator should work well with systems such as Looker and DropBoxMini when evidence or result artifacts are needed.

Relationship to other components

  • WEIC defines governed knowledge and source-backed system direction
  • Origin prepares source-derived material
  • DropBoxMini stores derived artifacts and results
  • Looker captures evidence from running systems
  • Infrastructure provides the controlled environment where execution happens