Commit 3057b30a authored by Dimitrios Gogos's avatar Dimitrios Gogos
Browse files

docs(platform): use the ADR-0061 adapter verbs in UC-OOP-06 and UC-OOP-07 

Scale calls update and reconciliation calls read_runtime_state, which
returns CapabilityRuntimeState. adapter.scale, query_status and
StatusResult no longer exist.

Co-Authored-By: default avatarClaude Opus 5.5 <noreply@anthropic.com>
parent 59eceb61

platform/architecture.md

+3 −3
Changes for platform/architecture.md: 3 added lines, 3 removed lines.
Original line number Diff line number Diff line
@@ -367,7 +367,7 @@ A running service instance needs to be scaled (more replicas) or terminated (res 
**Scale flow:**

1. OEG validates and publishes `command.srm.service.scale` with `service_instance_id` and `replicas`.

2. SRM looks up the `service_instance`, loads its `capability_instance`(s) and compute domain.

3. SRM calls `adapter.scale(external_id, replicas, compute_domain)`.

3. SRM calls `adapter.update` for each realization, with its `external_id` and the new replica count (ADR-0061).

4. TF-SDK translates to platform API (Helm upgrade, kubectl scale).

5. SRM updates the affected `capability_instance`(s).
@@ -391,9 +391,9 @@ SRM periodically checks whether its inventory matches actual infrastructure stat 


**Flow:**

1. SRM iterates over active `capability_instance`(s) per compute domain.

2. For each realization, SRM calls `adapter.query_status(external_id, compute_domain)`.

2. For each realization, SRM calls `adapter.read_runtime_state` with its `external_id` (ADR-0061).

3. TF-SDK queries the platform API (K8s pod status, Helm release status).

4. TF-SDK returns `StatusResult` (running, pending, failed, not_found).

4. TF-SDK returns `CapabilityRuntimeState` (`active`, `degraded`, `failed`, `deactivated`, `unknown`).

5. If SRM's inventory disagrees with actual state, SRM publishes `event.srm.inventory.drift_detected` (and `event.srm.inventory.drift_resolved` when it clears).



**Priority:** Should