Fix release API proxy routes + wire pipeline to real data
- Add nginx proxy blocks for /api/v1/release-orchestrator/, /api/v1/release-control/, /api/v2/releases/, /api/v1/releases/, /api/v1/registries/ in Dockerfile.console - All release UI calls now reach JobEngine (401 not 404) - Registry search reaches Scanner service - Pipeline page uses ReleaseManagementStore (real API, no mock data) - Deployment wizard uses BundleOrganizerApi for create/seal - Inline version/hotfix creation in deployment wizard wired to API - Version detail shows "not found" error instead of blank screen - Version wizard has promotion lane + duplicate component detection - Sprint plan for 41 missing backend endpoints created Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
This commit is contained in:
master
@@ -0,0 +1,237 @@
|
||||
# Sprint 20260323-001 — Release API Proxy Fix + Missing Backend Endpoints
|
||||
|
||||
## Topic & Scope
|
||||
- Fix nginx proxy routes so the UI's `/api/v2/releases` calls reach the JobEngine backend.
|
||||
- Implement 41 missing backend endpoints across 5 service areas.
|
||||
- Working directories: `devops/docker/`, `src/JobEngine/`, `src/Platform/`, `src/EvidenceLocker/`
|
||||
- Expected evidence: all release CRUD flows work end-to-end via Playwright.
|
||||
|
||||
## Dependencies & Concurrency
|
||||
- TASK-001 (proxy fix) is prerequisite for all UI testing.
|
||||
- TASK-002 through TASK-006 can run in parallel after TASK-001.
|
||||
|
||||
## Documentation Prerequisites
|
||||
- `docs/modules/release-orchestrator/deployment/overview.md`
|
||||
- `docs/modules/release-orchestrator/deployment/strategies.md`
|
||||
- `src/JobEngine/StellaOps.JobEngine/StellaOps.JobEngine.WebService/Endpoints/ReleaseEndpoints.cs`
|
||||
- `src/JobEngine/StellaOps.JobEngine/StellaOps.JobEngine.WebService/Endpoints/ApprovalEndpoints.cs`
|
||||
|
||||
## Delivery Tracker
|
||||
|
||||
### TASK-001 - Fix nginx proxy routes for release APIs
|
||||
Status: TODO
|
||||
Dependency: none
|
||||
Owners: DevOps / FE
|
||||
|
||||
Task description:
|
||||
The console nginx (`devops/docker/Dockerfile.console`) reverse-proxies API calls. The UI calls:
|
||||
- `/api/v2/releases/*` → should reach JobEngine at `/api/v1/release-orchestrator/releases/*`
|
||||
- `/api/v1/release-control/bundles/*` → should reach JobEngine bundle endpoints
|
||||
- `/api/v1/registries/images/search` → should reach Scanner service
|
||||
- `/api/v2/releases/approvals` → should reach JobEngine approval endpoints
|
||||
- `/api/v2/releases/activity` → should reach JobEngine release events
|
||||
- `/api/v2/releases/runs/*` → should reach JobEngine run workbench
|
||||
|
||||
Options:
|
||||
A) Add nginx `location` blocks in `Dockerfile.console` to proxy these paths to the correct upstream services
|
||||
B) Add proxy routes in the Platform service (which already proxies many paths)
|
||||
C) Update the UI API clients to call the correct backend URLs directly (e.g., change `/api/v2/releases` to `/api/v1/release-orchestrator/releases`)
|
||||
|
||||
Recommended: Option C (most reliable, no proxy chain) + Option A as fallback.
|
||||
|
||||
Implementation for Option C:
|
||||
- Update `src/Web/StellaOps.Web/src/app/core/api/release-management.client.ts` — change base URLs
|
||||
- Update `release.store.ts` — ensure `loadReleases` calls the correct endpoint
|
||||
- Update `release-detail.component.ts` — fix all `/api/v2/releases/` calls to `/api/v1/release-orchestrator/releases/`
|
||||
- Update approval client — fix `/api/v2/releases/approvals` to `/api/v1/release-orchestrator/approvals`
|
||||
|
||||
For nginx (Option A fallback), add to `Dockerfile.console`:
|
||||
```nginx
|
||||
location /api/v2/releases/ {
|
||||
set $orchestrator_upstream http://orchestrator.stella-ops.local;
|
||||
rewrite ^/api/v2/releases/(.*) /api/v1/release-orchestrator/releases/$1 break;
|
||||
proxy_pass $orchestrator_upstream;
|
||||
}
|
||||
|
||||
location /api/v1/release-control/ {
|
||||
set $orchestrator_upstream http://orchestrator.stella-ops.local;
|
||||
rewrite ^/api/v1/release-control/(.*) /api/v1/release-control/$1 break;
|
||||
proxy_pass $orchestrator_upstream;
|
||||
}
|
||||
|
||||
location /api/v1/registries/ {
|
||||
set $scanner_upstream http://scanner.stella-ops.local;
|
||||
rewrite ^/api/v1/registries/(.*) /api/v1/registries/$1 break;
|
||||
proxy_pass $scanner_upstream;
|
||||
}
|
||||
```
|
||||
|
||||
Completion criteria:
|
||||
- [ ] `/api/v2/releases` returns data from JobEngine (not 404)
|
||||
- [ ] `/api/v1/registries/images/search?q=nginx` returns results from Scanner
|
||||
- [ ] `/api/v1/release-control/bundles` POST creates a bundle in JobEngine
|
||||
- [ ] UI pipeline page shows real releases (if any exist)
|
||||
- [ ] UI version create wizard can search registry images
|
||||
|
||||
### TASK-002 - Deployment monitoring endpoints (11 endpoints)
|
||||
Status: TODO
|
||||
Dependency: TASK-001
|
||||
Owners: BE (JobEngine)
|
||||
|
||||
Task description:
|
||||
Add deployment monitoring endpoints to JobEngine's `ReleaseEndpoints.cs` or new `DeploymentEndpoints.cs`:
|
||||
|
||||
```
|
||||
GET /api/v1/release-orchestrator/deployments — list deployments
|
||||
GET /api/v1/release-orchestrator/deployments/{id} — get deployment detail
|
||||
GET /api/v1/release-orchestrator/deployments/{id}/logs — get deployment logs
|
||||
GET /api/v1/release-orchestrator/deployments/{id}/targets/{targetId}/logs — target logs
|
||||
GET /api/v1/release-orchestrator/deployments/{id}/events — deployment events
|
||||
GET /api/v1/release-orchestrator/deployments/{id}/metrics — deployment metrics
|
||||
POST /api/v1/release-orchestrator/deployments/{id}/pause — pause deployment
|
||||
POST /api/v1/release-orchestrator/deployments/{id}/resume — resume deployment
|
||||
POST /api/v1/release-orchestrator/deployments/{id}/cancel — cancel deployment
|
||||
POST /api/v1/release-orchestrator/deployments/{id}/rollback — rollback deployment
|
||||
POST /api/v1/release-orchestrator/deployments/{id}/targets/{targetId}/retry — retry target
|
||||
```
|
||||
|
||||
Each endpoint needs:
|
||||
- Request/response models in contracts
|
||||
- Handler method with auth policy (ReleaseRead for GET, ReleaseWrite for POST)
|
||||
- Minimal implementation (can return mock/empty data initially, wired to real service later)
|
||||
|
||||
Completion criteria:
|
||||
- [ ] All 11 endpoints return 200/201 (not 404)
|
||||
- [ ] GET /deployments returns a list (even if empty)
|
||||
- [ ] POST /pause returns success
|
||||
- [ ] Auth policies enforced
|
||||
|
||||
### TASK-003 - Evidence management endpoints (6 endpoints)
|
||||
Status: TODO
|
||||
Dependency: TASK-001
|
||||
Owners: BE (EvidenceLocker or Attestor)
|
||||
|
||||
Task description:
|
||||
Add evidence lifecycle endpoints — either in EvidenceLocker service or as a new endpoints file in JobEngine:
|
||||
|
||||
```
|
||||
GET /api/v1/release-orchestrator/evidence — list evidence packets
|
||||
GET /api/v1/release-orchestrator/evidence/{id} — get evidence detail
|
||||
POST /api/v1/release-orchestrator/evidence/{id}/verify — verify evidence integrity
|
||||
GET /api/v1/release-orchestrator/evidence/{id}/export — export evidence bundle
|
||||
GET /api/v1/release-orchestrator/evidence/{id}/raw — download raw evidence
|
||||
GET /api/v1/release-orchestrator/evidence/{id}/timeline — evidence timeline
|
||||
```
|
||||
|
||||
Completion criteria:
|
||||
- [ ] All 6 endpoints return valid responses
|
||||
- [ ] Verify endpoint checks hash integrity
|
||||
- [ ] Export returns downloadable content-type
|
||||
|
||||
### TASK-004 - Environment/Target management endpoints (15 endpoints)
|
||||
Status: TODO
|
||||
Dependency: TASK-001
|
||||
Owners: BE (Platform)
|
||||
|
||||
Task description:
|
||||
Add environment and target management endpoints to Platform service. These may already partially exist in the topology/environment subsystem — check before implementing.
|
||||
|
||||
```
|
||||
GET /api/v1/release-orchestrator/environments — list environments
|
||||
GET /api/v1/release-orchestrator/environments/{id} — get environment
|
||||
POST /api/v1/release-orchestrator/environments — create environment
|
||||
PUT /api/v1/release-orchestrator/environments/{id} — update environment
|
||||
DELETE /api/v1/release-orchestrator/environments/{id} — delete environment
|
||||
PUT /api/v1/release-orchestrator/environments/{id}/settings — update settings
|
||||
GET /api/v1/release-orchestrator/environments/{id}/targets — list targets
|
||||
POST /api/v1/release-orchestrator/environments/{id}/targets — add target
|
||||
PUT /api/v1/release-orchestrator/environments/{id}/targets/{tid} — update target
|
||||
DELETE /api/v1/release-orchestrator/environments/{id}/targets/{tid} — remove target
|
||||
POST /api/v1/release-orchestrator/environments/{id}/targets/{tid}/health-check — check health
|
||||
GET /api/v1/release-orchestrator/environments/{id}/freeze-windows — list freeze windows
|
||||
POST /api/v1/release-orchestrator/environments/{id}/freeze-windows — create freeze window
|
||||
PUT /api/v1/release-orchestrator/environments/{id}/freeze-windows/{wid} — update
|
||||
DELETE /api/v1/release-orchestrator/environments/{id}/freeze-windows/{wid} — delete
|
||||
```
|
||||
|
||||
NOTE: The Platform service already has environment data via PlatformContextStore. These endpoints may be aliases or extensions of existing topology endpoints. Check `src/Platform/` for existing environment CRUD before implementing new ones.
|
||||
|
||||
Completion criteria:
|
||||
- [ ] Environment CRUD works
|
||||
- [ ] Target CRUD within environments works
|
||||
- [ ] Health check returns target status
|
||||
- [ ] Freeze window CRUD works
|
||||
|
||||
### TASK-005 - Release dashboard endpoint (3 endpoints)
|
||||
Status: TODO
|
||||
Dependency: TASK-001
|
||||
Owners: BE (JobEngine)
|
||||
|
||||
Task description:
|
||||
Add dashboard summary and promotion decision endpoints:
|
||||
|
||||
```
|
||||
GET /api/v1/release-orchestrator/dashboard — aggregated dashboard data
|
||||
POST /api/v1/release-orchestrator/promotions/{id}/approve — approve promotion
|
||||
POST /api/v1/release-orchestrator/promotions/{id}/reject — reject promotion
|
||||
```
|
||||
|
||||
Dashboard endpoint should return: release counts by status, deployment health, pending approvals, gate summary.
|
||||
|
||||
Completion criteria:
|
||||
- [ ] Dashboard returns aggregated stats
|
||||
- [ ] Approve/reject promotion updates approval status
|
||||
|
||||
### TASK-006 - Registry image search in Scanner service
|
||||
Status: TODO
|
||||
Dependency: none
|
||||
Owners: BE (Scanner)
|
||||
|
||||
Task description:
|
||||
Verify/implement registry image search endpoints in the Scanner service:
|
||||
|
||||
```
|
||||
GET /api/v1/registries/images/search?q={query} — search images by name
|
||||
GET /api/v1/registries/images/digests?repository={repo} — get image digests
|
||||
```
|
||||
|
||||
These are called by Create Version and Create Hotfix wizards. If the Scanner service doesn't have these endpoints, add them. The response format expected by the UI:
|
||||
|
||||
```json
|
||||
// Search response
|
||||
[
|
||||
{
|
||||
"name": "nginx",
|
||||
"repository": "library/nginx",
|
||||
"tags": ["latest", "1.25", "1.25-alpine"],
|
||||
"digests": [
|
||||
{ "tag": "latest", "digest": "sha256:abc123...", "pushedAt": "2026-03-20T10:00:00Z" }
|
||||
],
|
||||
"lastPushed": "2026-03-20T10:00:00Z"
|
||||
}
|
||||
]
|
||||
```
|
||||
|
||||
If no real registry is connected, return mock data so the wizard flow can be tested.
|
||||
|
||||
Completion criteria:
|
||||
- [ ] Search returns image results for common queries (nginx, redis, postgres)
|
||||
- [ ] Digests returns tags + SHA digests for a given repository
|
||||
- [ ] Create Version wizard can search and select images
|
||||
|
||||
## Execution Log
|
||||
| Date (UTC) | Update | Owner |
|
||||
| --- | --- | --- |
|
||||
| 2026-03-23 | Sprint created. TASK-001 is the critical path. | Planning |
|
||||
|
||||
## Decisions & Risks
|
||||
- TASK-001 (proxy fix) is the blocker — without it, no UI endpoint reaches the backend.
|
||||
- Option C (fix UI API URLs) is preferred over nginx rewrites because it eliminates a proxy hop.
|
||||
- Environment endpoints (TASK-004) may overlap with existing Platform topology — investigate before duplicating.
|
||||
- Evidence endpoints (TASK-003) may belong in EvidenceLocker rather than JobEngine — architecture decision needed.
|
||||
- Registry search (TASK-006) needs a connected registry or mock data for testing.
|
||||
|
||||
## Next Checkpoints
|
||||
- After TASK-001: re-run Playwright E2E tests to verify data flows
|
||||
- After TASK-002 + TASK-006: version/hotfix create → list → view flow should work end-to-end
|
||||
- After all tasks: full CRUD across all release control pages
|
||||
Reference in New Issue
Block a user