3.1 KiB
3.1 KiB
| phase | slug | status | nyquist_compliant | wave_0_complete | created |
|---|---|---|---|---|---|
| 6 | background-worker | draft | false | false | 2026-05-15 |
Phase 6 — Validation Strategy
Per-phase validation contract for feedback sampling during execution.
Test Infrastructure
| Property | Value |
|---|---|
| Framework | go test |
| Config file | backend/go.mod |
| Quick run command | cd backend && go build ./cmd/worker/... |
| Full suite command | cd backend && go test ./... && go build ./... |
| Estimated runtime | ~10 seconds |
Sampling Rate
- After every task commit: Run
cd backend && go build ./cmd/worker/... - After every plan wave: Run
cd backend && go test ./... && go build ./... - Before
/gsd-verify-work: Full suite must be green - Max feedback latency: 15 seconds
Per-Task Verification Map
| Task ID | Plan | Wave | Requirement | Threat Ref | Secure Behavior | Test Type | Automated Command | File Exists | Status |
|---|---|---|---|---|---|---|---|---|---|
| 06-01-01 | 01 | 1 | WORK-01 | — | N/A | build | cd backend && go build ./cmd/worker/... |
✅ | ⬜ pending |
| 06-01-02 | 01 | 1 | WORK-01 | — | N/A | unit | cd backend && go test ./internal/worker/... |
❌ W0 | ⬜ pending |
| 06-02-01 | 02 | 1 | WORK-02 | — | N/A | unit | cd backend && go test ./internal/jobs/... |
❌ W0 | ⬜ pending |
| 06-03-01 | 03 | 2 | WORK-03 | — | N/A | manual | Worker logs show retry backoff entries | — | ⬜ pending |
| 06-04-01 | 04 | 2 | WORK-04 | — | N/A | build | cd backend && go build ./... |
✅ | ⬜ pending |
Status: ⬜ pending · ✅ green · ❌ red · ⚠️ flaky
Wave 0 Requirements
backend/internal/worker/worker_test.go— stubs for WORK-01 worker startupbackend/internal/jobs/jobs_test.go— stubs for WORK-02 periodic job registration
If none: "Existing infrastructure covers all phase requirements."
Manual-Only Verifications
| Behavior | Requirement | Why Manual | Test Instructions |
|---|---|---|---|
| Heartbeat logs "worker heartbeat" every ~1 min | WORK-02 | Requires running binary with real Postgres | just worker and observe logs for 2 min |
| Orphan cleanup runs on schedule and logs results | WORK-02 | Requires real DB + S3 state | Insert orphan row, start worker, wait for cleanup run |
| Failed job retried with backoff visible in logs | WORK-03 | Requires real river runtime | Register a job that fails, observe retry log entries |
| Web binary enqueues job, worker picks up within seconds | WORK-04 | Integration between two running binaries | just dev + enqueue trigger, observe worker logs |
Validation Sign-Off
- All tasks have
<automated>verify or Wave 0 dependencies - Sampling continuity: no 3 consecutive tasks without automated verify
- Wave 0 covers all MISSING references
- No watch-mode flags
- Feedback latency < 15s
nyquist_compliant: trueset in frontmatter
Approval: pending