**Evidence levels**: SOURCE = verified code present; CI PASS = focused automated tests/build passed; LIVE READ = public read-only endpoint observed; NOT VERIFIED = no independent safe live runtime evidence. These are deliberately different. Do NOT infer production DB/S3/MCP health from a public `online` response.
| Capability / feature | Exact representative path | Baseline evidence | Actual live conclusion | Changed by P01/P02/P03 |
|---|---|---|---|---|
| Next.js 15 public landing and app pages | `app/**`, `docker/Dockerfile.web` | Next.js CI build PASS and 22/22 static pages; public landing loads | LIVE READ for landing only | None |
| Compression and S3 archival | `core/Storage/s3Archive.ts`, `core/Storage/csvStorage.ts` | SOURCE + storage tests | No live S3 credentials or archive tests | None |
| Data browser, export CSV/JSON/NDJSON | `app/api/export/route.ts`, `app/api/data/route.ts` | SOURCE + TS tests | NOT live-tested | None |
| Storage quota / disk admission | `core/Storage/csvStorage.ts` | SOURCE + tests; 507 codes | NOT live-tested | None |
| Dashboard, client status, counts | `lib/vconUiHelpers.ts`, `app/vcon/clients/page.tsx` | SOURCE; client status helper tests; stale old regex CI-04 | No live user session proof | None |
| Prometheus and metrics bridge | `app/api/prometheus/route.ts`, `server/index.js` | SOURCE + tests | NOT live authenticated | None |
| Webhook alerts and SSRF/HMAC | `core/alerts/webhookNotifier.ts` | SOURCE + tests | NOT live-tested | None |
| CloudCmd file browser | `server/filesManager.js`, `tests/files-manager.test.ts` | SOURCE; files-manager suite 4 tests PASS | NOT live authenticated | None |
| MCP engine (17 permissions) | `server/mcp/mcpEngine.js`, `app/api/mcp/route.ts` | SOURCE + MCP server suite 3 tests PASS; write flags default false | NOT live connected/authenticated | None |
| Python SDK v2.2 + compat package | `sdk/python/datastack/datastack/**`, `sdk-python/**` | Windows CI **14/14** PASS | No real external SDK session | None |
| Node SDK v2.2 + offline spool | `sdk/node/datastack/**` | Native Windows SDK compile PASS; SDK-specific TS regression PASS | No live external SDK session | None |
| PWA/offline shell | `app/**`, `public/**` | SOURCE + phase06 suite PASS | Browser offline not live-tested | None |
| Docker web/proxy image | `docker/Dockerfile.web`, `docker/Dockerfile.proxy` | Web layer fail on containerd; proxy unrun | Existing Coolify instance appears online but CI image parity unverified | None |
| Health endpoint | `app/api/health/route.ts` | LIVE READ `status:online` on 2026-11-22T00:01:19Z | Confirmed public API liveness; not end-to-end readiness | None |
## Mandatory post-phase update convention
After EACH P01/P02/P03 close, amend only affected table rows:
-`Feature/capability` (exact added/changed behavior or `UNCHANGED`)
- P01: **CLOSED — 1/3 implementation phases**. CI-01…05 corrected exclusively in package-lock, obsolete bun.lock removal, scoped regression/type tests and Gitea CI harness. **No actual application, UI, backend, API, SDK or production functionality was added, removed or altered.** Gitea #59: committed-lock Linux/Windows clean install PASS; Node SDK builds PASS; all 24 TS suites exit0 with zero timeouts; Next standalone 22/22 pages PASS. One historical #59 test-only declaration TS error fixed; #60 strict TS zero errors and full ESLint PASS. Previous canonical Python SDK 14/14 (#48) preserved but not rerun. No live endpoint mutation or deployment; live data/Redis/S3/MCP actual operational status is not established.
- P02: **BLOCKED pending separate approved runtime remediation**. Three disposable fault-injection tests demonstrated existing behavior: positive ACK on Redis write failure with RAM-only data (R-01), duplicate appended event ID after Redis trim failure and retry in web-side buffer (R-02), and public health 200/online when Redis is down (R-03). These are **diagnostic tests, not production fixes**. Zero app/API/UI/SDK/runtime changes, no live deploy. CI #63. P01 remains closed (1/3 phases complete).
- P03: pending.
## P02 Diagnostic Feature Record — 2026-11-22
- **New production features: NONE.** This phase adds only three disposable failure-injection scenarios across two existing test suites; tests are not production handlers. Existing source/UI/auth/backend/SDK/MCP, Redis/PG/CSV services, Next/Express health interfaces and deployment remain unmodified.
- R-01: existing volatile RAM fallback and WS ACK behavior audited, *not corrected*. Do not mark ACK durability as working merely because a diagnostic test proves current behavior.
- R-02: existing CSV-before-Redis-trim order audited, *not corrected*. Exactly-once replay safety is **NOT VERIFIED**.
- R-03: current public Express health 200 plus `redisConnected:false` is liveness, not dependency readiness. Separate readiness remains **NOT IMPLEMENTED**.
- Gitea disposable CI #63: https://dev.ivib.cc/vibtools/DataStack/actions/runs/63. The previous P01 working results remain valid; no live read/write test or production deploy is included in P02.
- **Intended change:** failure cases on socket ingestion report `BUFFER_PERSIST_FAILED` instead of positive ACK when Redis RPUSH rejects; single and batch code paths tested by new disposable assertions. Ordinary success ACK, auth, limits and SDK public interfaces intended unchanged.
- **Intended change:** Redis replay flush on both ingestion/Next paths inspects existing client CSV partitions and filters already-stored event IDs before reappend; safe failure leaves Redis queue untrimmed. This prevents the tested post-append/LTRIM replay duplication but does **not** prove transactionally exactly-once delivery or zero performance impact.
- **No changes:** application UI/UX, auth, SDK, database schema, Docker, environment, public R-03 health, other features and production runtime. No deployment or live verification.
- **Status:** P02 code candidate locally committed, acceptance CI blocked by Gitea unavailability; **NOT CLOSED**, not production-ready. P01 remains only completed phase.
- **Changed in source, not deployed:** Port 4000 Socket.IO error ACK behavior on Redis write failure; no-Redis mode now waits for CSV flush before success; duplicate ACK checks actual Redis/CSV durability; mixed batch uncertainty gives negative ACK. EventId retry handling is made fail-closed with in-scope handler logic.
- **Changed in source, not deployed:** Port 4000 and web-buffer CSV replay after Redis LTRIM failure now checks persisted CSV/CSV.gz event IDs. Existing file rotation/compression and disk errors need focused tests; scanner I/O impacts high-volume performance and does not certify global exactly-once semantics.
- **Unchanged:** UI/UX/pages, public R-03 health shape/status, auth/JWT/SDK implementation, database/Prisma, Docker/Compose, environment, unrelated features, production data, live server and deployment.
- **Still NOT working as end-to-end guarantee:** HTTP `POST /api/events` currently may send success after Redis RPUSH fails while only RAM contains the event. This is an additional source-observed release blocker, not something fixed by WS error ACK. **P02 NOT CLOSED; P03 NOT STARTED.**
- **Evidence:** Gitea #66 initial remediation PASS; #68 exposed same-ID retry bug fixed in current branch; #70 final focused run awaiting terminal confirmation. The original PC main workspace's 16 staged P01 paths remain untouched.