fix(p02): close verified Redis ACK and CSV replay acceptance [skip ci]
This commit was merged in pull request #3.
This commit is contained in:
1 parent
fd57a3518d
commit
48fb277079
11 files changed
+799
-25
No files matched your search
@@ -89,3 +89,35 @@ Source freeze for P02: Gitea/GitHub main `667a140c4e9e75382bfdcdbe1d3e9eb96f6e80
|
||||
- **R-03 health:** `server/index.js:1926-1957` returns HTTP 200 `status:online` when Redis LLEN fails, while surfacing `redisConnected:false`; the historical route is **liveness**, not dependency readiness. `app/api/health/route.ts:95-111` likewise reports web liveness/buffer and does not prove database/Redis write readiness. A new readiness endpoint or altered public response requires separate approval.
|
||||
|
||||
Do not auto-fix source based solely on tests. Runtime code is READ-ONLY under the current approval. Any proven R-01/R-02 risk requiring behavior change needs a new exact method/file approval before application changes. P03 remains independent and unstarted.
|
||||
|
||||
|
||||
## P02 Runtime Remediation Candidate — 2026-11-22 (NOT YET VERIFIED)
|
||||
|
||||
Authorized follow-up baseline: Gitea/GitHub `fd57a3518d14bcb8367ce9ab9c3ec306bcfc9898` (GitLab same tree). Local isolated branch `p02-runtime-remediation-20261122`, implementation commit `44eee8877e40d1683f44502c2a0ba3d437a03b88`. Scope: **server/index.js** (`pushEventsToRedisOrFallback`, `flushBuffer`), **server/socket/handlers.js** (sync_data/batch_sync ACK and claim rollback), **core/data/buffer.ts** (`flush`), **tests/phase02-sync.test.ts** and **tests/phase03-storage-buffer.test.ts** only, plus notes/001–006. All other source is unchanged.
|
||||
|
||||
**Candidate R-01 behavior:** when a configured Redis client returns an RPUSH error, `pushEventsToRedisOrFallback` retains RAM fallback but throws a sanitized `BUFFER_PERSIST_FAILED` instead of returning a durable-looking buffer count. Socket single/batch ACK returns `success:false`; best-effort Redis idempotency-key removal and bounded in-process negative-ACK marker handle ordinary retries. This is a **candidate** pending tests. When Redis is intentionally absent, historical RAM fallback is deliberately unchanged; it is not a universal durable-ACK guarantee. If Redis claim deletion fails and the server restarts, stale Redis idempotency claims could still misclassify a retry; a full crash-atomic claim/queue protocol is **not established** by this patch.
|
||||
|
||||
**Candidate R-02 behavior:** Port 4000 `flushBuffer` and Port 3000 `DataBufferEngine.flush` stream-parse existing CSV/CSV.gz partitions for same-client/date/event IDs before replaying any buffered Redis or RAM batch, dropping already-persisted IDs while retaining Redis queue until the write/check completes. Scan/read/parse exceptions fail closed before LTRIM. In-scope test adds a direct Port 4000 writer + Redis LTRIM-failure/recovery test and updates the existing web-buffer replay fixture. **Not proven globally exactly-once:** cross-process CSV writers outside the Redis flush lock and unknown archival/partial-corruption cases remain; scanning partitions on each flush also has real disk-I/O cost, requiring throughput evaluation before production promotion.
|
||||
|
||||
**Candidate acceptance:** one disposable Gitea CI with 24/24 suites, strict TS, lint, Linux/Windows clean npm CI and production build; direct Port4000 replay tests; exact allowlist/diff audit and three-remote tree parity. The Gitea MCP/host was unavailable on the implementation attempt, so this is **NOT PASSED** and the P02 phase remains **BLOCKED / NOT CLOSED**. No live deployment or runtime production mutation.
|
||||
|
||||
|
||||
## P02 Final Acceptance Verification — 2026-11-22 (current superseding evidence)
|
||||
|
||||
Baseline published diagnosis: `fd57a3518d14bcb8367ce9ab9c3ec306bcfc9898`. Runtime candidate `76cd972bdeea9ef07b4b38e9d8b5b5ed3312ca5b` passed Gitea CI #66 (24/24 Node regression suites, Node SDK, strict TypeScript, ESLint, Next production build, Windows SDK). New final narrow patch `bf525f0d4e5e4781de7fddd2f5e7d3d602a80532`: `server/socket/handlers.js` plus two previously scoped TS tests only; final disposable Gitea CI #69 dispatched, terminal result to be recorded before phase closure.
|
||||
|
||||
Confirmed exact root cause in candidate: `server/socket/handlers.js` stores failed Redis persistence in `idempotencyMap` at `-Date.now()` while `cleanupMaps` deletes entries when `now - timestamp > 300000`; after the 60000ms cleanup window a failed marker can disappear despite an uncommitted Redis idempotency claim, permitting false-positive `DUPLICATE_DROPPED`. Arithmetic reproduction before code patch: at +61000ms, cleanup deletes a marker although Redis claim TTL (300000ms) is not over. Final patch uses a separate positive-timestamp failed-persistence marker and preserves negative ACK while the outstanding claim remains. A mixed batch with a failed duplicate must not return overall success while distinct records persist.
|
||||
|
||||
R-02: `server/index.js:843-897,1043-1058` and `core/data/buffer.ts:652-711,855-872` replay-scan existing same-client/day CSV, preserving Redis queue until a successful write/dedupe; fault tests include direct Port 4000 post-CSV LTRIM outage, rotated compressed .csv.gz containing quoted multiline payload and unrelated-date corrupt partition; malformed matching-date CSV must reject and leave Redis queue intact, with recovery after disposable fixture repair. Streaming parsing bounds memory, **not total disk I/O**; scan latency grows with matching partition size. No universal exactly-once guarantee, no live throughput claim and no claim of finite hard disk-I/O cap.
|
||||
|
||||
Public R-03 health remains historical liveness. Intentionally Redis-free RAM fallback and its historical successful ACK stay unchanged. No app/UI/auth/API schema/SDK/Docker/production-data/deployment changes. Work only in a new clean temp worktree; original PC repository's staged 16 P01 files and a separate pre-existing dirty candidate worktree preserved. No production readiness claim until CI #69 and final audit.
|
||||
|
||||
|
||||
### P02 Final Focused Correction — CI #69 to #72
|
||||
|
||||
Gitea [CI #69](https://dev.ivib.cc/vibtools/DataStack/actions/runs/69) was run against `bf525f0d4e5e4781de7fddd2f5e7d3d602a80532`: the new Redis stale-claim test timed out because it globally replaced `Date.now` during live Socket.IO ACK scheduling. This is a **test-harness failure**, not a proven backend regression; other new CSV.gz, partial-corruption, mixed-new/durable and Redis-free regression cases passed. An additional source-grounded edge case was found: `batch_sync` previously failed to reject a batch containing a non-durable claim PLUS an already accepted duplicate when `processed=0, duplicates>0`. Under exact approved scope, the `server/socket/handlers.js` branch changed from `if (volatileDuplicate && processed > 0)` to `if (volatileDuplicate)`, ensuring false overall ACK is never sent for any such mixture. Test now uses local simulated TTL arithmetic and real-clock Socket.IO retries; no global clock manipulation. Corrected commit `acca8be01a31f8d3972564cbaec8fd3117a0b551` sent to [focused CI #72](https://dev.ivib.cc/vibtools/DataStack/actions/runs/72); still no P02 closure or production deployment until PASS.
|
||||
|
||||
|
||||
### P02 definitive CI acceptance (2026-11-22)
|
||||
|
||||
Source-verified P02 final correction is `acca8be01a31f8d3972564cbaec8fd3117a0b551`. [Gitea CI #72](https://dev.ivib.cc/vibtools/DataStack/actions/runs/72) terminal **SUCCESS** (Linux 24/24 TypeScript suites, explicit R-01/R-02 fault cases, strict tsc, ESLint, Windows Node SDK, Next.js standalone build). Earlier #69 was cancelled after a test-only global `Date.now` Socket.IO ACK timeout; fixed by removing global clock override and adding direct arithmetic proof without changing timers. #72 proves the corrected test passes. Final source-path audit: exactly 11 approved paths (3 runtime, 2 tests, 6 notes). No keys/secrets, auth/SDK/UI/public health, dependency pins, CI workflows, deployment or database migrations changed. ACK remains **best-effort in intentionally Redis-free RAM mode**; Redis-configured RPUSH failure must return negative ACK; CSV deduplication guarantees only tested recovery windows, not globally atomic exactly-once. Partition scans are memory-streamed but disk reads scale with matching partition size, and production throughput is NOT certified. Release/installation still requires separate authorization.
|
||||
Reference in new issue
Block a user