fix: address M2 verifier findings in soak harness
This commit is contained in:
@@ -0,0 +1,17 @@
|
||||
---
|
||||
type: finding
|
||||
node_id: finding:m2-verifier-fixes
|
||||
title: "M2 verifier 三项 Important 补齐(不变量接线/网关保护/P3 验收)"
|
||||
date: 2026-07-21
|
||||
---
|
||||
|
||||
# M2 verifier 三项 Important 补齐
|
||||
|
||||
独立 verifier(全新上下文)验收 M2:0 Critical,核心库逐条达标;3 Important + 4 Minor 全部当日修复。
|
||||
|
||||
| # | 问题 | 修复 | 证据 |
|
||||
|---|---|---|---|
|
||||
| I1 | 不变量 1(记账归零/gate 可再准入)未接线进 harness | `run_soak._live_checks` 跑后活检 + 新增 `inv_gate_reenterable`(探针悬挂检测,拿到即归还) | P3 验收报告两项 PASS |
|
||||
| I2 | 签字网关保护(100/600)无落点 | `_guard` 硬门:压测 scope 必配 `__GLOBAL__` 键且 ≤ 签字值;.env.example SOAK 段写明 | 缺键拒跑 |
|
||||
| I3 | P3 中等规模验收未跑 | 300 calls × 2 workers 真实跑:297 ok,7 不变量 PASS,结构化基线 0.942,p50/p95=16.9s/36.4s;真实网关自然出现 22 次空补全全被 Transient 重试吸收 | tests/outputs/soak/soak_20260721_015028.md |
|
||||
| M4-M7 | settle 置位偏离未声明/对标未打钩/场景帽/故障存在性断言 | docstring 声明;chsanalyzer.md 十项打钩;`capped_budget`;`inv_fault_errors_present` | 单测 16 例 |
|
||||
@@ -30,6 +30,11 @@
|
||||
"id": "plan:m2-distributed",
|
||||
"label": "M2 分布式实现计划",
|
||||
"type": "plan"
|
||||
},
|
||||
{
|
||||
"id": "finding:m2-verifier-fixes",
|
||||
"label": "M2 verifier 三项 Important 补齐(不变量接线/网关保护/P3 验收)",
|
||||
"type": "finding"
|
||||
}
|
||||
],
|
||||
"links": [
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
# Research Wiki 索引
|
||||
|
||||
> 自动生成，更新时间：2026-07-21 03:46 UTC
|
||||
> 自动生成，更新时间：2026-07-21 06:00 UTC
|
||||
|
||||
## design (4)
|
||||
- [2026-07-20-m1-core-design](designs/2026-07-20-m1-core-design.md) `design:2026-07-20-m1-core-design`
|
||||
@@ -8,8 +8,9 @@
|
||||
- [M1 核心里程碑设计:公共签名冻结与治理栈落地](designs/m1-core-design.md) `design:m1-core-design`
|
||||
- [M2 分布式:Redis 治理后端+背压+Postgres 遥测+pricing+Embedding+压测 harness](designs/m2-distributed.md) `design:m2-distributed`
|
||||
|
||||
## finding (2)
|
||||
## finding (3)
|
||||
- [2026-07-20-m2-soak-workload](findings/2026-07-20-m2-soak-workload.md) `finding:2026-07-20-m2-soak-workload`
|
||||
- [M2 verifier 三项 Important 补齐(不变量接线/网关保护/P3 验收)](findings/m2-verifier-fixes.md) `finding:m2-verifier-fixes`
|
||||
- [M2 真实数据压测: 场景矩阵与数据清单](findings/m2-soak-workload.md) `finding:m2-soak-workload`
|
||||
|
||||
## plan (4)
|
||||
|
||||
@@ -17,3 +17,5 @@
|
||||
- [2026-07-21 03:46 UTC] 新增 plan: M2 分布式实现计划 (plan:m2-distributed)
|
||||
- [2026-07-21 03:46 UTC] 新增边: plan:m2-distributed --implements--> design:m2-distributed
|
||||
- [2026-07-21 03:46 UTC] 重建索引: 12 篇页面
|
||||
- [2026-07-21 06:00 UTC] 新增 finding: M2 verifier 三项 Important 补齐(不变量接线/网关保护/P3 验收) (finding:m2-verifier-fixes)
|
||||
- [2026-07-21 06:00 UTC] 重建索引: 13 篇页面
|
||||
|
||||
@@ -11,20 +11,22 @@
|
||||
| M2 后 | VLM scope 治理核心(governance/limiter/scripts/provider_gate/selector/streaming + VLM invoker) | Redis 六道闸+契约测试、多源选源、跨进程熔断、背压 stall |
|
||||
| M3 后 | OCR scope(MonkeyOcrParseInvoker → `OcrLayoutPort`)、`core/eval/judge.py` 收编,**全量迁移** | OCR 端口族 + MonkeyOCR transport |
|
||||
|
||||
**能力对标清单**(库必须逐项对等,任一缺失即库的边界缺口):
|
||||
**能力对标清单**(库必须逐项对等,任一缺失即库的边界缺口;✅ = M2 验收打钩 2026-07-21,证据为库测试):
|
||||
|
||||
| # | 能力 | 项目侧证据 |
|
||||
|---|---|---|
|
||||
| 1 | 六道闸原子限流(全局/单源 × 并发/RPM/TPM),拒绝零副作用 | `app/coordination/scripts.py:6-34` |
|
||||
| 2 | TPM 预扣入场、settle 按实际 usage 落回 acquire 窗口多退少补 | `limiter.py:55-62`、`scripts.py:44-48` |
|
||||
| 3 | 并发 lease 带 TTL(防进程死亡泄漏)+ release/settle 幂等 | `limiter.py:46-63`、`tests/contracts_limiter.py:66-73` |
|
||||
| 4 | 窗口 id 用 Redis 服务器时钟(多进程口径统一) | `limiter.py:95-98` |
|
||||
| 5 | 跨进程熔断:单探针半开、探针租约 TTL、epoch fencing、force_open | `scripts.py:82-225`、`provider_gate.py` |
|
||||
| 6 | 背压 stall 双条件判卡死(本地等够 + 全局无进展)+ mark_progress 全局活性 | `governance.py:270-281`、`limiter.py:193-209` |
|
||||
| 7 | 换源重试:失败跨源累计、退避取 Retry-After 较大值、源冷却备忘 | `governance.py:105-118, 244-261` |
|
||||
| 8 | 错误分类:429 body 细分 insufficient_quota、工件级失败不熔断 | `invokers.py:144-166`、`governance.py:237-239` |
|
||||
| 9 | scope 级不可用结构化错误(reason 7 种 + retry_after_s),供 worker 延期重投 | `errors.py:140-180`、`workers/tracking.py:406-428` |
|
||||
| 10 | 流式三层活性看门狗;thinking token 刷活性不计结果 | `streaming.py`、`invokers.py:55-79` |
|
||||
| # | 能力 | 项目侧证据 | M2 打钩 |
|
||||
|---|---|---|---|
|
||||
| 1 | 六道闸原子限流(全局/单源 × 并发/RPM/TPM),拒绝零副作用 | `app/coordination/scripts.py:6-34` | ✅ `backends/redis/limiter.py` Lua 逐字移植;契约 redis 参数全绿 |
|
||||
| 2 | TPM 预扣入场、settle 按实际 usage 落回 acquire 窗口多退少补 | `limiter.py:55-62`、`scripts.py:44-48` | ✅ 契约 `test_prededuct_and_settle_refund[redis]` |
|
||||
| 3 | 并发 lease 带 TTL(防进程死亡泄漏)+ release/settle 幂等 | `limiter.py:46-63`、`tests/contracts_limiter.py:66-73` | ✅ 真实等待变体 `test_variant_lease_expiry_reclaims_slot` + 幂等契约 |
|
||||
| 4 | 窗口 id 用 Redis 服务器时钟(多进程口径统一) | `limiter.py:95-98` | ✅ `test_rpm_window_rollover_resets_quota` + 跨连接 RPM 用例 |
|
||||
| 5 | 跨进程熔断:单探针半开、探针租约 TTL、epoch fencing、force_open | `scripts.py:82-225`、`provider_gate.py` | ✅ `backends/redis/breaker.py`;熔断契约 + 7 个真实等待变体 |
|
||||
| 6 | 背压 stall 双条件判卡死(本地等够 + 全局无进展)+ mark_progress 全局活性 | `governance.py:270-281`、`limiter.py:193-209` | ✅ `middleware/retry.py` + `test_backpressure` 四象限 + 跨连接进度可见 |
|
||||
| 7 | 换源重试:失败跨源累计、退避取 Retry-After 较大值、源冷却备忘 | `governance.py:105-118, 244-261` | ✅ M1 已交付(`test_retry`);M2 联合验证下重验 |
|
||||
| 8 | 错误分类:429 body 细分 insufficient_quota、工件级失败不熔断 | `invokers.py:144-166`、`governance.py:237-239` | ✅ M1 已交付(`test_openai_compat`/ResultInvalid 不熔断) |
|
||||
| 9 | scope 级不可用结构化错误(reason + retry_after_s),供 worker 延期重投 | `errors.py:140-180`、`workers/tracking.py:406-428` | ✅ M1 错误模型(5 值 reason 勘误后)+ M2 增 stalled 真实触发路径 |
|
||||
| 10 | 流式三层活性看门狗;thinking token 刷活性不计结果 | `streaming.py`、`invokers.py:55-79` | ✅ M1 已交付(`test_streaming`) |
|
||||
|
||||
> 注: OCR invoker 能力(MonkeyOCR 两端点)不在本清单——归 M3。
|
||||
|
||||
## 2. 现状盘点(行数 wc -l 实测)
|
||||
|
||||
|
||||
Reference in New Issue
Block a user