fix(reliability): handle closed diagnostics and harden test fixtures by lanyue-llk · Pull Request #725 · lexmount/moli

lanyue-llk · 2026-09-22T04:03:49Z

启动器关闭 stderr 后，浏览器记录脚本错误会因日志组件再次写入损坏管道而中止进程。本 PR 保留正常错误日志，关闭日志组件对自身写入失败的二次输出；同时修复 CI 产物校验和跨平台测试夹具，为后续功能 PR 提供稳定的测试前提。

修复机制

stderr：使用 tracing 的标准配置处理损坏诊断输出，并用真实子进程分别验证正常日志与关闭管道。
SQLite：保留 Linux 非 UTF-8 文件名场景，在 macOS 验证非 ASCII 文件名；先建立正常数据库再移动文件，隔离数据库创建与快照路径处理。
布局夹具：按实际 glyph baseline 与 padding 位移验证基线关系；滚动条场景使用仓库内 Ahem 字体和不可断词，固定输入而保留最终宽度与文本几何断言。
网络夹具：等待真实 socket 接入、保留 libcurl 端点和请求语义断言，避免把调度时序当作功能失败。
CI：验证下载结果和归档有效性，再解包发布产物。

flowchart TD
A[页面脚本错误] --> B[tracing 记录错误]
B --> C{stderr 可写?}
C -->|是| D[保留正常错误日志]
C -->|否| E[禁用日志组件自身写入失败的二次输出]
D --> F[浏览器继续完成请求]
E --> F

图示说明本 PR 的 stderr 故障处理路径。

对应验收

本 PR 不认领历史 1.1.9 的 34 个失败 case，验收对象是诊断管道、CI 归档及测试夹具的功能契约。

验证项	结果
当前 HEAD / main 基线	`f13e9f42` / `72413280`
本地验证版本	`22e161c8`，基于 `de1093d9c`；rebase 前后功能补丁逐字节相同
cargo fmt --all	通过
cargo clippy --workspace --all-targets --all-features -- -D warnings	通过
cargo nextest run --no-fail-fast	18,416 通过、0 失败；13 项仓库既有跳过
CI 脚本测试	`node --test .github/scripts/*.test.cjs`，13/13 通过
真实 release CLI 红/绿对照	main 关闭 stderr 后 SIGABRT、无页面输出；修复后 exit 0、页面内容完整；正常 stderr 仍记录脚本错误
当前 HEAD 完整门禁	由本 PR CI 验证

源码、门禁与构建参数证据、rebase 补丁一致性、release 二进制哈希与红绿结果、CI 脚本测试。本地门禁关闭调试符号以控制磁盘占用，未关闭断言或减少测试。

复现诊断管道专项：

cargo nextest run -E 'test(cli_uncaught_script_error_)'

范围与依赖

从 #640 拆出，原 PR 与原分支保留。本 PR 直接基于 main，是其他六个功能分区的共同前置。布局夹具的调整只固定字体输入和验证真实基线关系；布局生产逻辑仍在独立布局 PR 中。

完整对应关系：拆分 PR 索引、逐 case 历史失败及归属。

github-actions · 2026-09-22T05:16:54Z

Sequential Navigation Soak A/B

✅ HEAD completed the 200-navigation resilience and memory observation.

One browser process, one target, and one CDP session navigate CSDN → SegmentFault → Huaban → example.com repeatedly. This run issued 200 Page.navigate commands per binary.

Common ancestor 7241328085 → HEAD f13e9f427e; benchmark order: base-first; workflow: success.

Session resilience

Metric	Base	HEAD
Attempted / planned	200 / 200	200 / 200
Direct observable pass	200 / 200	200 / 200
Failures / recovered	0 / 0	0 / 0
Recovery failures	0	0
Lifecycle order violations	0	0
Network order violations	0	0
Wall time	304.5 s	302.5 s

A public-page failure is reported separately from an unrecoverable session. The soak requires all navigation attempts, zero failed recovery, zero lifecycle/network ordering violations, and complete resource evidence.

Process-tree memory

Metric	Base	HEAD	Δ
Boundary samples	200	200	0
Boundary RSS samples	200	200	0
Boundary PSS samples	200	200	0
Periodic peak RSS	224.3 MiB	222.7 MiB	-1.7 MiB
Periodic peak PSS	221.1 MiB	219.8 MiB	-1.3 MiB
RSS · first 10 avg	168.2 MiB	166.6 MiB	-1.5 MiB
RSS · last 10 avg	135.0 MiB	145.8 MiB	+10.8 MiB
RSS · last − first window	-33.2 MiB	-20.8 MiB	+12.3 MiB
RSS · warm slope / 100 nav	-23.0 MiB	-13.0 MiB	+10.0 MiB
PSS · first 10 avg	165.0 MiB	162.8 MiB	-2.2 MiB
PSS · last 10 avg	132.7 MiB	143.0 MiB	+10.3 MiB
PSS · last − first window	-32.3 MiB	-19.8 MiB	+12.6 MiB
PSS · warm slope / 100 nav	-22.5 MiB	-13.4 MiB	+9.2 MiB
Peak file descriptors	63	63	0
Peak threads	31	31	0

HEAD memory by 50-navigation quarter

HEAD range	Avg RSS	Final RSS	Peak RSS	Avg PSS	Final PSS	Peak FD
1–50	165.6 MiB	149.3 MiB	218.5 MiB	162.7 MiB	146.2 MiB	53
51–100	154.1 MiB	122.8 MiB	211.4 MiB	150.8 MiB	119.1 MiB	60
101–150	146.2 MiB	129.1 MiB	193.9 MiB	142.5 MiB	125.4 MiB	53
151–200	148.1 MiB	114.5 MiB	200.3 MiB	145.1 MiB	111.8 MiB	53

Raw reports and per-navigation boundary samples: workflow run and full sequential-navigation-soak-results artifact.

Public-site timing and memory are observational. Use the A/B deltas and trend shape as evidence, not as a deterministic performance threshold.

github-actions · 2026-09-22T05:31:29Z

Spider Bench A/B

✅ All benchmark browser service runs completed; results are informational.

Public HEAD: 189 / 240 rows; 39 / 48 sites produced rows; 9 unexpected empty sites.

✅ Deterministic fixture contract is clean.

Common ancestor 7241328085c8 → HEAD f13e9f427ebd; benchmark order: base-first; workflow: success.

Public 48-site run · informational

Metric	Base	HEAD	Δ
Run status	success	success	—
Extracted rows / contract	184 / 240	189 / 240	+5 (+2.72%)
Contract fill	76.67%	78.75%	+2.08 pp (+2.71%)
Sites with rows	38 / 48	39 / 48	+1 (+2.63%)
Sites meeting row contract	36 / 48	37 / 48	+1 (+2.78%)
Unexpected empty sites	10	9	-1 (-10.00%)
Partial-row sites	2	2	0 (0.00%)
Site outcomes	extracted: 36; HTTP error + rows: 2; empty: 7; HTTP error + empty: 1; snapshot error + empty: 2	extracted: 38; HTTP error + rows: 1; empty: 6; HTTP error + empty: 1; snapshot error + empty: 2	—
Observed duration	53.09 s	56.59 s	+3.50 s (+6.60%)
Estimated CPU	51.55 s	51.33 s	-0.23 s (-0.44%)
Peak PSS	261.71 MiB	266.55 MiB	+4.84 MiB (+1.85%)
Peak RSS	264.63 MiB	269.95 MiB	+5.31 MiB (+2.01%)
Peak threads	27	26	-1 (-3.70%)
Mismatched sites	0	0	0
Timeout with items	0	0	0

Category rows	Base	HEAD	Δ
news	35 / 40	35 / 40	0
stocks	20 / 40	20 / 40	0
tech	35 / 40	35 / 40	0
sports	27 / 40	32 / 40	+5
games	37 / 40	37 / 40	0
life	30 / 40	30 / 40	0

Public-site content, timing, and memory are noisy. This single A/B run reports evidence only; site outcome counts explain missing rows without treating them as a deterministic regression.

CPU and memory timelines

These are bounded 40-point views of the same complete process-tree samples used by the HTML report. CPU uses 100% per occupied logical core; memory lines are RSS first and PSS second when complete PSS samples are available.

Base

xychart-beta
    title "Base CPU"
    x-axis "Elapsed seconds" [1.547, 2.549, 3.05, 4.551, 6.052, 7.555, 9.056, 10.557, 12.06, 13.562, 14.562, 16.064, 17.564, 19.067, 20.569, 22.072, 23.573, 25.074, 26.576, 27.077, 28.077, 29.581, 31.083, 32.585, 34.087, 35.589, 37.091, 38.591, 40.093, 41.095, 42.596, 44.097, 45.598, 47.1, 48.601, 50.102, 51.604, 52.605]
    y-axis "CPU percent" 0 --> 500
    line [115.89, 105.86, 101.83, 79.99, 175.96, 159.59, 77.97, 163.82, 113.93, 65.92, 108.01, 0, 2, 87.84, 6, 33.92, 173.82, 123.94, 119.7, 99.91, 131.85, 103.8, 143.81, 149.85, 115.87, 17.96, 83.79, 41.97, 71.94, 145.83, 111.99, 4, 102.04, 123.81, 65.95, 99.93, 133.92, 193.8]

xychart-beta
    title "Base memory: RSS then PSS"
    x-axis "Elapsed seconds" [0.044, 1.547, 2.549, 3.05, 4.551, 6.052, 7.555, 9.056, 10.557, 12.06, 13.562, 14.562, 16.064, 17.564, 19.067, 20.569, 22.072, 23.573, 25.074, 26.576, 27.077, 28.077, 29.581, 31.083, 32.585, 34.087, 35.589, 37.091, 38.591, 40.093, 41.095, 42.596, 44.097, 45.598, 47.1, 48.601, 50.102, 51.604, 52.605, 53.091]
    y-axis "Memory MiB" 0 --> 500
    line [12.12, 107.49, 111.63, 116.66, 127.55, 145.28, 164.77, 178.4, 162.89, 158.65, 170.92, 186.45, 165.65, 142.94, 157.61, 146.13, 141.54, 163.15, 158.71, 204.52, 264.63, 193.14, 212.74, 168.65, 173.21, 162.89, 143.94, 144.91, 146.83, 150.64, 155.36, 166.38, 150.1, 159.81, 176.25, 148.55, 151.03, 162.1, 188.36, 0]
    line [8.57, 103.8, 108.08, 113.34, 124.01, 141.85, 161.31, 174.95, 159.17, 155.71, 167.85, 184.22, 162.71, 140, 154.67, 143.19, 138.59, 160.27, 155.76, 201.61, 261.71, 190.35, 210.02, 165.62, 170.43, 159.72, 140.78, 141.99, 143.6, 147.41, 152.18, 163.24, 147.01, 156.69, 173.08, 145.51, 147.89, 158.98, 185.3, 0]

HEAD

xychart-beta
    title "HEAD CPU"
    x-axis "Elapsed seconds" [1.538, 3.042, 4.544, 6.548, 8.05, 9.552, 11.055, 12.558, 14.064, 15.567, 17.569, 19.07, 20.576, 21.125, 22.126, 23.629, 25.13, 26.632, 28.634, 29.134, 30.136, 31.668, 33.171, 34.671, 36.173, 37.679, 39.18, 41.181, 42.687, 44.189, 45.692, 47.193, 48.698, 50.199, 52.201, 53.703, 55.207, 56.207]
    y-axis "CPU percent" 0 --> 500
    line [121.85, 137.34, 105.93, 105.72, 115.64, 103.88, 129.82, 121.86, 91.93, 115.71, 87.7, 85.98, 17.85, 0, 0, 31.95, 113.89, 113.83, 99.87, 102, 103.77, 90.51, 43.9, 107.97, 111.87, 37.97, 65.98, 9.99, 109.91, 127.92, 87.89, 13.99, 93.49, 117.87, 51.96, 95.91, 113.7, 218.06]

xychart-beta
    title "HEAD memory: RSS then PSS"
    x-axis "Elapsed seconds" [0.033, 1.538, 3.042, 4.544, 6.548, 8.05, 9.552, 11.055, 12.558, 14.064, 15.567, 17.569, 19.07, 20.576, 21.125, 22.126, 23.629, 25.13, 26.632, 28.634, 29.134, 30.136, 31.668, 33.171, 34.671, 36.173, 37.679, 39.18, 41.181, 42.687, 44.189, 45.692, 47.193, 48.698, 50.199, 52.201, 53.703, 55.207, 56.207, 56.593]
    y-axis "Memory MiB" 0 --> 500
    line [11.96, 108.16, 123.34, 147.4, 178.59, 175.79, 187.84, 159.5, 156.5, 167.26, 183.47, 146.21, 150.18, 138.3, 138.3, 137.21, 140.54, 175.67, 148.95, 225.77, 269.95, 190.63, 164.51, 156.1, 186.71, 165.38, 142.38, 152.29, 151.88, 186.61, 183.75, 183.14, 153.11, 177.29, 175.45, 153.17, 153.71, 165.23, 198.54, 0]
    line [9.22, 104.91, 120.01, 144.2, 175.96, 172.48, 184.36, 156.17, 153.23, 163.85, 180.1, 142.74, 146.78, 134.84, 134.84, 133.75, 137.08, 172.3, 145.43, 222.41, 266.55, 187.33, 162.56, 152.75, 183.29, 161.99, 138.9, 148.85, 148.47, 183.4, 180.32, 179.73, 149.68, 173.89, 166.79, 149.68, 150.23, 161.77, 195.2, 0]

Deterministic fixture · required contract

The fixture exercises 15 routes: 8 are expected to emit rows and 7 intentionally exercise empty, loading, or timeout behavior.

Metric	Base	HEAD	Δ
Run status	success	success	—
Extracted rows / contract	40 / 40	40 / 40	0 (0.00%)
Contract fill	100.00%	100.00%	0.00 pp (0.00%)
Sites with rows	8 / 15	8 / 15	0 (0.00%)
Sites meeting row contract	15 / 15	15 / 15	0 (0.00%)
Unexpected empty sites	0	0	0
Partial-row sites	0	0	0
Site outcomes	extracted: 7; HTTP error + rows: 1; empty: 3; timeout + empty: 4	extracted: 7; HTTP error + rows: 1; empty: 3; timeout + empty: 4	—
Observed duration	19.99 s	19.99 s	-0.00 s (-0.01%)
Estimated CPU	0.40 s	0.42 s	+0.02 s (+4.96%)
Peak PSS	56.97 MiB	57.20 MiB	+0.22 MiB (+0.39%)
Peak RSS	60.85 MiB	61.05 MiB	+0.21 MiB (+0.34%)
Peak threads	18	18	0 (0.00%)
Mismatched sites	0	0	0
Timeout with items	0	0	0

Full HTML, JSON, CSV, logs, and page snapshots: workflow run and full spider-bench-results artifact.

The public-web A/B is informational. The exact HEAD fixture contract runs as a separate required CI check; timing and public-site content deltas remain non-blocking.

lanyue-llk · 2026-09-22T05:31:46Z

由 #734 替代：基于 main@2793b2407，PR 作者和所有新增提交的 GitHub author/committer 已核验关联 lanyue-llk。原分支、提交和本 PR 的讨论保留；这里只关闭被替代的 PR，不删除分支。新的 CI 与逐 case 验收记录归属替代 PR，旧版本测试结果不冒认为新版本结果。

github-actions · 2026-09-22T05:32:19Z

CI Regression Report

Source CI run · source state at render: in_progress · artifacts: 4/5

Check	Status	Signal
Release regression	⚪	artifact unavailable or invalid
Frontend differential	✅	1,020/1,020 cases matched; 0 issues
Agent episodes	✅	8/8 Moli episodes passed; 0 failures
Runtime/CDP contracts	✅	26 contract cases; 0 failures
CDP smoke	✅	48/48 groups passed; 535 scenarios

Release regression — ⚪ artifact unavailable or invalid

Artifact unavailable or invalid. See the source CI run for infrastructure details.

Frontend differential — ✅ 1,020/1,020 cases matched; 0 issues

Match	DOM mismatch	Diagnostic mismatch	Moli error	Reference error	Infrastructure error	Mismatched frames	Duration
1,020	0	0	0	0	0	0	8m 50.71s

Agent episodes · Moli vs Chromium — ✅ 8/8 Moli episodes passed; 0 failures

Fast contract	Episodes	Passed	Assertions	Ready p50	Episode p50
Moli	8	8	47/47	3.91 ms	25.75 ms
Chromium	8	8	47/47	297.97 ms	79.74 ms

Operation p50	Chromium	Moli	Delta	Delta %
navigate	41.17 ms	6.67 ms	−34.50 ms	−83.80%
observe	0.77 ms	0.84 ms	+0.06 ms	+8.27%
fill	2.69 ms	3.03 ms	+0.34 ms	+12.53%
click	2.75 ms	3.47 ms	+0.72 ms	+26.31%

Idle-resource episode	Peak RSS	Peak PSS	Average CPU	Peak CPU	Peak processes	Sampler
Moli	63.68 MiB	61.90 MiB	0.68%	9.99%	1	✅
Chromium	1254.94 MiB	405.90 MiB	2.11%	88.18%	13	⚠️ PSS partial

Runtime and CDP session contracts — ✅ 26 contract cases; 0 failures

Suite	Cases	Failures	Median case p50	Median case p95	Median peak PSS p50
Synthetic fetch	13	0	23.54 ms	28.38 ms	513.00 KiB
Long-lived CDP session	13	0	21.25 ms	23.27 ms	—

CDP smoke — ✅ 48/48 groups passed; 535 scenarios

Workers: 4 · cumulative group time: 148.14 s · failed groups: 0

Slowest group	Scenarios	Duration	Status
`target-lifecycle`	24	16.79 s	✅
`multi-page`	45	11.59 s	✅
`chromium-cdp`	56	10.78 s	✅
`dom-input`	27	10.76 s	✅
`layout-screenshot`	5	7.70 s	✅

All artifact fields are parsed by the trusted default-branch renderer; missing or invalid inputs remain visible as unavailable.

fix(reliability): harden diagnostics and test infrastructure

f13e9f4

lanyue-llk mentioned this pull request Sep 22, 2026

fix(reliability): harden diagnostics and test infrastructure #734

Merged

lanyue-llk closed this Sep 22, 2026

lanyue-llk mentioned this pull request Sep 22, 2026

fix(reliability): harden diagnostics and test infrastructure #755

Merged

Navigation Menu

Repository navigation

fix(reliability): handle closed diagnostics and harden test fixtures - #725

fix(reliability): handle closed diagnostics and harden test fixtures#725
lanyue-llk wants to merge 1 commit into
mainfrom
codex/browser-engineering-reliability

lanyue-llk commented Sep 22, 2026 •

edited

Loading

Uh oh!

github-actions Bot commented Sep 22, 2026

Uh oh!

github-actions Bot commented Sep 22, 2026

Uh oh!

lanyue-llk commented Sep 22, 2026

Uh oh!

github-actions Bot commented Sep 22, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Conversation

lanyue-llk commented Sep 22, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

修复机制

对应验收

范围与依赖

Uh oh!

github-actions Bot commented Sep 22, 2026

Sequential Navigation Soak A/B

Session resilience

Process-tree memory

HEAD memory by 50-navigation quarter

Uh oh!

github-actions Bot commented Sep 22, 2026

Spider Bench A/B

Public 48-site run · informational

CPU and memory timelines

Base

HEAD

Deterministic fixture · required contract

Uh oh!

lanyue-llk commented Sep 22, 2026

Uh oh!

github-actions Bot commented Sep 22, 2026

CI Regression Report

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

lanyue-llk commented Sep 22, 2026 •

edited

Loading