Skip to content

fix(indexeddb): finish upgrade opens after transaction completion - #427

Open
ldm0 wants to merge 1 commit into
mainfrom
fix/indexeddb-upgrade-order
Open

fix(indexeddb): finish upgrade opens after transaction completion#427
ldm0 wants to merge 1 commit into
mainfrom
fix/indexeddb-upgrade-order

Conversation

@ldm0

@ldm0 ldm0 commented Sep 7, 2026

Copy link
Copy Markdown
Member

Dependency / merge order

Stacked on #426 (fix/svg-create-rect) so this PR contains exactly one IndexedDB commit. The dependency is the combined SVG-gated application-startup smoke, not the IndexedDB implementation. Merge #426 first, then rebase this commit onto main and retarget this PR; if #426 is squash-merged, do not merge the old SVG commit a second time.

Summary

  • Keep the exact open request attached to the upgrade transaction until accepted requests and their microtasks finish.
  • Publish open success/error in a separate task after transaction completion; cover abort rollback and a connection closed before result delivery.
  • Preserve FIFO among waiting open/delete requests for the same database and remove eager/duplicate lifecycle paths.
  • Preserve task-boundary, migration, microtask, rollback, deletion-order, and combined application-startup regressions. The standalone SVGRect contract lives in fix(svg): expose detached SVGRect values #426.

Validation

  • Local: cargo fmt --all and git diff --check passed.
  • Per request, Clippy, nextest, and runtime smoke are left to CI; they were not rerun locally for this split.

@github-actions

github-actions Bot commented Sep 7, 2026 •
edited
Loading

Copy link
Copy Markdown

Spider Bench A/B

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

Public HEAD: 184 / 240 rows; 38 / 48 sites produced rows; 10 unexpected empty sites.

✅ Deterministic fixture contract is clean.

Common ancestor 3a5cf7f1239b → HEAD bd7ddea62445; benchmark order: head-first; workflow: success.

Public 48-site run · informational

Metric Base HEAD Δ
Run status success success —
Extracted rows / contract 189 / 240 184 / 240 -5 (-2.65%)
Contract fill 78.75% 76.67% -2.08 pp (-2.64%)
Sites with rows 39 / 48 38 / 48 -1 (-2.56%)
Sites meeting row contract 37 / 48 36 / 48 -1 (-2.70%)
Unexpected empty sites 9 10 +1 (+11.11%)
Partial-row sites 2 2 0 (0.00%)
Site outcomes extracted: 38; HTTP error + rows: 1; empty: 6; HTTP error + empty: 1; snapshot error + empty: 2 extracted: 36; HTTP error + rows: 2; empty: 7; HTTP error + empty: 1; snapshot error + empty: 2 —
Observed duration 42.37 s 47.71 s +5.34 s (+12.60%)
Estimated CPU 43.39 s 42.40 s -0.99 s (-2.28%)
Peak PSS 198.76 MiB 211.77 MiB +13.01 MiB (+6.55%)
Peak RSS 202.42 MiB 215.53 MiB +13.11 MiB (+6.47%)
Peak threads 26 27 +1 (+3.85%)
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 32 / 40 27 / 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" [1.037, 2.54, 3.54, 4.542, 6.043, 7.043, 8.543, 9.544, 10.545, 12.047, 13.048, 14.049, 15.552, 16.553, 17.553, 19.053, 20.054, 21.556, 22.558, 23.559, 25.061, 26.061, 26.562, 27.063, 28.565, 29.566, 30.568, 32.07, 32.57, 33.071, 34.072, 35.574, 36.075, 36.576, 38.078, 39.079, 40.08, 41.581]
    y-axis "CPU percent" 0 --> 200
    line [41.9, 123.8, 152.07, 161.84, 173.55, 166.07, 180.1, 145.93, 135.64, 127.68, 119.94, 87.78, 95.77, 29.98, 4, 142.04, 131.94, 151.64, 121.8, 127.85, 79.81, 115.96, 185.87, 139.59, 25.98, 91.97, 69.84, 97.91, 135.91, 141.7, 97.92, 11.97, 0, 45.88, 129.82, 165.8, 67.96, 175.79]
Loading 
xychart-beta
    title "Base memory: RSS then PSS"
    x-axis "Elapsed seconds" [0.036, 1.037, 2.54, 3.54, 4.542, 6.043, 7.043, 8.543, 9.544, 10.545, 12.047, 13.048, 14.049, 15.552, 16.553, 17.553, 19.053, 20.054, 21.556, 22.558, 23.559, 25.061, 26.061, 26.562, 27.063, 28.565, 29.566, 30.568, 32.07, 32.57, 33.071, 34.072, 35.574, 36.075, 36.576, 38.078, 39.079, 40.08, 41.581, 42.369]
    y-axis "Memory MiB" 0 --> 500
    line [12.09, 90.2, 113.02, 130.8, 157.85, 152.74, 154.4, 185.31, 165.12, 182.18, 163.36, 160.86, 168.84, 152.07, 139.67, 143.35, 180.76, 166.02, 162.9, 173.07, 191.93, 148, 181.76, 175.86, 179.2, 153.08, 148.95, 152.38, 169.01, 202.42, 177.38, 184.41, 156.41, 156.41, 167.63, 187.11, 167.83, 164.83, 180.98, 0]
    line [8.72, 86.52, 109.38, 127.15, 154.19, 148.99, 150.63, 181.55, 161.35, 178.44, 159.46, 157.09, 165.07, 148.31, 135.91, 139.6, 177.59, 162.11, 159.13, 169.3, 185.31, 144.25, 177.89, 172.09, 175.44, 149.31, 145.18, 148.61, 165.2, 198.76, 173.66, 180.64, 152.64, 152.64, 164.13, 183.34, 163.83, 162.74, 177.44, 0]
Loading

HEAD

xychart-beta
    title "HEAD CPU"
    x-axis "Elapsed seconds" [1.035, 1.536, 2.537, 4.039, 5.542, 6.543, 8.045, 9.546, 10.546, 12.049, 13.551, 14.051, 14.551, 16.051, 17.551, 18.552, 20.055, 21.558, 22.56, 24.063, 25.566, 26.566, 28.069, 29.571, 30.572, 32.075, 33.576, 34.578, 36.078, 37.581, 38.581, 40.084, 41.587, 42.587, 44.09, 45.591, 46.593, 47.093]
    y-axis "CPU percent" 0 --> 500
    line [45.97, 115.86, 107.88, 119.78, 167.89, 167.79, 101.74, 197.83, 67.95, 145.88, 69.96, 0, 2, 2, 128, 23.95, 7.99, 95.81, 25.94, 37.95, 57.94, 119.96, 111.74, 121.87, 69.83, 109.87, 23.94, 39.99, 81.98, 59.86, 121.84, 77.91, 1.99, 125.85, 167.58, 93.79, 111.92, 209.83]
Loading 
xychart-beta
    title "HEAD memory: RSS then PSS"
    x-axis "Elapsed seconds" [0.035, 1.035, 1.536, 2.537, 4.039, 5.542, 6.543, 8.045, 9.546, 10.546, 12.049, 13.551, 14.051, 14.551, 16.051, 17.551, 18.552, 20.055, 21.558, 22.56, 24.063, 25.566, 26.566, 28.069, 29.571, 30.572, 32.075, 33.576, 34.578, 36.078, 37.581, 38.581, 40.084, 41.587, 42.587, 44.09, 45.591, 46.593, 47.093, 47.708]
    y-axis "Memory MiB" 0 --> 500
    line [12.05, 91.4, 105.63, 115.39, 140.26, 162.53, 168.31, 156.54, 159.75, 164.79, 174.47, 162.61, 162.61, 140.21, 143.57, 156.66, 156.91, 150.97, 140.03, 130.47, 133.71, 145.48, 164.5, 192.32, 156.96, 154.02, 171.73, 155.99, 147.04, 153.21, 157.48, 215.53, 191.57, 153.5, 172.02, 167.91, 154.25, 176.02, 195.77, 0]
    line [8.72, 87.69, 101.91, 111.68, 136.54, 158.83, 164.57, 152.91, 155.99, 161.07, 170.81, 158.89, 158.88, 136.49, 139.84, 152.93, 150.97, 147.25, 136.3, 126.78, 129.98, 141.75, 160.08, 188.6, 153.23, 150.07, 165.87, 152.33, 143.34, 149.48, 153.76, 211.77, 187.84, 149.78, 168.69, 166.31, 150.57, 172.37, 192.18, 0]
Loading

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 19.88 s 19.88 s +0.00 s (+0.01%)
Estimated CPU 0.42 s 0.43 s +0.01 s (+2.38%)
Peak PSS 57.46 MiB 57.29 MiB -0.17 MiB (-0.29%)
Peak RSS 60.95 MiB 60.89 MiB -0.06 MiB (-0.10%)
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.

@github-actions

github-actions Bot commented Sep 7, 2026 •
edited
Loading

Copy link
Copy Markdown

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 3a5cf7f123 → HEAD bd7ddea624; benchmark order: head-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 476.1 s 389.1 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 198.7 MiB 214.0 MiB +15.2 MiB
Periodic peak PSS 196.3 MiB 209.2 MiB +13.0 MiB
RSS · first 10 avg 160.8 MiB 156.3 MiB -4.5 MiB
RSS · last 10 avg 166.4 MiB 165.6 MiB -0.9 MiB
RSS · last − first window +5.6 MiB +9.2 MiB +3.6 MiB
RSS · warm slope / 100 nav +2.7 MiB +3.0 MiB +0.4 MiB
PSS · first 10 avg 158.0 MiB 152.8 MiB -5.2 MiB
PSS · last 10 avg 163.1 MiB 163.3 MiB +0.2 MiB
PSS · last − first window +5.1 MiB +10.5 MiB +5.4 MiB
PSS · warm slope / 100 nav +2.6 MiB +3.6 MiB +1.0 MiB
Peak file descriptors 61 62 1
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
1–50 159.8 MiB 148.8 MiB 189.5 MiB 156.4 MiB 145.5 MiB 51
51–100 162.1 MiB 145.2 MiB 184.5 MiB 159.0 MiB 141.9 MiB 49
101–150 164.3 MiB 155.8 MiB 197.5 MiB 161.3 MiB 152.8 MiB 51
151–200 165.1 MiB 147.6 MiB 190.6 MiB 162.5 MiB 145.0 MiB 50

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

github-actions Bot commented Sep 7, 2026 •
edited
Loading

Copy link
Copy Markdown

CI Regression Report

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

Check Status Signal
Release regression ✅ HEAD/base failures 0/0; raw binary +0.016301%
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 ✅ 49/49 groups passed; 542 scenarios
Release regression — ✅ HEAD/base failures 0/0; raw binary +0.016301%

Package and image size

Metric Base HEAD Delta Delta %
Raw binary 149.35 MiB 149.38 MiB +24.93 KiB +0.016301%
Stripped binary 103.68 MiB 103.70 MiB +24.00 KiB +0.022606%
gzip binary 50.25 MiB 50.27 MiB +14.78 KiB +0.028714%
Rootfs 156.26 MiB 156.28 MiB +20.00 KiB +0.012499%
gzip rootfs 59.08 MiB 59.09 MiB +9.51 KiB +0.015714%

Startup latency and PSS

Case Base p50 HEAD p50 Delta Base PSS HEAD PSS Delta
Serve ready 27.20 ms 27.12 ms −0.08 ms 4.72 MiB 3.66 MiB −1.06 MiB
CDP first page 52.00 ms 50.98 ms −1.02 ms 4.52 MiB 4.12 MiB −409.50 KiB
CDP warm page 13.78 ms 13.16 ms −0.63 ms 48.49 MiB 48.29 MiB −208.00 KiB
CLI about:blank 14.57 ms 14.39 ms −0.18 ms 3.30 MiB 1.86 MiB −1.44 MiB
CLI local JS 18.54 ms 17.17 ms −1.37 ms 3.82 MiB 2.46 MiB −1.36 MiB

Concurrency matrix

Concurrency Base p50 HEAD p50 Delta Base PSS HEAD PSS Delta HEAD failures Unstable B/H
1 16.91 ms 16.53 ms −0.39 ms 792.00 KiB 820.00 KiB +28.00 KiB 0 0/0
5 18.74 ms 20.22 ms +1.49 ms 13.72 MiB 27.64 MiB +13.92 MiB 0 0/0
10 18.84 ms 19.72 ms +0.88 ms 15.60 MiB 27.50 MiB +11.90 MiB 0 0/0
25 19.42 ms 19.24 ms −0.17 ms 22.48 MiB 27.00 MiB +4.51 MiB 0 0/0
100 19.67 ms 20.12 ms +0.45 ms 13.78 MiB 26.73 MiB +12.94 MiB 0 0/0
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 6m 24.37s
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 3.45 ms 24.99 ms
Chromium 8 8 47/47 262.09 ms 87.21 ms
Operation p50 Chromium Moli Delta Delta %
navigate 44.37 ms 6.85 ms −37.52 ms −84.56%
observe 1.67 ms 0.86 ms −0.81 ms −48.35%
fill 3.11 ms 3.26 ms +0.15 ms +4.89%
click 5.11 ms 3.62 ms −1.49 ms −29.18%
Idle-resource episode Peak RSS Peak PSS Average CPU Peak CPU Peak processes Sampler
Moli 63.92 MiB 61.22 MiB 0.64% 10.02% 1 ✅
Chromium 1286.87 MiB 336.21 MiB 1.82% 90.07% 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 18.07 ms 19.24 ms 607.00 KiB
Long-lived CDP session 13 0 19.54 ms 20.56 ms —
CDP smoke — ✅ 49/49 groups passed; 542 scenarios

Workers: 4 · cumulative group time: 149.98 s · failed groups: 0

Slowest group Scenarios Duration Status
target-lifecycle 24 16.30 s ✅
chromium-cdp 56 13.19 s ✅
multi-page 45 11.00 s ✅
dom-input 27 10.32 s ✅
layout-screenshot 5 7.63 s ✅

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

Base automatically changed from fix/svg-create-rect to main September 8, 2026 12:38
@ldm0
ldm0 force-pushed the fix/indexeddb-upgrade-order branch 3 times, most recently from faf8c14 to 7a7ffdd Compare September 9, 2026 17:44
@ldm0
fix(indexeddb): finish upgrade opens after transaction completion
bd7ddea 
Retain the exact open request until normal transaction request draining and checkpoint deactivation complete. Publish open success or error in a separate task, handle aborted/closed provisional connections, and preserve FIFO among waiting open/delete requests.

Remove eager upgrade-completion and duplicate blocked-request dispatch paths. Update selected-task boundary regressions and add Chromium-calibrated CDP coverage for migration, microtasks, rollback, deletion order and SVG-gated application startup.
@ldm0
ldm0 force-pushed the fix/indexeddb-upgrade-order branch from 7a7ffdd to bd7ddea Compare September 12, 2026 22:52
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

1 participant

@ldm0