docs(perf): record rejected owned server mesh batches by louistrue · Pull Request #4072 · LTplus-AG/ifc-lite

louistrue · 2026-09-07T02:21:00Z

The owned server mesh-batch candidate failed its prespecified three-model HTTP continuation rule despite passing every declared exact-output, raw payload, cache-replay and cleanup gate. Record the rejection and its lesson without landing the implementation or expanding to the wider corpus.

The evidence retains all six fresh-process arms, startup separately from complete-SSE-plus-data-model readiness, sampled RSS, the frozen thresholds, and exact source/compiler/artifact identities. Single-pair results establish neither precise causality nor a physical-footprint change. Downloaded-fixture skips and ignored Rust tests remain explicit.

Validation: the candidate passed all six new ownership/pipeline tests, the full Rust workspace (2,851 passed, 36 ignored; external-fixture skip output retained), and cargo clippy --workspace --all-targets -- -D warnings. For this documentation-only change, JSON arithmetic/gates, privacy scanning, patch digests and git diff --check pass. Temporary-index application of the archived zero-context patch on public main 1b95c6652 reconstructs the exact measured candidate tree. No new performance measurement is implied by that source reconstruction.

Revert-oracle exemption: this PR changes only the performance ledger and non-executed evidence (JSON, Markdown and an archived patch). The oracle classifies evidence JSON as production code, but there is no application implementation to revert or regression test to make red. The maintainer-authorized revert-oracle-exempt label covers that artifact-classification false positive; all other checks and reviews remain required.

Closes #4066.

vercel · 2026-09-07T02:21:05Z

The latest updates on your projects. Learn more about Vercel for GitHub.

2 Skipped Deployments

Project	Deployment	Actions	Updated
ifc-lite-dev	Ignored		Sep 7, 2026 3:53am UTC
ifc-lite-viewer-embed	Ignored		Sep 7, 2026 3:53am UTC

coderabbitai · 2026-09-07T02:21:10Z

Warning

Review limit reached

Run on-demand review

_{This review includes 5 billable files and costs up to $1.25.}

Or wait 7 minutes for your next included review.

Check out review usage here.

View limit details

Limit details: You’ve used all 2 included reviews currently available.

Learn how review limits work.

Review configuration:

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Team

Run ID: 0c2b9492-547c-4564-82dd-e3db02d08359

📥 Commits

Reviewing files that changed from the base of the PR and between d2534a8 and 23ae029.

📒 Files selected for processing (5)

scripts/perf/README.md
scripts/perf/evidence/rejected-owned-batches-2026-09-07/README.md
scripts/perf/evidence/rejected-owned-batches-2026-09-07/measured-candidate.patch
scripts/perf/evidence/rejected-owned-batches-2026-09-07/screen.json
scripts/perf/evidence/rejected-owned-batches-2026-09-07/source-archive.json

_{Comment @coderabbitai help to get the list of available commands.}

cursor · 2026-09-07T02:21:11Z

Bugbot couldn't run - usage limit reached

Bugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit.

A user or team admin can review and increase usage limits in the Cursor dashboard.

(requestId: serverGenReqId_40f20aef-04b9-46d9-9175-baf642965208)

chatgpt-codex-connector · 2026-09-07T02:21:12Z

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-09-07T02:23:36.409850Z	`3ad3eff`	PR opened

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

github-actions · 2026-09-07T02:23:59Z

Claude review - no findings for `3ad3effb1`

Reviewed this diff and found nothing to flag.

cursor · 2026-09-07T02:28:27Z

Bugbot couldn't run - usage limit reached

Bugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit.

A user or team admin can review and increase usage limits in the Cursor dashboard.

(requestId: serverGenReqId_0b7fa417-a665-4d06-9e23-a4ab0a881cd0)

github-actions · 2026-09-07T02:30:44Z

Claude review - no findings for `be6869cb1` (partial)

Reviewed everything that fit the model prompt and found nothing to flag there.

⚠️ PARTIAL REVIEW: 2 changed file(s) were NOT shown to the reviewer -- too large to fit the model prompt, or too large for GitHub to return a patch for (#3679):

scripts/perf/evidence/server-pgo-counter-http-4059/screen.json
scripts/perf/evidence/server-pgo-darwin-2026-09-07/pre-screen-provenance.json

Nothing vouches for those files. This verdict covers only the files that were reviewed.

cursor · 2026-09-07T02:48:53Z

Bugbot couldn't run - usage limit reached

Bugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit.

A user or team admin can review and increase usage limits in the Cursor dashboard.

(requestId: serverGenReqId_4d3df061-2eca-45cb-9b7a-2e3eec99ad6e)

cursor · 2026-09-07T02:50:59Z

Bugbot couldn't run - usage limit reached

Bugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit.

A user or team admin can review and increase usage limits in the Cursor dashboard.

(requestId: serverGenReqId_f024f8dd-6d1e-4864-87d9-041b336be5e9)

github-actions · 2026-09-07T02:53:21Z

Claude review - no findings for `964b656fd` (partial)

Reviewed everything that fit the model prompt and found nothing to flag there.

⚠️ PARTIAL REVIEW: 3 changed file(s) were NOT shown to the reviewer -- too large to fit the model prompt, or too large for GitHub to return a patch for (#3679):

scripts/perf/evidence/server-pgo-counter-http-4059/screen.json
scripts/perf/evidence/server-pgo-darwin-2026-09-07/pre-screen-provenance.json
scripts/perf/evidence/server-pgo-full-value-http-4059/screen.json

Nothing vouches for those files. This verdict covers only the files that were reviewed.

cursor · 2026-09-07T03:01:51Z

Bugbot couldn't run - usage limit reached

Bugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit.

A user or team admin can review and increase usage limits in the Cursor dashboard.

(requestId: serverGenReqId_589fa600-62c7-4a89-adc1-0d722b29ea58)

github-actions · 2026-09-07T03:04:11Z

Claude review - no findings for `085e89c94` (partial)

Reviewed everything that fit the model prompt and found nothing to flag there.

⚠️ PARTIAL REVIEW: 3 changed file(s) were NOT shown to the reviewer -- too large to fit the model prompt, or too large for GitHub to return a patch for (#3679):

scripts/perf/evidence/server-pgo-counter-http-4059/screen.json
scripts/perf/evidence/server-pgo-darwin-2026-09-07/pre-screen-provenance.json
scripts/perf/evidence/server-pgo-full-value-http-4059/screen.json

Nothing vouches for those files. This verdict covers only the files that were reviewed.

cursor · 2026-09-07T03:53:33Z

Bugbot couldn't run - usage limit reached

Bugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit.

A user or team admin can review and increase usage limits in the Cursor dashboard.

(requestId: serverGenReqId_beb3a2ed-0feb-4138-80ed-d22ddba2a03b)

github-actions · 2026-09-07T03:56:40Z

Claude review - no findings for `23ae0296c`

Reviewed this diff and found nothing to flag.

87 files and 2.3 MB of rejected-experiment blobs and unrun harnesses, none of which CI executes and none of which any scripts/perf entrypoint reads. The tree needed a CI bypass to exist: every one of the seven evidence-only PRs that built it (#4060, #4061, #4068, #4069, #4070, #4071, #4072 -- each touching nothing but the tree and the ledger) carried revert-oracle-exempt, which skips the lane entirely. #4069's body says why: "The oracle classified retained measurement JSON as production code and failed because it has no changed application test." Three of its READMEs specify zero-context patches so blank context lines do not trip the whitespace gate. The repo keeps what CI runs and what the next agent must read before spiking; the rest lives in git history at a named SHA. Each of the nine ledger sections that linked into the tree becomes one bullet in the section matching its verdict, in the house style the older #1445 entry already uses: verdict, the mechanism that failed, the headline number with its cohort size, the issue and PR, and a `git show` pointer pinned to 4fbbe8d. Five went to "Dead ends", four to "Shipped wins", and the #3978 harness smoke folded into the manual-readiness section as one sentence. #4031's verdict joins them: a dead end that was recorded only in an issue comment and absent from the ledger entirely. Two things stay inline rather than behind a pointer, because they are what a re-spike gets wrong: the PGO RUSTFLAGS decisions (empty control, -Cprofile- generate with its Darwin section alignment for counter-only training only, -Cprofile-use, and the build-std/target split that makes the probe's profile and the server's non-interchangeable), and the Y-up orientation rule (x, y, z) -> (x, z, -y). The archived patches are the only public copy of their mechanism. d979e92e4, 3e675edea, 67c3f6d31 and bdc38d30c are on no remote (`git branch -r --contains` is empty for all four), so each bullet cites its patch path plus a public apply base checked reachable on origin/main: 96ea5f0, e409924, 1b95c66. The two chained follow-up patches (later-tests.patch reproducing bdc38d30c, test-followup.patch reproducing c0ef3e802) get a pointer each for the same reason. All five patches were applied for real against their cited bases with `git apply --cached --unidiff-zero` into a temporary index; all five succeed and both follow-ups chain after their measured patch. 7509432 is NOT on origin/main -- it is the local pre-squash measured source -- so it is not cited as an apply base. Every one of the 15 `git show` pointers resolves. .github/workflows/test.yml's path-filter comment names the tree. The claim is historical and stays true, so it is only marked as removed so a future grep does not chase the path. No workflow logic changes. `git grep perf/evidence` now returns that comment and the pinned pointers, nothing else. Deleting the tree also removes the only Python project root outside rust/python and tools/ifcopenshell_reference, and with it the dependabot noise source behind #4073 and #4074. Closes #4112. Claude-Session: https://claude.ai/code/session_0193douQ6sTYHE65DJmyAei9 Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>

louistrue added the revert-oracle-exempt Maintainer-approved: executable test-observation check is inapplicable to this refactor label Sep 7, 2026

github-actions Bot added the llm-reviewed A review was verified as posted for this PR's head. label Sep 7, 2026

louistrue force-pushed the docs/4066-owned-batches-verdict-t3 branch from 3ad3eff to be6869c Compare September 7, 2026 02:28

github-actions Bot removed the llm-reviewed A review was verified as posted for this PR's head. label Sep 7, 2026

louistrue force-pushed the docs/4066-owned-batches-verdict-t3 branch from be6869c to 75ad161 Compare September 7, 2026 02:48

louistrue force-pushed the docs/4066-owned-batches-verdict-t3 branch from 75ad161 to 964b656 Compare September 7, 2026 02:50

louistrue force-pushed the docs/4066-owned-batches-verdict-t3 branch from 964b656 to 085e89c Compare September 7, 2026 03:01

docs(perf): record rejected owned server batch screen

23ae029

louistrue force-pushed the docs/4066-owned-batches-verdict-t3 branch from 085e89c to 23ae029 Compare September 7, 2026 03:53

github-actions Bot added the llm-reviewed A review was verified as posted for this PR's head. label Sep 7, 2026

louistrue merged commit 1bd9ccf into main Sep 7, 2026
33 checks passed

louistrue deleted the docs/4066-owned-batches-verdict-t3 branch September 7, 2026 06:56

Uh oh!

Uh oh!

Conversation

louistrue commented Sep 7, 2026

Uh oh!

vercel Bot commented Sep 7, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

coderabbitai Bot commented Sep 7, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Review limit reached

Uh oh!

cursor Bot commented Sep 7, 2026

Bugbot couldn't run - usage limit reached

Uh oh!

chatgpt-codex-connector Bot commented Sep 7, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codex Review Summary

Uh oh!

github-actions Bot commented Sep 7, 2026

Claude review - no findings for 3ad3effb1

Uh oh!

cursor Bot commented Sep 7, 2026

Bugbot couldn't run - usage limit reached

Uh oh!

github-actions Bot commented Sep 7, 2026

Claude review - no findings for be6869cb1 (partial)

Uh oh!

cursor Bot commented Sep 7, 2026

Bugbot couldn't run - usage limit reached

Uh oh!

cursor Bot commented Sep 7, 2026

Bugbot couldn't run - usage limit reached

Uh oh!

github-actions Bot commented Sep 7, 2026

Claude review - no findings for 964b656fd (partial)

Uh oh!

cursor Bot commented Sep 7, 2026

Bugbot couldn't run - usage limit reached

Uh oh!

github-actions Bot commented Sep 7, 2026

Claude review - no findings for 085e89c94 (partial)

Uh oh!

cursor Bot commented Sep 7, 2026

Bugbot couldn't run - usage limit reached

Uh oh!

github-actions Bot commented Sep 7, 2026

Claude review - no findings for 23ae0296c

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

vercel Bot commented Sep 7, 2026 •

edited

Loading

coderabbitai Bot commented Sep 7, 2026 •

edited

Loading

chatgpt-codex-connector Bot commented Sep 7, 2026 •

edited

Loading

Claude review - no findings for `3ad3effb1`

Claude review - no findings for `be6869cb1` (partial)

Claude review - no findings for `964b656fd` (partial)

Claude review - no findings for `085e89c94` (partial)

Claude review - no findings for `23ae0296c`