[snissn]

Current readiness checkpoint

Published head: e05e6aa5bfd9c4786be89335d4be8f4a57a9cf50; base: 78e4de8642b4e177a7ec24b87e18c51c4e0bd684; tree: 9e3e933bb576f17c0e6c6ac968f823106ada0b45.

This supersedes the earlier scheduling deferral. The candidate preserves original PR history and changes only the vector allocation-test setup/checked Close and maintained discovery inventory. Production behavior, test routes and strict 100-sample zero-allocation assertion remain unchanged; #5390 is not a product dependency.

Focused normal/race and affected owning-scope checks passed on the identical reviewed tree. Successful isolated normal-child execution provides source-bound indirect strict-zero evidence; race intentionally skips allocation measurement.

Current-head CI passed all 40 TreeDB jobs, including required aggregate and Windows nativewire. All 56 current checks pass. Exact-head Codex review is clean; no unresolved review threads.

Historical bdad Windows failure remains preserved with cause INCONCLUSIVE. New-head success does not establish its causal repair or retroactively waive it.

Final live policy and ownership checks, an authorized merge decision, and merged-main qualification remain.

Summary by CodeRabbit

• Tests
  • Improved resource cleanup and setup checks in a vector index search test.

[coderabbitai]

No actionable comments were generated in the recent review.

Walkthrough

The CI discovery-source checksum changed. A vector-index test now reports database close errors and completes allocation setup after rebuilding the index.

Changes: CI Discovery Checksum

Layer / File(s)	Summary
Update discovery-source checksum .github/ci/ci_impact.json	The configured discovery-source SHA-256 value changed.

Vector Allocation Test

Layer / File(s)	Summary
Complete vector allocation test setup TreeDB/collections/vector_index_search_test.go	The test now reports database close errors and calls allocation setup after rebuilding the index.

Estimated code review effort: 2 (Simple) | ~5 minutes

Change: Other

Merge Risk: Minimal · up to e05e6

This updates a test's allocation-measurement setup and the CI discovery digest. It has no production behavior impact. Required CI, including the separately reported Windows failure, must still pass under the normal merge policy.

Pre-merge checks: 4

Check name	Status	Explanation
Title check		The title clearly and concisely describes the main change: completing vector setup before measuring buffer allocations in the TreeDB test.
Linked Issues check		Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check		Check skipped because no linked issues were found for this pull request.
Description Check		Check skipped - CodeRabbit’s high-level summary is enabled.

Finishing Touches

Generate unit tests (beta)

• Commit to this branch
• Create a new PR

• Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

Share

• X
• Mastodon
• Reddit
• LinkedIn

Recent review info

Run configuration

• Source mode: Auto
• Repository: Unavailable
• Branch: Unavailable
• File: No configuration file from this repository applied
• Commit: Unavailable
• Layers: Organization UI
• Review profile: CHILL
• Plan: Advanced
• Run ID: 3574d4e0-dba4-49d9-9983-6517264719f9

Commits

Reviewing files that changed from the base of the PR and between 78e4de8 and e05e6aa.

Files selected for processing: 2

• .github/ci/ci_impact.json
• TreeDB/collections/vector_index_search_test.go

Included review availability: This review used your included allowance. Your plan provides up to 1 included review per hour; 0 remain after this review.

[snissn]

Required CI remains blocking. Actual race8 job126309294647/run41749058940 checked out67559eb19c5292e9ec74cb3c17c11841af0321df (head e9f2ca1 into29b), Ubuntu26.04/linux-amd64 Go1.26.9. Both TestSplitVectorInsertTerminalTailRecoveryV1 children stop at 'fixture requires fixed WAL v3', lines223/256, before mutation/reopen. This matches the actual #5385 failure; no DATA RACE warning was observed. Source permits stream WAL on positively identified tmpfs/RAMFS, but actual CI medium/header is unknown. One bounded independent fixture inquiry is active; no rerun/skip/waiver or production-cause claim.

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has insufficient remaining quota. Try requesting a review at a lower effort level.

[chatgpt-codex-connector]

Codex Review Summary: ✅ Completed

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

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-11-11T19:33:08.949530Z	e05e6aa	New commits

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.

[snissn]

Current-head required CI remains blocked. Run41765342305/job126362286885 failed TestSparseCatalogConsumerRejectsTamperedRouteV1 (33.29s) at fixed_peer_sparse_catalog_contract_v1_test.go:313: context deadline exceeded; nativewire package exited 1 after551.060s. Actual checkout was synthetic b516d0d1aa57dafa0fac8a999f5368bac4760e73, tree 4ce486d6eaf45d99f6b58f8c866a3308b2397b62, equal to current bdad84c8 candidate. No DATA RACE signature appears in the retained job log. This signature is distinct from the earlier fixed-WAL fixture failure and the separately owned #5373 Raft failure.

Full decoded log SHA256 a96e09d5632c51d9f5652a4f7397bfb1f0f3a672d01a6e6e5949127616c6b5a5, evidence 065846c116b66b66fa7354691b690a1bd7139de29b965b8e93d5ec129922d640. One existing worker has a bounded source-only diagnosis of the actual failed statement and readiness/deadline/cleanup path. No unchanged retry, timeout change, skip, flake conclusion or waiver is selected.

The owning vector allocation/parallel/recovery tests independently passed their actual normal and race entries; normal supports only the source-bound indirect strict-zero proof, race correctness only. That scoped evidence does not clear this required Windows failure or permit merge.

[snissn]

Preserved pre-replan body (2026-11-04). Current issue body governs scheduling; original evidence and gates remain.

The public vector buffer allocation test starts warming immediately after rebuilding its index. Complete setup through the existing checkpoint-and-StateToken helper before warm/measured searches, and check deferred Close errors.

The current head is bdad84c837cdc9203ef5a411d964bc6b1a44f642, tree 4ce486d6eaf45d99f6b58f8c866a3308b2397b62, normally integrated with actual main 4c0f30343c96e5756a1e58e86f4de140ce2a8a5a. Both fixture prerequisites #5387/#5388 have merged. The two-file diff remains the original test repair and generated CI discovery binding. The complete tree, setup helper and actual collection builder are unchanged from the reviewed provisional integration; no production, journal or Stats delta is imported.

Preserve five 3D rows, exact TopK3/EfSearch5/MaxDecodedBlocks1, production stats, no-document caller-buffer route, child90s/parent2min isolation, actual 100-sample buffered zero-allocation assertion and race correctness-only behavior. Owned-route allocations stay measured with no new threshold. Current-head CI must actually enter the strict normal and race correctness paths plus affected owning scope and all required repository checks.

Historical head e9f2ca1d retains named normal PASS (job126309294897) and race PASS (126309294647). Its required race8/aggregate failed the separate terminal-tail fixed-WAL precondition before fault/reopen checks; actual old header/filesystem and cause are unknown. Historical 12-versus-zero evidence remains unexplained and retained. Neither prior integrated passes nor later fixture merges qualify this new head without its own CI.

Independent source revalidation 15d7a9c7, separate-author ROOT payload review 2aa1ff03, and actual-main publication proof 599e79d9 found no blocker. All exact-head CI, completed review, final live policy/merge gates and merged-main qualification remain required. No product speedup or parent acceptance is claimed; #5225 and final matched Mongo obligations remain open.