Skip to content

fix(memory-core): skip qmd zero-hit search sync#90030

Merged
vincentkoc merged 1 commit into
openclaw:mainfrom
sahibzada-allahyar:tui-90023-qmd-zero-hit
Jul 1, 2026
Merged

fix(memory-core): skip qmd zero-hit search sync#90030
vincentkoc merged 1 commit into
openclaw:mainfrom
sahibzada-allahyar:tui-90023-qmd-zero-hit

Conversation

@sahibzada-allahyar

@sahibzada-allahyar sahibzada-allahyar commented Jun 3, 2026 •
edited by vincentkoc
Loading

Copy link
Copy Markdown
Contributor

What Problem This Solves

Fixes #90023.

Long-lived QMD-backed memory_search calls can turn a normal zero-hit result into synchronous index maintenance by awaiting manager.sync({ reason: "search", force: true }). QMD maps that call to update/embed work, which can stall an interactive turn for tens of seconds.

Why This Change Was Made

The generic zero-hit repair remains useful for builtin/fallback managers, but QMD already owns its search and update lifecycle. The repair is therefore backend- and lifecycle-specific:

  • long-lived QMD managers return the zero-hit result without forcing synchronous update work;
  • one-shot CLI QMD managers keep the bootstrap sync/retry because they do not have a background lifecycle;
  • non-QMD managers keep the existing sync/retry behavior.

This is narrower than removing zero-hit repair for every backend and does not add configuration or timeout policy.

User Impact

Interactive QMD memory misses return without entering QMD update/embed maintenance. Existing builtin/fallback freshness repair and one-shot CLI bootstrap behavior remain unchanged.

Evidence

  • Canonical issue QMD memory_search zero-hit queries trigger synchronous force sync and stall interactive turns #90023 includes repeated live QMD reports showing 69-75 second zero-hit stalls and sub-second results after bypassing the synchronous QMD update path.
  • node scripts/run-vitest.mjs extensions/memory-core/src/tools.test.ts - 22 passed
  • Blacksmith Testbox tbx_01kweq20wqs24wspdyf03kg2zh - corepack pnpm check:changed passed
  • Fresh autoreview on the final local diff - no accepted/actionable findings
  • git diff --check origin/main...HEAD - clean

Focused coverage proves:

  • long-lived QMD zero-hit search performs one search and zero forced syncs while preserving runtime debug;
  • one-shot QMD zero-hit search still syncs and retries;
  • non-QMD zero-hit search still syncs and retries.

Hosted CI is required on the pushed exact head before merge.

@openclaw-barnacle openclaw-barnacle Bot added extensions: memory-core Extension: memory-core size: XS triage: needs-real-behavior-proof Candidate: external PR needs after-fix proof from a real setup. labels Jun 3, 2026
@clawsweeper

clawsweeper Bot commented Jun 3, 2026 •
edited
Loading

Copy link
Copy Markdown
Contributor

Codex review: needs maintainer review before merge. Reviewed June 21, 2026, 2:22 PM ET / 18:22 UTC.

Summary
The PR changes memory_search so QMD zero-hit results skip the generic forced sync retry path while non-QMD backends keep that retry behavior, and it adds a regression test for the QMD guard.

PR surface: Source +4, Tests +23. Total +27 across 2 files.

Reproducibility: yes. source-reproducible: current main and v2026.6.9 still await a forced sync after zero memory_search results, and QMD sync awaits update maintenance. I did not run a live large-corpus QMD timing repro in this read-only pass, but the linked issue and PR body include live QMD behavior evidence.

Review metrics: 1 noteworthy metric.

  • QMD zero-hit fallback: 1 backend-specific retry condition changed. QMD no longer uses the generic forced sync/retry after empty results, so maintainers should review the freshness tradeoff before merge.

Stored data model
Persistent data-model change detected: vector/embedding metadata: extensions/memory-core/src/tools.test.ts. Confirm migration or upgrade compatibility proof before merge.

Root-cause cluster
Relationship: fixed_by_candidate
Canonical: #90023
Summary: This PR is a candidate fix for the canonical QMD zero-hit forced-sync stall issue; overlapping open PRs exist, but none is a merged or clearly safer replacement for closing this branch.

Members:

Proposal only: this assessment does not dispatch repair, suppress jobs, mutate sibling items, close, or merge anything.

Merge readiness
Overall: 🐚 platinum hermit
Proof: 🦞 diamond lobster
Patch quality: 🐚 platinum hermit
Result: ready for maintainer review.

Overall follows the weaker of proof and patch quality, so missing proof can cap an otherwise strong patch.

Rank-up moves:

  • Refresh the dirty branch against current main.
  • Have maintainers choose between the QMD-only skip and a bounded/background refresh variant.

Risk before merge

  • [P1] The PR is currently dirty/unmergeable on GitHub and needs a refresh before merge.
  • [P2] Skipping QMD zero-hit forced sync changes fallback behavior for stale QMD indexes: users can now receive prompt empty results instead of immediate search-time freshness repair.
  • [P1] Several open sibling PRs overlap this memory_search path, but none is a merged or clearly safer canonical replacement for this branch; maintainers need to choose the landing shape.

Maintainer options:

  1. Refresh and review the QMD guard
    Resolve the dirty branch while preserving the QMD-only skip, then have maintainers explicitly accept the latency-versus-freshness behavior.
  2. Accept prompt empty QMD results
    Maintainers can intentionally accept that QMD zero-hit searches return empty results instead of forcing immediate search-time freshness repair.
  3. Replace with nonblocking refresh
    If immediate QMD freshness repair must remain, replace the skip with a bounded, async, or debounced refresh outside the interactive tool wait.

Next step before merge

  • [P2] The remaining action is maintainer review of the dirty branch, overlapping implementation PRs, and QMD freshness policy rather than an automated replacement fix.

Security
Cleared: No concrete security or supply-chain concern found; the diff is limited to memory-core runtime branching and a colocated regression test.

Review details

Best possible solution:

Land one refreshed canonical memory-core fix that keeps QMD update work out of the interactive zero-hit memory_search deadline while preserving non-QMD retry behavior or replacing QMD repair with bounded/background refresh if immediate freshness must stay.

Do we have a high-confidence way to reproduce the issue?

Yes, source-reproducible: current main and v2026.6.9 still await a forced sync after zero memory_search results, and QMD sync awaits update maintenance. I did not run a live large-corpus QMD timing repro in this read-only pass, but the linked issue and PR body include live QMD behavior evidence.

Is this the best way to solve the issue?

Yes as a narrow mitigation, pending maintainer acceptance of the QMD freshness tradeoff. A bounded or background QMD refresh is the safer alternative if maintainers want to preserve immediate repair without blocking the interactive tool path.

AGENTS.md: found and applied where relevant.

Codex review notes: model internal, reasoning high; reviewed against bda05dbc2f15.

Label changes

Label justifications:

  • P2: The PR targets a focused memory-core availability bug with a linked source-repro issue and limited blast radius.
  • merge-risk: 🚨 compatibility: Skipping QMD's zero-hit forced sync changes fallback behavior for users whose QMD index is stale during search.
  • rating: 🐚 platinum hermit: Overall readiness is 🐚 platinum hermit; proof is 🦞 diamond lobster and patch quality is 🐚 platinum hermit.
  • status: 👀 ready for maintainer look: ClawSweeper has no concrete contributor-facing blocker left for this PR. Sufficient (live_output): The PR body includes after-fix QMD CLI and MCP live output from isolated state showing zero-hit results returning through the QMD backend without the forced update path.
  • proof: sufficient: Contributor real behavior proof is sufficient. The PR body includes after-fix QMD CLI and MCP live output from isolated state showing zero-hit results returning through the QMD backend without the forced update path.
Evidence reviewed

PR surface:

Source +4, Tests +23. Total +27 across 2 files.

View PR surface stats
Area Files Added Removed Net
Source 1 5 1 +4
Tests 1 23 0 +23
Docs 0 0 0 0
Config 0 0 0 0
Generated 0 0 0 0
Other 0 0 0 0
Total 2 28 1 +27

Acceptance criteria:

  • [P1] node scripts/run-vitest.mjs extensions/memory-core/src/tools.test.ts.
  • [P1] node scripts/run-vitest.mjs extensions/memory-core/src/tools.citations.test.ts.
  • [P1] git diff --check.

What I checked:

Likely related people:

  • TurboTheTurtle: Recent merged work kept memory_search in transient QMD mode and touched both the tool and QMD manager surfaces adjacent to this behavior. (role: recent memory_search and QMD lifecycle contributor; confidence: high; commits: ddacb7ba39d4, 6aff1e8f9ea4; files: extensions/memory-core/src/tools.ts, extensions/memory-core/src/memory/qmd-manager.ts)
  • dreamhunter2333: Authored the recent memory_search AbortSignal/deadline work around the same interactive timeout wrapper that this stall path runs under. (role: recent memory_search timeout contributor; confidence: medium; commits: 8d72cb9401e5; files: extensions/memory-core/src/tools.ts, extensions/memory-core/src/tools.test.ts, packages/memory-host-sdk/src/host/types.ts)
  • osolmaz: Recent work changed memory index identity, paused-index retry behavior, and QMD rerank/search configuration near zero-hit and stale-index behavior. (role: adjacent memory index and QMD config contributor; confidence: medium; commits: a4b4fed41287, 0dbf17471b41; files: extensions/memory-core/src/tools.ts, extensions/memory-core/src/memory/qmd-manager.ts, packages/memory-host-sdk/src/host/backend-config.ts)
  • steipete: Recent commits touched QMD search process timeouts, QMD numeric config, and memory recall stall handling near this failure mode. (role: recent memory-core hardening and SDK boundary contributor; confidence: medium; commits: 1ac037d94894, c36ba9ea7a2d, 9e2bd8b2f7eb; files: extensions/memory-core/src/tools.ts, extensions/memory-core/src/memory/qmd-manager.ts, packages/memory-host-sdk/src/host/backend-config.ts)
  • vincentkoc: Recent QMD watcher/probe work and review history are adjacent to QMD readiness, status, and search behavior touched by this PR. (role: recent QMD area contributor; confidence: medium; commits: dbcae5b78d9f, 6aff1e8f9ea4; files: extensions/memory-core/src/memory/qmd-manager.ts, extensions/memory-core/src/tools.ts)
What the crustacean ranks mean
  • 🦀 challenger crab: rare, exceptional readiness with strong proof, clean implementation, and convincing validation.
  • 🦞 diamond lobster: very strong readiness with only minor maintainer review expected.
  • 🐚 platinum hermit: good normal PR, likely mergeable with ordinary maintainer review.
  • 🦐 gold shrimp: useful signal, but proof or patch confidence is still limited.
  • 🦪 silver shellfish: thin signal; proof, validation, or implementation needs work.
  • 🧂 unranked krab: not merge-ready because proof is missing/unusable or there are serious correctness or safety concerns.
  • 🌊 off-meta tidepool: rating does not apply to this item.

Shiny media proof means a screenshot, video, or linked artifact directly shows the changed behavior. Runtime, network, CSP, and security claims still need visible diagnostics.

How this review workflow works
  • ClawSweeper keeps one durable marker-backed review comment per issue or PR.
  • Re-runs edit this comment so the latest verdict, findings, and automation markers stay together instead of adding duplicate bot comments.
  • A fresh review can be triggered by eligible @clawsweeper re-review comments, exact-item GitHub events, scheduled/background review runs, or manual workflow dispatch.
  • PR/issue authors and users with repository write access can comment @clawsweeper re-review or @clawsweeper re-run on an open PR or issue to request a fresh review only.
  • Maintainers can also comment @clawsweeper review to request a fresh review only.
  • Fresh-review commands do not start repair, autofix, rebase, CI repair, or automerge.
  • Maintainer-only repair and merge flows require explicit commands such as @clawsweeper autofix, @clawsweeper automerge, @clawsweeper fix ci, or @clawsweeper address review.
  • Maintainers can comment @clawsweeper explain to ask for more context, or @clawsweeper stop to stop active automation.

@clawsweeper clawsweeper Bot added the rating: 🌊 off-meta tidepool PR readiness rating does not apply to this item. label Jun 3, 2026
@sahibzada-allahyar

sahibzada-allahyar commented Jun 4, 2026

Copy link
Copy Markdown
Contributor Author

@clawsweeper re-review

@clawsweeper

clawsweeper Bot commented Jun 4, 2026 •
edited
Loading

Copy link
Copy Markdown
Contributor

🦞🧹
ClawSweeper re-review requested.

I asked ClawSweeper to review this item again.
Action: item re-review queued (workflow sweep.yml, event repository_dispatch).
Result: the existing ClawSweeper review comment will be edited in place when the review finishes.

Re-review progress:

@clawsweeper clawsweeper Bot added rating: 🦪 silver shellfish Thin PR readiness signal; proof, validation, or implementation needs work. status: 📣 needs proof The PR needs real behavior proof before ClawSweeper can clear the contributor ask. P2 Normal backlog priority with limited blast radius. merge-risk: 🚨 compatibility 🚨 May break existing users, config, migrations, defaults, or upgrade paths. and removed rating: 🌊 off-meta tidepool PR readiness rating does not apply to this item. labels Jun 4, 2026
@sahibzada-allahyar

sahibzada-allahyar commented Jun 4, 2026

Copy link
Copy Markdown
Contributor Author

@clawsweeper re-review

@clawsweeper

clawsweeper Bot commented Jun 4, 2026 •
edited
Loading

Copy link
Copy Markdown
Contributor

🦞👀
ClawSweeper picked this up.

Command router queued. I will update this comment with the next step.

Re-review progress:

@sahibzada-allahyar

sahibzada-allahyar commented Jun 4, 2026

Copy link
Copy Markdown
Contributor Author

@clawsweeper re-review

@clawsweeper

clawsweeper Bot commented Jun 4, 2026

Copy link
Copy Markdown
Contributor

🦞👀
ClawSweeper picked this up.

Command router queued. I will update this comment with the next step.

@sahibzada-allahyar

sahibzada-allahyar commented Jun 4, 2026

Copy link
Copy Markdown
Contributor Author

@clawsweeper re-review

@clawsweeper

clawsweeper Bot commented Jun 4, 2026 •
edited
Loading

Copy link
Copy Markdown
Contributor

🦞👀
ClawSweeper picked this up.

Command router queued. I will update this comment with the next step.

Re-review progress:

@clawsweeper clawsweeper Bot added rating: 🌊 off-meta tidepool PR readiness rating does not apply to this item. proof: sufficient ClawSweeper judged the real behavior proof convincing. rating: 🐚 platinum hermit Good normal PR readiness with ordinary maintainer review expected. status: 👀 ready for maintainer look ClawSweeper has no concrete contributor-facing blocker left for this PR. and removed rating: 🦪 silver shellfish Thin PR readiness signal; proof, validation, or implementation needs work. status: 📣 needs proof The PR needs real behavior proof before ClawSweeper can clear the contributor ask. rating: 🌊 off-meta tidepool PR readiness rating does not apply to this item. labels Jun 4, 2026
@openclaw-barnacle openclaw-barnacle Bot added proof: supplied External PR includes structured after-fix real behavior proof. and removed triage: needs-real-behavior-proof Candidate: external PR needs after-fix proof from a real setup. proof: sufficient ClawSweeper judged the real behavior proof convincing. labels Jun 4, 2026
@clawsweeper clawsweeper Bot added the proof: sufficient ClawSweeper judged the real behavior proof convincing. label Jun 4, 2026
@openclaw-barnacle openclaw-barnacle Bot removed the proof: sufficient ClawSweeper judged the real behavior proof convincing. label Jun 4, 2026
@clawsweeper clawsweeper Bot added the proof: sufficient ClawSweeper judged the real behavior proof convincing. label Jun 4, 2026
@clawsweeper clawsweeper Bot mentioned this pull request Jun 10, 2026
Closed
This was referenced Jun 19, 2026
Closed
Closed
Open
Closed
Open
Closed
Closed
@clawsweeper clawsweeper Bot mentioned this pull request Jul 1, 2026
Closed
@vincentkoc
vincentkoc force-pushed the tui-90023-qmd-zero-hit branch from 6fe705b to 5b3b116 Compare July 1, 2026 11:01
@vincentkoc vincentkoc self-assigned this Jul 1, 2026
@vincentkoc

vincentkoc commented Jul 1, 2026

Copy link
Copy Markdown
Member

Land-ready at 5b3b116e4944e908de406eb522fe92c2237f1ed5.

  • Fix is limited to long-lived QMD zero-hit searches; one-shot CLI bootstrap retry and non-QMD retry behavior remain intact.
  • Focused proof: node scripts/run-vitest.mjs extensions/memory-core/src/tools.test.ts (22 passed).
  • Exact-head changed gate: Blacksmith Testbox tbx_01kwen8xbmyvahz8tj3zdc2bte passed check:changed.
  • Fresh autoreview completed with no accepted/actionable findings after the lifecycle and type-narrowing fixes.
  • Hosted CI run 28512792755 is green. Its initial tooling shard was killed by SIGKILL after the preceding suites passed; rerun job 84520045826 passed unchanged.

This is the canonical fix for #90023. PRs #90117 and #98572 are broader/weaker duplicates and can close after merge.

@vincentkoc
vincentkoc force-pushed the tui-90023-qmd-zero-hit branch from 5b3b116 to 6bc28ac Compare July 1, 2026 11:32
@clawsweeper clawsweeper Bot mentioned this pull request Jul 1, 2026
Closed
1 task
@vincentkoc
vincentkoc force-pushed the tui-90023-qmd-zero-hit branch from 6bc28ac to 4ea0549 Compare July 1, 2026 11:46
@vincentkoc
vincentkoc merged commit 90e31be into openclaw:main Jul 1, 2026
87 checks passed
@steipete steipete mentioned this pull request Jul 9, 2026
Closed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

@vincentkoc vincentkoc

Labels

extensions: memory-core Extension: memory-core merge-risk: 🚨 compatibility 🚨 May break existing users, config, migrations, defaults, or upgrade paths. P2 Normal backlog priority with limited blast radius. proof: sufficient ClawSweeper judged the real behavior proof convincing. proof: supplied External PR includes structured after-fix real behavior proof. rating: 🐚 platinum hermit Good normal PR readiness with ordinary maintainer review expected. size: S status: 👀 ready for maintainer look ClawSweeper has no concrete contributor-facing blocker left for this PR.

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

QMD memory_search zero-hit queries trigger synchronous force sync and stall interactive turns

2 participants

@sahibzada-allahyar @vincentkoc