HNSW merge-time graph reuse can leave the entry node with no neighbours, dropping kNN recall to zero

Finding LFV-005Area: search / HNSWPriority: highImpact: silent wrong resultsNovelty: new

Finding page for LFV-005

AI-assisted finding, confirmed by automated checks, not yet triaged by a human Lucene committer. This site files nothing upstream.

What it shows

You delete a cluster of documents around the HNSW entry node and merge, for example with forceMerge(1), so the merge reuses and repairs the biggest segment's graph. You expect kNN recall to stay about the same. Instead the repair cannot reconnect an entry node that lost every neighbour, every search starts at that isolated node and returns only it (recall@10 0.906 to 0.000 in the repro), and CheckIndex reports no problem.

The test shows that with the default codec, deleting the 31-doc cluster that the HNSW entry node (doc 171) links into and calling forceMerge(1) reuses the graph but leaves the entry with no neighbours, so kNN search returns only that node.

Expected vs actual

Expected: a k=10 KnnFloatVectorQuery returns 10 hits among the 1419 live docs. Actual (-ea and -da): 1 hit (doc 171).

Files

• src/org/apache/lucene/index/TestHnswMergeReuseIsolatedEntry.java (JUnit test)

Run it

• Tested on Lucene main e1faeba08bde (2026-11-02): still reproduces.
• Tested on Lucene main 645d4fe14a20 (2026-10-28): reproduced.
• Recorded outcome of run 1 on Lucene main 645d4fe14a20: 1 test(s) run, 1 failed.
• How to read the outcome: a failing test is the finding reproducing (the failure message states expected against actual).
• Seeds: a rerun with the same seed replays the same random choices. Assertions off (-da -dsa -Dtests.asserts=false) is the production-like setting.

# Lucene checkout at the tested commit, built with: ./gradlew --no-daemon assemble   (JDK 25)
LUCENE=/path/to/lucene; V=12.0.0-SNAPSHOT
CP=$LUCENE/lucene/core/build/libs/lucene-core-$V.jar
CP=$CP:$LUCENE/lucene/codecs/build/libs/lucene-codecs-$V.jar
CP=$CP:$LUCENE/lucene/test-framework/build/libs/lucene-test-framework-$V.jar
CP=$CP:randomizedtesting-runner-2.9.2.jar:junit-4.13.2.jar:hamcrest-3.0.jar   # test-framework dependencies (Maven Central)
CP=$CP:junit-platform-console-standalone-6.1.3.jar   # JUnit Platform console launcher (Maven Central)

javac -d classes -cp "$CP" $(find src -name '*.java')

# run 1: assertions on, seed DEADBEEF
java -ea -esa --add-modules jdk.incubator.vector --enable-native-access=ALL-UNNAMED -Dtests.seed=DEADBEEF \
     -cp "classes:$CP" org.junit.platform.console.ConsoleLauncher \
     execute --select-class org.apache.lucene.index.TestHnswMergeReuseIsolatedEntry

Gradle alternative: copy the test into the matching module's src/test/ tree of the checkout and run ./gradlew -p lucene/<module> test --tests <class> -Ptests.seed=<seed> (add -Ptests.asserts=false for a run without assertions).

Formal models and scripts this finding's part of the hunt used: study-hnsw-concurrent-merge

Feedback

Open an issue in lucene-ai-bug-hunt. Please do not open issues or pull requests in apache/lucene about this finding unless a Lucene committer agrees it should be filed.

Snapshot of 2026-11-04. AI-assisted; not yet triaged by a human Lucene committer. Feedback: open an issue in lucene-ai-bug-hunt.