Concurrent HNSW build retries a lost entry-node promotion with the wrong query vector

Finding LFV-061Area: search / HNSWPriority: lowImpact: silent wrong resultsNovelty: new

Finding page for LFV-061

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

What it shows

You build or merge HNSW graphs with more than one merge worker. You expect each node's neighbours to be scored against that node's vector. Instead, when an entry-node promotion is lost to another worker and retried, the shared scorer is not reset, so wrong scores are stored and the persisted topology changes.

The test shows that when two HnswConcurrentMergeBuilder workers both try to promote a new entry node and worker 0 loses, worker 0's retry of the new upper levels scores with its shared scorer still pointed at another node (node 0), not at the node it inserts (node 1).

Expected vs actual

Expected every stored neighbour score to be sim(owner, neighbour), e.g. level 1: 1 -> 3 = 0.027027028 (1/37). Actual: 0.02 (1/50 = sim(0, 3)).

Files

• src/org/apache/lucene/util/hnsw/TestHnswConcurrentMergeRetryScorer.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.util.hnsw.TestHnswConcurrentMergeRetryScorer

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.