ByteBuffersDirectory.rename ends with a leftover unconditional files.remove(source), so a file that another thread creates (createOutput) or renames onto the source name while the rename runs is silently deleted, although that thread's call returned normally

Finding LFV-836Area: store / ByteBuffersDirectory (rename vs. a concurrent createOutput or rename onto the source name)Priority: lowImpact: data corruptionNovelty: new

Finding page for LFV-836

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

What it shows

You share one ByteBuffersDirectory between two threads: one renames file 'a' to a new name while the other creates a new 'a' (createOutput, write, close) or renames a temp file onto 'a', and you expect the new 'a' to stay until someone deletes or renames it, as it does on FSDirectory. Instead, now and then the new file vanishes: rename has already moved its own entry with the checked files.remove(source, file) and then runs a leftover unconditional files.remove(source), which can only remove a different entry, the one the other thread has just put under 'a'. Neither call reports an error; afterwards listAll lacks the file and openInput throws NoSuchFileException (stress tests lose between 1 and a few hundred files per 100,000 rounds on every run).

The test shows that ByteBuffersDirectory.rename("a", "b") deletes a new file "a" created right after its checked files.remove(source, file), through the leftover files.remove(source); the test forces that order by swapping the private files map for one that runs createOutput("a") at that point (what another thread's createOutput does in the race).

Expected vs actual

Expected listAll() = [a, b] (nobody deleted the new 'a'). Actual: [b].

Files

src/org/apache/lucene/store/TestByteBuffersDirectoryRenameDropsNewSource.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.
This repro reads private Lucene state by reflection (setAccessible), so it depends on internals that may change.

# 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.store.TestByteBuffersDirectoryRenameDropsNewSource

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: bytebuffers-output-block-expansion-recycler-and-view-aliasing

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.