[CASSANDRA-20119] Some benchmark tests leak LocalLog-related thread

• Type: Bug
• Status: Triage Needed
• Priority: Normal
• Resolution: Unresolved
• Platform: All
• Assignee: Unassigned
• Reporter: Stefan Miklosovic
• Created: 27/Nov/24 17:02
• Updated: 27/Nov/24 17:03

Description

I am running a benchmark, like CacheLoaderBench and there are these messages when iteration is over

<JMH had finished, but forked VM did not exit, are there stray running threads? Waiting 9 seconds more...>

Non-finished threads:

Thread[GlobalLogFollower,5,system]
  at java.base@11.0.12/jdk.internal.misc.Unsafe.park(Native Method)
  at java.base@11.0.12/java.util.concurrent.locks.LockSupport.park(LockSupport.java:323)
  at app//org.apache.cassandra.utils.concurrent.WaitQueue$Standard$AbstractSignal.await(WaitQueue.java:290)
  at app//org.apache.cassandra.utils.concurrent.WaitQueue$Standard$AbstractSignal.await(WaitQueue.java:283)
  at app//org.apache.cassandra.tcm.log.LocalLog$Async$AsyncRunnable.run(LocalLog.java:816)
  at app//org.apache.cassandra.concurrent.InfiniteLoopExecutor.loop(InfiniteLoopExecutor.java:121)
  at app//org.apache.cassandra.concurrent.InfiniteLoopExecutor$$Lambda$233/0x0000000800393840.run(Unknown Source)
  at app//io.netty.util.concurrent.FastThreadLocalRunnable.run(FastThreadLocalRunnable.java:30)
  at java.base@11.0.12/java.lang.Thread.run(Thread.java:829)
Thread[DestroyJavaVM,5,main]

I noticed this is its teardown

    @TearDown(Level.Trial)
    public void teardown()
    {
        CQLTester.tearDownClass();
        CommitLog.instance.stopUnsafe(true);
        CQLTester.cleanup();
    }

when I do this it will be OK

   @TearDown(Level.Trial)
    public void teardown()
    {
        CQLTester.tearDownClass();
        CommitLog.instance.stopUnsafe(true);
        ClusterMetadataService.instance().log().close();
        CQLTester.cleanup();
    }

Worth to say that benchark eventually finishes. It is not stuck, JMH just waits and checks for 30 seconds and then it seems to "kill it".