[OAK-3098] CopyOnWrite might block Async indexer thread indefinitely

• Type: Bug
• Status: Closed
• Priority: Minor
• Resolution: Fixed
• Fix Version/s: 1.0.18, 1.2.3, 1.3.3, 1.4
• Component/s: lucene
• Assignee: Chetan Mehrotra
• Reporter: Chetan Mehrotra
• Created: 13/Jul/15 16:50
• Updated: 02/Mar/16 11:55
• Resolved: 15/Jul/15 11:36

Description

CopyOnWrite logic uses a CountDownLatch to sync up on close call. In case of config change which might lead to closing of the the IndexCopier this latch might not be released causing the async index thread to hang

      sun.misc.Unsafe.park(Native Method)
      java.util.concurrent.locks.LockSupport.park(LockSupport.java:186)
      java.util.concurrent.locks.AbstractQueuedSynchronizer.parkAndCheckInterrupt(AbstractQueuedSynchronizer.java:834)
      java.util.concurrent.locks.AbstractQueuedSynchronizer.doAcquireSharedInterruptibly(AbstractQueuedSynchronizer.java:994)
      java.util.concurrent.locks.AbstractQueuedSynchronizer.acquireSharedInterruptibly(AbstractQueuedSynchronizer.java:1303)
      java.util.concurrent.CountDownLatch.await(CountDownLatch.java:236)
      org.apache.jackrabbit.oak.plugins.index.lucene.IndexCopier$CopyOnWriteDirectory.close(IndexCopier.java:648)
      org.apache.jackrabbit.oak.plugins.index.lucene.LuceneIndexEditorContext.closeWriter(LuceneIndexEditorContext.java:196)
      org.apache.jackrabbit.oak.plugins.index.lucene.LuceneIndexEditor.leave(LuceneIndexEditor.java:207)
      org.apache.jackrabbit.oak.plugins.index.IndexUpdate.leave(IndexUpdate.java:219)
      org.apache.jackrabbit.oak.spi.commit.VisibleEditor.leave(VisibleEditor.java:63)
      org.apache.jackrabbit.oak.spi.commit.EditorDiff.process(EditorDiff.java:56)
      org.apache.jackrabbit.oak.plugins.index.AsyncIndexUpdate.updateIndex(AsyncIndexUpdate.java:366)
      org.apache.jackrabbit.oak.plugins.index.AsyncIndexUpdate.run(AsyncIndexUpdate.java:311)
      org.apache.sling.commons.scheduler.impl.QuartzJobExecutor.execute(QuartzJobExecutor.java:105)
      org.quartz.core.JobRunShell.run(JobRunShell.java:207)
      java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
      java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
      java.lang.Thread.run(Thread.java:745)

Ideally this latch should have a timeout or closing logic of IndexCopier ensures that this is released