Uploaded image for project: 'Cassandra'
  1. Cassandra
  2. CASSANDRA-8474

Error in commit log allocator thread

    XMLWordPrintableJSON

Details

    • Bug
    • Status: Resolved
    • Low
    • Resolution: Fixed
    • 2.1.3
    • None
    • None

    Description

      The dtest counters_test.py:TestCounters.upgrade_test is intermittently failing on 2.1-HEAD because of the following error in the system.log

      ERROR [COMMIT-LOG-ALLOCATOR] 2014-12-12 14:18:07,342 StorageService.java:366 - Stopping gossiper
      WARN  [COMMIT-LOG-ALLOCATOR] 2014-12-12 14:18:07,342 StorageService.java:274 - Stopping gossip by operato
      r request
      INFO  [COMMIT-LOG-ALLOCATOR] 2014-12-12 14:18:07,342 Gossiper.java:1341 - Announcing shutdown
      ERROR [COMMIT-LOG-ALLOCATOR] 2014-12-12 14:18:09,349 CassandraDaemon.java:170 - Exception in thread Threa
      d[COMMIT-LOG-ALLOCATOR,5,main]
      java.lang.AssertionError: java.lang.InterruptedException
              at org.apache.cassandra.net.OutboundTcpConnection.enqueue(OutboundTcpConnection.java:107) ~[main/
      :na]
              at org.apache.cassandra.net.MessagingService.sendOneWay(MessagingService.java:682) ~[main/:na]
              at org.apache.cassandra.net.MessagingService.sendOneWay(MessagingService.java:648) ~[main/:na]
              at org.apache.cassandra.gms.Gossiper.stop(Gossiper.java:1345) ~[main/:na]
              at org.apache.cassandra.service.StorageService.stopGossiping(StorageService.java:275) ~[main/:na]
              at org.apache.cassandra.service.StorageService.stopTransports(StorageService.java:367) ~[main/:na
      ]
              at org.apache.cassandra.db.commitlog.CommitLog.handleCommitError(CommitLog.java:365) ~[main/:na]
              at org.apache.cassandra.db.commitlog.CommitLogSegmentManager$1.runMayThrow(CommitLogSegmentManage
      r.java:164) ~[main/:na]
              at org.apache.cassandra.utils.WrappedRunnable.run(WrappedRunnable.java:28) ~[main/:na]
              at java.lang.Thread.run(Thread.java:745) ~[na:1.7.0_67]
      Caused by: java.lang.InterruptedException: null
              at java.util.concurrent.locks.AbstractQueuedSynchronizer.acquireInterruptibly(AbstractQueuedSynch
      ronizer.java:1219) ~[na:1.7.0_67]
              at java.util.concurrent.locks.ReentrantLock.lockInterruptibly(ReentrantLock.java:340) ~[na:1.7.0_
      67]
              at java.util.concurrent.LinkedBlockingQueue.put(LinkedBlockingQueue.java:338) ~[na:1.7.0_67]
              at org.apache.cassandra.net.OutboundTcpConnection.enqueue(OutboundTcpConnection.java:103) ~[main/:na]
              ... 9 common frames omitted
      

      I have attached the system.log files of both nodes used, as well as the test being run.

      I don't see this problem at all while running against 2.0-HEAD. I can reproduce this very far back into 2.1's history. With 2.1.0-rc2, I see

      ERROR [COMMIT-LOG-ALLOCATOR] 2014-12-12 15:21:04,069 CassandraDaemon.java:166 - Exception in thread Thread[COMMIT-LOG-ALLOCATOR,5,main]
      org.apache.cassandra.io.FSWriteError: java.nio.channels.ClosedByInterruptException
              at org.apache.cassandra.db.commitlog.CommitLogSegment.<init>(CommitLogSegment.java:178) ~[main/:na]
              at org.apache.cassandra.db.commitlog.CommitLogSegment.recycle(CommitLogSegment.java:373) ~[main/:na]
              at org.apache.cassandra.db.commitlog.CommitLogSegmentManager$3.call(CommitLogSegmentManager.java:334) ~[main/:na]
              at org.apache.cassandra.db.commitlog.CommitLogSegmentManager$3.call(CommitLogSegmentManager.java:331) ~[main/:na]
              at org.apache.cassandra.db.commitlog.CommitLogSegmentManager$1.runMayThrow(CommitLogSegmentManager.java:148) ~[main/:na]
              at org.apache.cassandra.utils.WrappedRunnable.run(WrappedRunnable.java:28) ~[main/:na]
              at java.lang.Thread.run(Thread.java:745) ~[na:1.7.0_67]
      Caused by: java.nio.channels.ClosedByInterruptException: null
              at java.nio.channels.spi.AbstractInterruptibleChannel.end(AbstractInterruptibleChannel.java:202) ~[na:1.7.0_67]
              at sun.nio.ch.FileChannelImpl.map(FileChannelImpl.java:919) ~[na:1.7.0_67]
              at org.apache.cassandra.db.commitlog.CommitLogSegment.<init>(CommitLogSegment.java:167) ~[main/:na]
              ... 6 common frames omitted

      Is there something we're doing wrong in the test?

      Attachments

        1. 8474.txt
          1.0 kB
          Benedict Elliott Smith
        2. counter_tests.py
          10 kB
          Philip Thompson
        3. node2.log
          186 kB
          Philip Thompson
        4. node1.log
          186 kB
          Philip Thompson

        Activity

          People

            benedict Benedict Elliott Smith
            philipthompson Philip Thompson
            Benedict Elliott Smith
            Philip Thompson Philip Thompson
            Votes:
            0 Vote for this issue
            Watchers:
            1 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved: