Persistit r382
The Akiban Server build job hung last week in what seemed to be well exercised Persistit code:
http://172.16.20.104:8080/job/server-build/3211/console
Slightly snipped thread dump of interest:
"main" prio=10 tid=0x0000000040234000 nid=0x4d0e waiting on condition [0x00007f097a572000]
java.lang.Thread.State: WAITING (parking)
at sun.misc.Unsafe.park(Native Method)
- parking to wait for <0x00000000e087ac08> (a java.util.concurrent.locks.ReentrantLock$NonfairSync)
at java.util.concurrent.locks.LockSupport.park(LockSupport.java:156)
at java.util.concurrent.locks.AbstractQueuedSynchronizer.parkAndCheckInterrupt(AbstractQueuedSynchronizer.java:811)
at java.util.concurrent.locks.AbstractQueuedSynchronizer.acquireQueued(AbstractQueuedSynchronizer.java:842)
at java.util.concurrent.locks.AbstractQueuedSynchronizer.acquire(AbstractQueuedSynchronizer.java:1178)
at java.util.concurrent.locks.ReentrantLock$NonfairSync.lock(ReentrantLock.java:186)
at java.util.concurrent.locks.ReentrantLock.lock(ReentrantLock.java:262)
at java.util.concurrent.ArrayBlockingQueue.clear(ArrayBlockingQueue.java:572)
at com.persistit.CleanupManager.clear(CleanupManager.java:194)
- locked <0x00000000e087ac60> (a com.persistit.CleanupManager)
at com.persistit.Persistit.releaseAllResources(Persistit.java:1843)
at com.persistit.Persistit.crash(Persistit.java:1795)
at com.akiban.server.service.tree.TreeServiceImpl.crash(TreeServiceImpl.java:280)
Comments from Peter:
Well the IOTaskRunnable#crash method calls deprecated Thread#stop(). It's deprecated because it can leave monitors in inconsistent states. I suspect that's what happened here. Perhaps we can modify that crash method to try interrupting the Thread a few times before calling stop it, so yes, I guess we should file a low-priority bug.
And I believe this is the only time this type of hang as been observed. As such, marking as low priority.
