[ 
https://issues.apache.org/jira/browse/AMQ-4274?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Timothy Bish resolved AMQ-4274.
-------------------------------

    Resolution: Fixed

fixed on trunk.
                
> Potential deadlock between FailoverTransport and AbstractInactivityMonitor
> --------------------------------------------------------------------------
>
>                 Key: AMQ-4274
>                 URL: https://issues.apache.org/jira/browse/AMQ-4274
>             Project: ActiveMQ
>          Issue Type: Bug
>          Components: Transport
>    Affects Versions: 5.7.0
>            Reporter: Timothy Bish
>            Assignee: Timothy Bish
>             Fix For: 5.8.0
>
>
> Its possible for an operation that's doing a send via oneway in 
> FailoverTransport to deadlock with a Keep Alive write in the Inactivity 
> Monitor.  
> {code}
> Found one Java-level deadlock:
> =============================
> "U.GeoCodingIncBuilder.1":
>   waiting for ownable synchronizer 0x00002aaac04e29e8, (a 
> java.util.concurrent.locks.ReentrantLock$NonfairSync),
>   which is held by "ActiveMQ Session Task-42904"
> "ActiveMQ Session Task-42904":
>   waiting for ownable synchronizer 0x00002ab3797e7348, (a 
> java.util.concurrent.locks.ReentrantReadWriteLock$NonfairSync),
>   which is held by "ActiveMQ InactivityMonitor Worker"
> "ActiveMQ InactivityMonitor Worker":
>   waiting to lock monitor 0x00002ab729f36a70 (object 0x00002aaac04f11d8, a 
> java.lang.Object),
>   which is held by "ActiveMQ Session Task-42904"
> Java stack information for the threads listed above:
> ===================================================
> "U.GeoCodingIncBuilder.1":
>       at sun.misc.Unsafe.park(Native Method)
>       - parking to wait for  <0x00002aaac04e29e8> (a 
> java.util.concurrent.locks.ReentrantLock$NonfairSync)
>       at java.util.concurrent.locks.LockSupport.park(LockSupport.java:158)
>       at 
> java.util.concurrent.locks.AbstractQueuedSynchronizer.parkAndCheckInterrupt(AbstractQueuedSynchronizer.java:747)
>       at 
> java.util.concurrent.locks.AbstractQueuedSynchronizer.acquireQueued(AbstractQueuedSynchronizer.java:778)
>       at 
> java.util.concurrent.locks.AbstractQueuedSynchronizer.acquire(AbstractQueuedSynchronizer.java:1114)
>       at 
> java.util.concurrent.locks.ReentrantLock$NonfairSync.lock(ReentrantLock.java:186)
>       at java.util.concurrent.locks.ReentrantLock.lock(ReentrantLock.java:262)
>       at 
> org.apache.activemq.transport.MutexTransport.oneway(MutexTransport.java:66)
>       at 
> org.apache.activemq.transport.ResponseCorrelator.oneway(ResponseCorrelator.java:60)
>       at 
> org.apache.activemq.ActiveMQConnection.doAsyncSendPacket(ActiveMQConnection.java:1290)
>       at 
> org.apache.activemq.ActiveMQConnection.asyncSendPacket(ActiveMQConnection.java:1284)
>       at org.apache.activemq.ActiveMQSession.<init>(ActiveMQSession.java:252)
>       at 
> org.apache.activemq.ActiveMQConnection.createSession(ActiveMQConnection.java:332)
>       at 
> linqmap.ipc.impl.jms.SingleJmsFactory.createSession(SingleJmsFactory.java:492)
>       at 
> linqmap.ipc.impl.jms.SingleJmsFactory.createSendSession(SingleJmsFactory.java:318)
>       at linqmap.ipc.impl.jms.JmsQueue.send(JmsQueue.java:117)
>       at linqmap.ipc.queues.wrappers.AsyncSender$2.run(AsyncSender.java:81)
>       at 
> java.util.concurrent.ThreadPoolExecutor$Worker.runTask(ThreadPoolExecutor.java:885)
>       at 
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:907)
>       at java.lang.Thread.run(Thread.java:619)
> "ActiveMQ Session Task-42904":
>       at sun.misc.Unsafe.park(Native Method)
>       - parking to wait for  <0x00002ab3797e7348> (a 
> java.util.concurrent.locks.ReentrantReadWriteLock$NonfairSync)
>       at java.util.concurrent.locks.LockSupport.park(LockSupport.java:158)
>       at 
> java.util.concurrent.locks.AbstractQueuedSynchronizer.parkAndCheckInterrupt(AbstractQueuedSynchronizer.java:747)
>       at 
> java.util.concurrent.locks.AbstractQueuedSynchronizer.doAcquireShared(AbstractQueuedSynchronizer.java:877)
>       at 
> java.util.concurrent.locks.AbstractQueuedSynchronizer.acquireShared(AbstractQueuedSynchronizer.java:1197)
>       at 
> java.util.concurrent.locks.ReentrantReadWriteLock$ReadLock.lock(ReentrantReadWriteLock.java:594)
>       at 
> org.apache.activemq.transport.AbstractInactivityMonitor.oneway(AbstractInactivityMonitor.java:268)
>       at 
> org.apache.activemq.transport.TransportFilter.oneway(TransportFilter.java:85)
>       at 
> org.apache.activemq.transport.WireFormatNegotiator.oneway(WireFormatNegotiator.java:104)
>       at 
> org.apache.activemq.transport.failover.FailoverTransport.oneway(FailoverTransport.java:640)
>       - locked <0x00002aaac04f11d8> (a java.lang.Object)
>       at 
> org.apache.activemq.transport.MutexTransport.oneway(MutexTransport.java:68)
>       at 
> org.apache.activemq.transport.ResponseCorrelator.oneway(ResponseCorrelator.java:60)
>       at 
> org.apache.activemq.ActiveMQConnection.doAsyncSendPacket(ActiveMQConnection.java:1290)
>       at 
> org.apache.activemq.ActiveMQConnection.asyncSendPacket(ActiveMQConnection.java:1284)
>       at 
> org.apache.activemq.ActiveMQSession.asyncSendPacket(ActiveMQSession.java:1898)
>       at 
> org.apache.activemq.ActiveMQSession.sendAck(ActiveMQSession.java:2064)
>       at 
> org.apache.activemq.ActiveMQSession.sendAck(ActiveMQSession.java:2059)
>       at 
> org.apache.activemq.ActiveMQMessageConsumer.acknowledge(ActiveMQMessageConsumer.java:1061)
>       - locked <0x00002aaac06c7280> (a java.util.LinkedList)
>       at 
> org.apache.activemq.ActiveMQSession.acknowledge(ActiveMQSession.java:1604)
>       at 
> org.apache.activemq.ActiveMQMessageConsumer$1.execute(ActiveMQMessageConsumer.java:552)
>       at 
> org.apache.activemq.command.ActiveMQMessage.acknowledge(ActiveMQMessage.java:97)
>       at linqmap.ipc.impl.jms.JmsQueue.onMessage(JmsQueue.java:262)
>       at 
> org.apache.activemq.ActiveMQMessageConsumer.dispatch(ActiveMQMessageConsumer.java:1321)
>       - locked <0x00002aaac06c0b70> (a java.lang.Object)
>       at 
> org.apache.activemq.ActiveMQSessionExecutor.dispatch(ActiveMQSessionExecutor.java:131)
>       at 
> org.apache.activemq.ActiveMQSessionExecutor.iterate(ActiveMQSessionExecutor.java:202)
>       at 
> org.apache.activemq.thread.PooledTaskRunner.runTask(PooledTaskRunner.java:129)
>       at 
> org.apache.activemq.thread.PooledTaskRunner$1.run(PooledTaskRunner.java:47)
>       at 
> java.util.concurrent.ThreadPoolExecutor$Worker.runTask(ThreadPoolExecutor.java:885)
>       at 
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:907)
>       at java.lang.Thread.run(Thread.java:619)
> "ActiveMQ InactivityMonitor Worker":
>       at 
> org.apache.activemq.transport.failover.FailoverTransport.handleTransportFailure(FailoverTransport.java:252)
>       - waiting to lock <0x00002aaac04f11d8> (a java.lang.Object)
>       at 
> org.apache.activemq.transport.failover.FailoverTransport$3.onException(FailoverTransport.java:209)
>       at 
> org.apache.activemq.transport.TransportFilter.onException(TransportFilter.java:101)
>       at 
> org.apache.activemq.transport.WireFormatNegotiator.onException(WireFormatNegotiator.java:160)
>       at 
> org.apache.activemq.transport.AbstractInactivityMonitor.onException(AbstractInactivityMonitor.java:295)
>       at 
> org.apache.activemq.transport.AbstractInactivityMonitor$3.run(AbstractInactivityMonitor.java:168)
>       at 
> java.util.concurrent.ThreadPoolExecutor$Worker.runTask(ThreadPoolExecutor.java:885)
>       at 
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:907)
>       at java.lang.Thread.run(Thread.java:619)
> Found 1 deadlock.
> {code}
> The deadlock occurs when the write check task detects a failure and calls the 
> onException method while holding the write lock side of the RW lock in the 
> monitor.  Since the FailoverTransport holds its reconnect lock for the 
> duration of the oneway call and the onException method of failover transport 
> tries to lock that same mutex things can lock if the oneway call was waiting 
> on the read lock side of the monitors R/W lock. 
> The solution is to ensure that we always unlock the writelock before we call 
> the next transports onException method but after we've set the failed flag so 
> that any waiting oneway calls will fail and throw their IOException 
> indicating the transport has already failed.  This will free the failover 
> transport up to do its normal failure recovery processing.  

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

Reply via email to