Details
-
Bug
-
Status: Closed
-
Major
-
Resolution: Fixed
-
qpid-java-6.0
-
None
Description
JE Transaction aborted in the middle of commit in Replicated Environment can cause abrupt Broker shutdown
######################################################################## # # Unhandled Exception java.lang.IllegalStateException: Transaction Id -42 has been closed. in Thread IO-/127.0.0.1:59532 # # Exiting # ######################################################################## java.lang.IllegalStateException: Transaction Id -42 has been closed. at com.sleepycat.je.Transaction.checkOpen(Transaction.java:894) at com.sleepycat.je.Transaction.doCommit(Transaction.java:592) at com.sleepycat.je.Transaction.commit(Transaction.java:410) at org.apache.qpid.server.store.berkeleydb.replication.ReplicatedEnvironmentFacade.commitAsync(ReplicatedEnvironmentFacade.java:302) at org.apache.qpid.server.store.berkeleydb.AbstractBDBMessageStore.commitTranAsyncImpl(AbstractBDBMessageStore.java:794) at org.apache.qpid.server.store.berkeleydb.AbstractBDBMessageStore.access$1200(AbstractBDBMessageStore.java:74) at org.apache.qpid.server.store.berkeleydb.AbstractBDBMessageStore$BDBTransaction.commitTranAsync(AbstractBDBMessageStore.java:1364) at org.apache.qpid.server.txn.LocalTransaction.commitAsync(LocalTransaction.java:399) at org.apache.qpid.server.protocol.v0_8.AMQChannel.commit(AMQChannel.java:1220) at org.apache.qpid.server.protocol.v0_8.AMQChannel.receiveTxCommit(AMQChannel.java:3622) at org.apache.qpid.codec.ServerDecoder.processMethod(ServerDecoder.java:228) at org.apache.qpid.codec.AMQDecoder.processFrame(AMQDecoder.java:191) at org.apache.qpid.server.protocol.v0_8.BrokerDecoder.doProcessFrame(BrokerDecoder.java:114) at org.apache.qpid.server.protocol.v0_8.BrokerDecoder.access$000(BrokerDecoder.java:36) at org.apache.qpid.server.protocol.v0_8.BrokerDecoder$1.run(BrokerDecoder.java:78) at org.apache.qpid.server.protocol.v0_8.BrokerDecoder$1.run(BrokerDecoder.java:74) at java.security.AccessController.doPrivileged(Native Method) at org.apache.qpid.server.protocol.v0_8.BrokerDecoder.processFrame(BrokerDecoder.java:73) at org.apache.qpid.codec.AMQDecoder.processInput(AMQDecoder.java:173) at org.apache.qpid.codec.AMQDecoder.decode(AMQDecoder.java:114) at org.apache.qpid.codec.ServerDecoder.decodeBuffer(ServerDecoder.java:43) at org.apache.qpid.server.protocol.v0_8.AMQPConnection_0_8$1.run(AMQPConnection_0_8.java:266) at org.apache.qpid.server.protocol.v0_8.AMQPConnection_0_8$1.run(AMQPConnection_0_8.java:258) at java.security.AccessController.doPrivileged(Native Method) at org.apache.qpid.server.protocol.v0_8.AMQPConnection_0_8.received(AMQPConnection_0_8.java:257) at org.apache.qpid.server.transport.MultiVersionProtocolEngine.received(MultiVersionProtocolEngine.java:142) at org.apache.qpid.server.transport.NonBlockingConnection.processAmqpData(NonBlockingConnection.java:547) at org.apache.qpid.server.transport.NonBlockingConnectionPlainDelegate.processData(NonBlockingConnectionPlainDelegate.java:58) at org.apache.qpid.server.transport.NonBlockingConnection.doRead(NonBlockingConnection.java:446) at org.apache.qpid.server.transport.NonBlockingConnection.doWork(NonBlockingConnection.java:253) at org.apache.qpid.server.transport.NetworkConnectionScheduler.processConnection(NetworkConnectionScheduler.java:108) at org.apache.qpid.server.transport.SelectorThread$ConnectionProcessor.processConnection(SelectorThread.java:499) at org.apache.qpid.server.transport.SelectorThread$SelectionTask.performSelect(SelectorThread.java:337) at org.apache.qpid.server.transport.SelectorThread$SelectionTask.run(SelectorThread.java:86) at org.apache.qpid.server.transport.SelectorThread.run(SelectorThread.java:457) at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145) at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615) at java.lang.Thread.run(Thread.java:745)
In my tests syncing data to disk in commit thread failed (due to majority lost) causing abort of pending commit jobs. IllegalStateException was reported IO-Threads waiting for commit finish.
try-catch blocks in ReplicatedEnvironmentFacade#commit and ReplicatedEnvironmentFacade#commitAsync do not handle IllegalStateException on environment restart as in the rest of the code.
Cathcing and handling all RuntimeExceptions should fix the issue