[JBoss JIRA] Created: (ISPN-819) Deadlock detection seem to not detect deadlock case between two nodes on the same cache.
11:51 a.m.
Deadlock detection seem to not detect deadlock case between two nodes on the same cache.
----------------------------------------------------------------------------------------
Key: ISPN-819
URL: https://jira.jboss.org/browse/ISPN-819
Project: Infinispan
Issue Type: Bug
Components: Locking and Concurrency
Affects Versions: 4.2.0.CR3
Environment: Windows XP, 4.2.0-CR3
Reporter: Thomas Jean
Assignee: Manik Surtani
Hello,
after activating deadlock detection with 500 ms of spin duration, i launch a classic
deadlock scenario.
Node 1 acquire local lock.
Node 1 try to acquire cluster wide lock.
Node 2 acquire local lock.
Node 2 try to acquire cluster wide lock.
Then deadlock is not detected and the TimeoutException occurs:
org.infinispan.util.concurrent.TimeoutException: Unable to acquire lock after [20 seconds]
on key [counter_key] for requestor [DldGlobalTrans
action{coinToss=3136454835479173852, isMarkedForRollback=false, lockIntention=counter_key,
affectedKeys=[counter_key], locksAtOrigin=[]} Glob
alTransaction:<nodeDeadLock-11465>:5:local]! Lock held by
[DldGlobalTransaction{coinToss=3164596253027655679, isMarkedForRollback=false, lock
Intention=null, affectedKeys=[], locksAtOrigin=[]}
GlobalTransaction:<nodeDeadLock-16603>:1:remote].
I give my configuration file for ISPN and a Class TwoNodesDeadlocks.java to reproduce the
problem.
--
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
11:54 a.m.
New subject: [JBoss JIRA] Updated: (ISPN-819) Deadlock detection seem to not detect deadlock case between two nodes on the same cache.
[
https://jira.jboss.org/browse/ISPN-819?page=com.atlassian.jira.plugin.sys...
]
Thomas Jean updated ISPN-819:
-----------------------------
Attachment: nodeDeadLock-45547.log
nodeDeadLock-43429.log
AgentManager_InfiniSpan_Config_Generic.xml
log4j_Generic.properties
TwoNodesDeadlocks.java
Here are the file to reproduce the problem
Deadlock detection seem to not detect deadlock case between two nodes
on the same cache.
----------------------------------------------------------------------------------------
Key: ISPN-819
URL: https://jira.jboss.org/browse/ISPN-819
Project: Infinispan
Issue Type: Bug
Components: Locking and Concurrency
Affects Versions: 4.2.0.CR3
Environment: Windows XP, 4.2.0-CR3
Reporter: Thomas Jean
Assignee: Manik Surtani
Attachments: AgentManager_InfiniSpan_Config_Generic.xml,
log4j_Generic.properties, nodeDeadLock-43429.log, nodeDeadLock-45547.log,
TwoNodesDeadlocks.java
Hello,
after activating deadlock detection with 500 ms of spin duration, i launch a classic
deadlock scenario.
Node 1 acquire local lock.
Node 1 try to acquire cluster wide lock.
Node 2 acquire local lock.
Node 2 try to acquire cluster wide lock.
Then deadlock is not detected and the TimeoutException occurs:
org.infinispan.util.concurrent.TimeoutException: Unable to acquire lock after [20
seconds] on key [counter_key] for requestor [DldGlobalTrans
action{coinToss=3136454835479173852, isMarkedForRollback=false,
lockIntention=counter_key, affectedKeys=[counter_key], locksAtOrigin=[]} Glob
alTransaction:<nodeDeadLock-11465>:5:local]! Lock held by
[DldGlobalTransaction{coinToss=3164596253027655679, isMarkedForRollback=false, lock
Intention=null, affectedKeys=[], locksAtOrigin=[]}
GlobalTransaction:<nodeDeadLock-16603>:1:remote].
I give my configuration file for ISPN and a Class TwoNodesDeadlocks.java to reproduce
the problem.
--
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
12:13 p.m.
New subject: [JBoss JIRA] Updated: (ISPN-819) Deadlock detection seem to not detect deadlock case between two nodes on the same cache.
[
https://jira.jboss.org/browse/ISPN-819?page=com.atlassian.jira.plugin.sys...
]
Manik Surtani updated ISPN-819:
-------------------------------
Fix Version/s: 4.2.0.CR4
4.2.0.Final
Priority: Critical (was: Major)
Deadlock detection seem to not detect deadlock case between two nodes
on the same cache.
----------------------------------------------------------------------------------------
Key: ISPN-819
URL: https://jira.jboss.org/browse/ISPN-819
Project: Infinispan
Issue Type: Bug
Components: Locking and Concurrency
Affects Versions: 4.2.0.CR3
Environment: Windows XP, 4.2.0-CR3
Reporter: Thomas Jean
Assignee: Manik Surtani
Priority: Critical
Fix For: 4.2.0.CR4, 4.2.0.Final
Attachments: AgentManager_InfiniSpan_Config_Generic.xml,
log4j_Generic.properties, nodeDeadLock-43429.log, nodeDeadLock-45547.log,
TwoNodesDeadlocks.java
Hello,
after activating deadlock detection with 500 ms of spin duration, i launch a classic
deadlock scenario.
Node 1 acquire local lock.
Node 1 try to acquire cluster wide lock.
Node 2 acquire local lock.
Node 2 try to acquire cluster wide lock.
Then deadlock is not detected and the TimeoutException occurs:
org.infinispan.util.concurrent.TimeoutException: Unable to acquire lock after [20
seconds] on key [counter_key] for requestor [DldGlobalTrans
action{coinToss=3136454835479173852, isMarkedForRollback=false,
lockIntention=counter_key, affectedKeys=[counter_key], locksAtOrigin=[]} Glob
alTransaction:<nodeDeadLock-11465>:5:local]! Lock held by
[DldGlobalTransaction{coinToss=3164596253027655679, isMarkedForRollback=false, lock
Intention=null, affectedKeys=[], locksAtOrigin=[]}
GlobalTransaction:<nodeDeadLock-16603>:1:remote].
I give my configuration file for ISPN and a Class TwoNodesDeadlocks.java to reproduce
the problem.
--
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
5:46 p.m.
New subject: [JBoss JIRA] Assigned: (ISPN-819) Deadlock detection seem to not detect deadlock case between two nodes on the same cache.
[
https://jira.jboss.org/browse/ISPN-819?page=com.atlassian.jira.plugin.sys...
]
Mircea Markus reassigned ISPN-819:
----------------------------------
Assignee: Mircea Markus (was: Manik Surtani)
Deadlock detection seem to not detect deadlock case between two nodes
on the same cache.
----------------------------------------------------------------------------------------
Key: ISPN-819
URL: https://jira.jboss.org/browse/ISPN-819
Project: Infinispan
Issue Type: Bug
Components: Locking and Concurrency
Affects Versions: 4.2.0.CR3
Environment: Windows XP, 4.2.0-CR3
Reporter: Thomas Jean
Assignee: Mircea Markus
Priority: Critical
Fix For: 4.2.0.CR4, 4.2.0.Final
Attachments: AgentManager_InfiniSpan_Config_Generic.xml,
log4j_Generic.properties, nodeDeadLock-43429.log, nodeDeadLock-45547.log,
TwoNodesDeadlocks.java
Hello,
after activating deadlock detection with 500 ms of spin duration, i launch a classic
deadlock scenario.
Node 1 acquire local lock.
Node 1 try to acquire cluster wide lock.
Node 2 acquire local lock.
Node 2 try to acquire cluster wide lock.
Then deadlock is not detected and the TimeoutException occurs:
org.infinispan.util.concurrent.TimeoutException: Unable to acquire lock after [20
seconds] on key [counter_key] for requestor [DldGlobalTrans
action{coinToss=3136454835479173852, isMarkedForRollback=false,
lockIntention=counter_key, affectedKeys=[counter_key], locksAtOrigin=[]} Glob
alTransaction:<nodeDeadLock-11465>:5:local]! Lock held by
[DldGlobalTransaction{coinToss=3164596253027655679, isMarkedForRollback=false, lock
Intention=null, affectedKeys=[], locksAtOrigin=[]}
GlobalTransaction:<nodeDeadLock-16603>:1:remote].
I give my configuration file for ISPN and a Class TwoNodesDeadlocks.java to reproduce
the problem.
--
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
11:15 a.m.
New subject: [JBoss JIRA] Commented: (ISPN-819) Deadlock detection seem to not detect deadlock case between two nodes on the same cache.
[
https://jira.jboss.org/browse/ISPN-819?page=com.atlassian.jira.plugin.sys...
]
Mircea Markus commented on ISPN-819:
------------------------------------
This is now fixed. UT was added for both replication and distribution. The attached
TwoNodesDeadlocks now fast fails with an DeadlockDetectedException.
Deadlock detection seem to not detect deadlock case between two nodes
on the same cache.
----------------------------------------------------------------------------------------
Key: ISPN-819
URL: https://jira.jboss.org/browse/ISPN-819
Project: Infinispan
Issue Type: Bug
Components: Locking and Concurrency
Affects Versions: 4.2.0.CR3
Environment: Windows XP, 4.2.0-CR3
Reporter: Thomas Jean
Assignee: Mircea Markus
Priority: Critical
Fix For: 4.2.0.CR4, 4.2.0.Final
Attachments: AgentManager_InfiniSpan_Config_Generic.xml,
log4j_Generic.properties, nodeDeadLock-43429.log, nodeDeadLock-45547.log,
TwoNodesDeadlocks.java
Hello,
after activating deadlock detection with 500 ms of spin duration, i launch a classic
deadlock scenario.
Node 1 acquire local lock.
Node 1 try to acquire cluster wide lock.
Node 2 acquire local lock.
Node 2 try to acquire cluster wide lock.
Then deadlock is not detected and the TimeoutException occurs:
org.infinispan.util.concurrent.TimeoutException: Unable to acquire lock after [20
seconds] on key [counter_key] for requestor [DldGlobalTrans
action{coinToss=3136454835479173852, isMarkedForRollback=false,
lockIntention=counter_key, affectedKeys=[counter_key], locksAtOrigin=[]} Glob
alTransaction:<nodeDeadLock-11465>:5:local]! Lock held by
[DldGlobalTransaction{coinToss=3164596253027655679, isMarkedForRollback=false, lock
Intention=null, affectedKeys=[], locksAtOrigin=[]}
GlobalTransaction:<nodeDeadLock-16603>:1:remote].
I give my configuration file for ISPN and a Class TwoNodesDeadlocks.java to reproduce
the problem.
--
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
3:29 p.m.
New subject: [JBoss JIRA] Resolved: (ISPN-819) Deadlock detection seem to not detect deadlock case between two nodes on the same cache.
[
https://jira.jboss.org/browse/ISPN-819?page=com.atlassian.jira.plugin.sys...
]
Mircea Markus resolved ISPN-819.
--------------------------------
Resolution: Done
fixed and integrated
Deadlock detection seem to not detect deadlock case between two nodes
on the same cache.
----------------------------------------------------------------------------------------
Key: ISPN-819
URL: https://jira.jboss.org/browse/ISPN-819
Project: Infinispan
Issue Type: Bug
Components: Locking and Concurrency
Affects Versions: 4.2.0.CR3
Environment: Windows XP, 4.2.0-CR3
Reporter: Thomas Jean
Assignee: Mircea Markus
Priority: Critical
Fix For: 4.2.0.CR4, 4.2.0.Final
Attachments: AgentManager_InfiniSpan_Config_Generic.xml,
log4j_Generic.properties, nodeDeadLock-43429.log, nodeDeadLock-45547.log,
TwoNodesDeadlocks.java
Hello,
after activating deadlock detection with 500 ms of spin duration, i launch a classic
deadlock scenario.
Node 1 acquire local lock.
Node 1 try to acquire cluster wide lock.
Node 2 acquire local lock.
Node 2 try to acquire cluster wide lock.
Then deadlock is not detected and the TimeoutException occurs:
org.infinispan.util.concurrent.TimeoutException: Unable to acquire lock after [20
seconds] on key [counter_key] for requestor [DldGlobalTrans
action{coinToss=3136454835479173852, isMarkedForRollback=false,
lockIntention=counter_key, affectedKeys=[counter_key], locksAtOrigin=[]} Glob
alTransaction:<nodeDeadLock-11465>:5:local]! Lock held by
[DldGlobalTransaction{coinToss=3164596253027655679, isMarkedForRollback=false, lock
Intention=null, affectedKeys=[], locksAtOrigin=[]}
GlobalTransaction:<nodeDeadLock-16603>:1:remote].
I give my configuration file for ISPN and a Class TwoNodesDeadlocks.java to reproduce
the problem.
--
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
10:53 a.m.
New subject: [JBoss JIRA] Updated: (ISPN-819) Deadlock detection seem to not detect deadlock case between two nodes on the same cache.
[
https://issues.jboss.org/browse/ISPN-819?page=com.atlassian.jira.plugin.s...
]
Thomas Jean updated ISPN-819:
-----------------------------
Attachment: ThreeNodesDeadlocks.java
This the case to reproduce the not detected three nodes deadlock for the same cache.
Deadlock detection seem to not detect deadlock case between two nodes
on the same cache.
----------------------------------------------------------------------------------------
Key: ISPN-819
URL: https://issues.jboss.org/browse/ISPN-819
Project: Infinispan
Issue Type: Bug
Components: Locking and Concurrency
Affects Versions: 4.2.0.CR3
Environment: Windows XP, 4.2.0-CR3
Reporter: Thomas Jean
Assignee: Mircea Markus
Priority: Critical
Fix For: 4.2.0.CR4, 4.2.0.Final
Attachments: AgentManager_InfiniSpan_Config_Generic.xml,
log4j_Generic.properties, nodeDeadLock-43429.log, nodeDeadLock-45547.log,
ThreeNodesDeadlocks.java, TwoNodesDeadlocks.java
Hello,
after activating deadlock detection with 500 ms of spin duration, i launch a classic
deadlock scenario.
Node 1 acquire local lock.
Node 1 try to acquire cluster wide lock.
Node 2 acquire local lock.
Node 2 try to acquire cluster wide lock.
Then deadlock is not detected and the TimeoutException occurs:
org.infinispan.util.concurrent.TimeoutException: Unable to acquire lock after [20
seconds] on key [counter_key] for requestor [DldGlobalTrans
action{coinToss=3136454835479173852, isMarkedForRollback=false,
lockIntention=counter_key, affectedKeys=[counter_key], locksAtOrigin=[]} Glob
alTransaction:<nodeDeadLock-11465>:5:local]! Lock held by
[DldGlobalTransaction{coinToss=3164596253027655679, isMarkedForRollback=false, lock
Intention=null, affectedKeys=[], locksAtOrigin=[]}
GlobalTransaction:<nodeDeadLock-16603>:1:remote].
I give my configuration file for ISPN and a Class TwoNodesDeadlocks.java to reproduce
the problem.
--
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
10:55 a.m.
New subject: [JBoss JIRA] Commented: (ISPN-819) Deadlock detection seem to not detect deadlock case between two nodes on the same cache.
[
https://issues.jboss.org/browse/ISPN-819?page=com.atlassian.jira.plugin.s...
]
Thomas Jean commented on ISPN-819:
----------------------------------
Hello,
I try the same test with three nodes and deadlock are not detected.
To test it use the same steps to reproduce with launching three nodes and attached file
ThreeNodesDeadlocks.java .
Deadlock detection seem to not detect deadlock case between two nodes
on the same cache.
----------------------------------------------------------------------------------------
Key: ISPN-819
URL: https://issues.jboss.org/browse/ISPN-819
Project: Infinispan
Issue Type: Bug
Components: Locking and Concurrency
Affects Versions: 4.2.0.CR3
Environment: Windows XP, 4.2.0-CR3
Reporter: Thomas Jean
Assignee: Mircea Markus
Priority: Critical
Fix For: 4.2.0.CR4, 4.2.0.Final
Attachments: AgentManager_InfiniSpan_Config_Generic.xml,
log4j_Generic.properties, nodeDeadLock-43429.log, nodeDeadLock-45547.log,
ThreeNodesDeadlocks.java, TwoNodesDeadlocks.java
Hello,
after activating deadlock detection with 500 ms of spin duration, i launch a classic
deadlock scenario.
Node 1 acquire local lock.
Node 1 try to acquire cluster wide lock.
Node 2 acquire local lock.
Node 2 try to acquire cluster wide lock.
Then deadlock is not detected and the TimeoutException occurs:
org.infinispan.util.concurrent.TimeoutException: Unable to acquire lock after [20
seconds] on key [counter_key] for requestor [DldGlobalTrans
action{coinToss=3136454835479173852, isMarkedForRollback=false,
lockIntention=counter_key, affectedKeys=[counter_key], locksAtOrigin=[]} Glob
alTransaction:<nodeDeadLock-11465>:5:local]! Lock held by
[DldGlobalTransaction{coinToss=3164596253027655679, isMarkedForRollback=false, lock
Intention=null, affectedKeys=[], locksAtOrigin=[]}
GlobalTransaction:<nodeDeadLock-16603>:1:remote].
I give my configuration file for ISPN and a Class TwoNodesDeadlocks.java to reproduce
the problem.
--
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
12:42 p.m.
New subject: [JBoss JIRA] Commented: (ISPN-819) Deadlock detection seem to not detect deadlock case between two nodes on the same cache.
[
https://issues.jboss.org/browse/ISPN-819?page=com.atlassian.jira.plugin.s...
]
Mircea Markus commented on ISPN-819:
------------------------------------
I've just tried to run the same. The test stops indeed, but received exception is:
2010-12-15 11:36:36,644 INFO [ThreeNodesDeadlocks] (main) Before LOCK...
2010-12-15 11:36:36,876 ERROR [InvocationContextInterceptor] (main) Execution error:
org.infinispan.util.concurrent.locks.DeadlockDetectedException: Deadlock found and we
DldGlobalTransaction{coinToss=3780946122068384327, isMarkedForRollback=false,
lockIntention=counter_key, affectedKeys=[counter_key], locksAtOrigin=[]}
GlobalTransaction:<nodeDeadLock-51901>:1:local shall not continue. Other tx is
DldGlobalTransaction{coinToss=6506972304784572625, isMarkedForRollback=false,
lockIntention=null, affectedKeys=[], locksAtOrigin=[]}
GlobalTransaction:<nodeDeadLock-40589>:1:remote
at
org.infinispan.util.concurrent.locks.DeadlockDetectingLockManager.lockAndRecord(DeadlockDetectingLockManager.java:84)
at org.infinispan.container.EntryFactoryImpl.acquireLock(EntryFactoryImpl.java:220)
at
org.infinispan.container.EntryFactoryImpl.wrapEntryForWriting(EntryFactoryImpl.java:155)
at
org.infinispan.container.EntryFactoryImpl.wrapEntryForWriting(EntryFactoryImpl.java:107)
at
org.infinispan.interceptors.LockingInterceptor.lockKeysForLockCommand(LockingInterceptor.java:215)
at
org.infinispan.interceptors.LockingInterceptor.visitLockControlCommand(LockingInterceptor.java:182)
at
org.infinispan.commands.control.LockControlCommand.acceptVisitor(LockControlCommand.java:143)
at
org.infinispan.interceptors.base.CommandInterceptor.invokeNextInterceptor(CommandInterceptor.java:118)
at
org.infinispan.interceptors.DeadlockDetectingInterceptor.handleDataCommand(DeadlockDetectingInterceptor.java:91)
at
org.infinispan.interceptors.DeadlockDetectingInterceptor.visitLockControlCommand(DeadlockDetectingInterceptor.java:73)
at
org.infinispan.commands.control.LockControlCommand.acceptVisitor(LockControlCommand.java:143)
at
org.infinispan.interceptors.base.CommandInterceptor.invokeNextInterceptor(CommandInterceptor.java:118)
at
org.infinispan.interceptors.base.CommandInterceptor.handleDefault(CommandInterceptor.java:132)
at
org.infinispan.commands.AbstractVisitor.visitLockControlCommand(AbstractVisitor.java:147)
at
org.infinispan.commands.control.LockControlCommand.acceptVisitor(LockControlCommand.java:143)
at
org.infinispan.interceptors.base.CommandInterceptor.invokeNextInterceptor(CommandInterceptor.java:118)
at
org.infinispan.interceptors.base.CommandInterceptor.handleDefault(CommandInterceptor.java:132)
at
org.infinispan.commands.AbstractVisitor.visitLockControlCommand(AbstractVisitor.java:147)
at
org.infinispan.commands.control.LockControlCommand.acceptVisitor(LockControlCommand.java:143)
at
org.infinispan.interceptors.base.CommandInterceptor.invokeNextInterceptor(CommandInterceptor.java:118)
at
org.infinispan.interceptors.TxInterceptor.enlistReadAndInvokeNext(TxInterceptor.java:169)
at
org.infinispan.interceptors.TxInterceptor.visitLockControlCommand(TxInterceptor.java:125)
at
org.infinispan.commands.control.LockControlCommand.acceptVisitor(LockControlCommand.java:143)
at
org.infinispan.interceptors.base.CommandInterceptor.invokeNextInterceptor(CommandInterceptor.java:118)
at
org.infinispan.interceptors.base.CommandInterceptor.handleDefault(CommandInterceptor.java:132)
at
org.infinispan.commands.AbstractVisitor.visitLockControlCommand(AbstractVisitor.java:147)
at
org.infinispan.commands.control.LockControlCommand.acceptVisitor(LockControlCommand.java:143)
at
org.infinispan.interceptors.base.CommandInterceptor.invokeNextInterceptor(CommandInterceptor.java:118)
at
org.infinispan.interceptors.InvocationContextInterceptor.handleAll(InvocationContextInterceptor.java:87)
at
org.infinispan.interceptors.InvocationContextInterceptor.visitLockControlCommand(InvocationContextInterceptor.java:63)
at
org.infinispan.commands.control.LockControlCommand.acceptVisitor(LockControlCommand.java:143)
at org.infinispan.interceptors.InterceptorChain.invoke(InterceptorChain.java:273)
at org.infinispan.CacheDelegate.lock(CacheDelegate.java:311)
at org.infinispan.CacheDelegate.lock(CacheDelegate.java:303)
at ThreeNodesDeadlocks.launchOperation(ThreeNodesDeadlocks.java:61)
at ThreeNodesDeadlocks.main(ThreeNodesDeadlocks.java:98)
This is to point out that a deadlock was found, and it's correct/expected. Do you
receive timeoutexceptions (these might indicate a problem indeed) ? Otherwise this is
expected behavior, as even with deadlock enabled one of the processes does give up.
It's only that it gives up earlier and allows the other one to continue.
Deadlock detection seem to not detect deadlock case between two nodes
on the same cache.
----------------------------------------------------------------------------------------
Key: ISPN-819
URL: https://issues.jboss.org/browse/ISPN-819
Project: Infinispan
Issue Type: Bug
Components: Locking and Concurrency
Affects Versions: 4.2.0.CR3
Environment: Windows XP, 4.2.0-CR3
Reporter: Thomas Jean
Assignee: Mircea Markus
Priority: Critical
Fix For: 4.2.0.CR4, 4.2.0.Final
Attachments: AgentManager_InfiniSpan_Config_Generic.xml,
log4j_Generic.properties, nodeDeadLock-43429.log, nodeDeadLock-45547.log,
ThreeNodesDeadlocks.java, TwoNodesDeadlocks.java
Hello,
after activating deadlock detection with 500 ms of spin duration, i launch a classic
deadlock scenario.
Node 1 acquire local lock.
Node 1 try to acquire cluster wide lock.
Node 2 acquire local lock.
Node 2 try to acquire cluster wide lock.
Then deadlock is not detected and the TimeoutException occurs:
org.infinispan.util.concurrent.TimeoutException: Unable to acquire lock after [20
seconds] on key [counter_key] for requestor [DldGlobalTrans
action{coinToss=3136454835479173852, isMarkedForRollback=false,
lockIntention=counter_key, affectedKeys=[counter_key], locksAtOrigin=[]} Glob
alTransaction:<nodeDeadLock-11465>:5:local]! Lock held by
[DldGlobalTransaction{coinToss=3164596253027655679, isMarkedForRollback=false, lock
Intention=null, affectedKeys=[], locksAtOrigin=[]}
GlobalTransaction:<nodeDeadLock-16603>:1:remote].
I give my configuration file for ISPN and a Class TwoNodesDeadlocks.java to reproduce
the problem.
--
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
1:06 p.m.
New subject: [JBoss JIRA] Updated: (ISPN-819) Deadlock detection seem to not detect deadlock case between two nodes on the same cache.
[
https://issues.jboss.org/browse/ISPN-819?page=com.atlassian.jira.plugin.s...
]
Mircea Markus updated ISPN-819:
-------------------------------
Attachment: testSetup.zip
for easy setup of the tests:
- mvn clean install -Dmaven.test.skip.exec=true -Pdistribution
- cd target/distribution
- unzip infinispan-5.0.0-SNAPSHOT-bin.zip
- infinispan-5.0.0-SNAPSHOT
- unzip testStup.zip
Deadlock detection seem to not detect deadlock case between two nodes
on the same cache.
----------------------------------------------------------------------------------------
Key: ISPN-819
URL: https://issues.jboss.org/browse/ISPN-819
Project: Infinispan
Issue Type: Bug
Components: Locking and Concurrency
Affects Versions: 4.2.0.CR3
Environment: Windows XP, 4.2.0-CR3
Reporter: Thomas Jean
Assignee: Mircea Markus
Priority: Critical
Fix For: 4.2.0.CR4, 4.2.0.Final
Attachments: AgentManager_InfiniSpan_Config_Generic.xml,
log4j_Generic.properties, nodeDeadLock-43429.log, nodeDeadLock-45547.log, testSetup.zip,
ThreeNodesDeadlocks.java, TwoNodesDeadlocks.java
Hello,
after activating deadlock detection with 500 ms of spin duration, i launch a classic
deadlock scenario.
Node 1 acquire local lock.
Node 1 try to acquire cluster wide lock.
Node 2 acquire local lock.
Node 2 try to acquire cluster wide lock.
Then deadlock is not detected and the TimeoutException occurs:
org.infinispan.util.concurrent.TimeoutException: Unable to acquire lock after [20
seconds] on key [counter_key] for requestor [DldGlobalTrans
action{coinToss=3136454835479173852, isMarkedForRollback=false,
lockIntention=counter_key, affectedKeys=[counter_key], locksAtOrigin=[]} Glob
alTransaction:<nodeDeadLock-11465>:5:local]! Lock held by
[DldGlobalTransaction{coinToss=3164596253027655679, isMarkedForRollback=false, lock
Intention=null, affectedKeys=[], locksAtOrigin=[]}
GlobalTransaction:<nodeDeadLock-16603>:1:remote].
I give my configuration file for ISPN and a Class TwoNodesDeadlocks.java to reproduce
the problem.
--
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
3:33 p.m.
New subject: [JBoss JIRA] Commented: (ISPN-819) Deadlock detection seem to not detect deadlock case between two nodes on the same cache.
[
https://issues.jboss.org/browse/ISPN-819?page=com.atlassian.jira.plugin.s...
]
Thomas Jean commented on ISPN-819:
----------------------------------
ok, i reproduce the problem but this is a new situation.
Then I will make a new issue and link with this one.
Deadlock detection seem to not detect deadlock case between two nodes
on the same cache.
----------------------------------------------------------------------------------------
Key: ISPN-819
URL: https://issues.jboss.org/browse/ISPN-819
Project: Infinispan
Issue Type: Bug
Components: Locking and Concurrency
Affects Versions: 4.2.0.CR3
Environment: Windows XP, 4.2.0-CR3
Reporter: Thomas Jean
Assignee: Mircea Markus
Priority: Critical
Fix For: 4.2.0.CR4, 4.2.0.Final
Attachments: AgentManager_InfiniSpan_Config_Generic.xml,
log4j_Generic.properties, nodeDeadLock-43429.log, nodeDeadLock-45547.log, testSetup.zip,
ThreeNodesDeadlocks.java, TwoNodesDeadlocks.java
Hello,
after activating deadlock detection with 500 ms of spin duration, i launch a classic
deadlock scenario.
Node 1 acquire local lock.
Node 1 try to acquire cluster wide lock.
Node 2 acquire local lock.
Node 2 try to acquire cluster wide lock.
Then deadlock is not detected and the TimeoutException occurs:
org.infinispan.util.concurrent.TimeoutException: Unable to acquire lock after [20
seconds] on key [counter_key] for requestor [DldGlobalTrans
action{coinToss=3136454835479173852, isMarkedForRollback=false,
lockIntention=counter_key, affectedKeys=[counter_key], locksAtOrigin=[]} Glob
alTransaction:<nodeDeadLock-11465>:5:local]! Lock held by
[DldGlobalTransaction{coinToss=3164596253027655679, isMarkedForRollback=false, lock
Intention=null, affectedKeys=[], locksAtOrigin=[]}
GlobalTransaction:<nodeDeadLock-16603>:1:remote].
I give my configuration file for ISPN and a Class TwoNodesDeadlocks.java to reproduce
the problem.
--
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
4:09 p.m.
New subject: [JBoss JIRA] Commented: (ISPN-819) Deadlock detection seem to not detect deadlock case between two nodes on the same cache.
[
https://issues.jboss.org/browse/ISPN-819?page=com.atlassian.jira.plugin.s...
]
Mircea Markus commented on ISPN-819:
------------------------------------
@Thomas are you still getting TimeoutExceptions?
Deadlock detection seem to not detect deadlock case between two nodes
on the same cache.
----------------------------------------------------------------------------------------
Key: ISPN-819
URL: https://issues.jboss.org/browse/ISPN-819
Project: Infinispan
Issue Type: Bug
Components: Locking and Concurrency
Affects Versions: 4.2.0.CR3
Environment: Windows XP, 4.2.0-CR3
Reporter: Thomas Jean
Assignee: Mircea Markus
Priority: Critical
Fix For: 4.2.0.CR4, 4.2.0.Final
Attachments: AgentManager_InfiniSpan_Config_Generic.xml,
log4j_Generic.properties, nodeDeadLock-43429.log, nodeDeadLock-45547.log, testSetup.zip,
ThreeNodesDeadlocks.java, TwoNodesDeadlocks.java
Hello,
after activating deadlock detection with 500 ms of spin duration, i launch a classic
deadlock scenario.
Node 1 acquire local lock.
Node 1 try to acquire cluster wide lock.
Node 2 acquire local lock.
Node 2 try to acquire cluster wide lock.
Then deadlock is not detected and the TimeoutException occurs:
org.infinispan.util.concurrent.TimeoutException: Unable to acquire lock after [20
seconds] on key [counter_key] for requestor [DldGlobalTrans
action{coinToss=3136454835479173852, isMarkedForRollback=false,
lockIntention=counter_key, affectedKeys=[counter_key], locksAtOrigin=[]} Glob
alTransaction:<nodeDeadLock-11465>:5:local]! Lock held by
[DldGlobalTransaction{coinToss=3164596253027655679, isMarkedForRollback=false, lock
Intention=null, affectedKeys=[], locksAtOrigin=[]}
GlobalTransaction:<nodeDeadLock-16603>:1:remote].
I give my configuration file for ISPN and a Class TwoNodesDeadlocks.java to reproduce
the problem.
--
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
4:21 p.m.
New subject: [JBoss JIRA] Commented: (ISPN-819) Deadlock detection seem to not detect deadlock case between two nodes on the same cache.
[
https://issues.jboss.org/browse/ISPN-819?page=com.atlassian.jira.plugin.s...
]
Thomas Jean commented on ISPN-819:
----------------------------------
When launching with two nodes, all works good: no timeout exception.
Deadlock detection seem to not detect deadlock case between two nodes
on the same cache.
----------------------------------------------------------------------------------------
Key: ISPN-819
URL: https://issues.jboss.org/browse/ISPN-819
Project: Infinispan
Issue Type: Bug
Components: Locking and Concurrency
Affects Versions: 4.2.0.CR3
Environment: Windows XP, 4.2.0-CR3
Reporter: Thomas Jean
Assignee: Mircea Markus
Priority: Critical
Fix For: 4.2.0.CR4, 4.2.0.Final
Attachments: AgentManager_InfiniSpan_Config_Generic.xml,
log4j_Generic.properties, nodeDeadLock-43429.log, nodeDeadLock-45547.log, testSetup.zip,
ThreeNodesDeadlocks.java, TwoNodesDeadlocks.java
Hello,
after activating deadlock detection with 500 ms of spin duration, i launch a classic
deadlock scenario.
Node 1 acquire local lock.
Node 1 try to acquire cluster wide lock.
Node 2 acquire local lock.
Node 2 try to acquire cluster wide lock.
Then deadlock is not detected and the TimeoutException occurs:
org.infinispan.util.concurrent.TimeoutException: Unable to acquire lock after [20
seconds] on key [counter_key] for requestor [DldGlobalTrans
action{coinToss=3136454835479173852, isMarkedForRollback=false,
lockIntention=counter_key, affectedKeys=[counter_key], locksAtOrigin=[]} Glob
alTransaction:<nodeDeadLock-11465>:5:local]! Lock held by
[DldGlobalTransaction{coinToss=3164596253027655679, isMarkedForRollback=false, lock
Intention=null, affectedKeys=[], locksAtOrigin=[]}
GlobalTransaction:<nodeDeadLock-16603>:1:remote].
I give my configuration file for ISPN and a Class TwoNodesDeadlocks.java to reproduce
the problem.
--
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
4:25 p.m.
New subject: [JBoss JIRA] Commented: (ISPN-819) Deadlock detection seem to not detect deadlock case between two nodes on the same cache.
[
https://issues.jboss.org/browse/ISPN-819?page=com.atlassian.jira.plugin.s...
]
Mircea Markus commented on ISPN-819:
------------------------------------
Right, it's not clear to me what the problem is this time. If you don't get
TimeoutException (but DeadlockDetectedException) then it's not a problem - it's
expected behavior. Disregarding the number of nodes.
Deadlock detection seem to not detect deadlock case between two nodes
on the same cache.
----------------------------------------------------------------------------------------
Key: ISPN-819
URL: https://issues.jboss.org/browse/ISPN-819
Project: Infinispan
Issue Type: Bug
Components: Locking and Concurrency
Affects Versions: 4.2.0.CR3
Environment: Windows XP, 4.2.0-CR3
Reporter: Thomas Jean
Assignee: Mircea Markus
Priority: Critical
Fix For: 4.2.0.CR4, 4.2.0.Final
Attachments: AgentManager_InfiniSpan_Config_Generic.xml,
log4j_Generic.properties, nodeDeadLock-43429.log, nodeDeadLock-45547.log, testSetup.zip,
ThreeNodesDeadlocks.java, TwoNodesDeadlocks.java
Hello,
after activating deadlock detection with 500 ms of spin duration, i launch a classic
deadlock scenario.
Node 1 acquire local lock.
Node 1 try to acquire cluster wide lock.
Node 2 acquire local lock.
Node 2 try to acquire cluster wide lock.
Then deadlock is not detected and the TimeoutException occurs:
org.infinispan.util.concurrent.TimeoutException: Unable to acquire lock after [20
seconds] on key [counter_key] for requestor [DldGlobalTrans
action{coinToss=3136454835479173852, isMarkedForRollback=false,
lockIntention=counter_key, affectedKeys=[counter_key], locksAtOrigin=[]} Glob
alTransaction:<nodeDeadLock-11465>:5:local]! Lock held by
[DldGlobalTransaction{coinToss=3164596253027655679, isMarkedForRollback=false, lock
Intention=null, affectedKeys=[], locksAtOrigin=[]}
GlobalTransaction:<nodeDeadLock-16603>:1:remote].
I give my configuration file for ISPN and a Class TwoNodesDeadlocks.java to reproduce
the problem.
--
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
4:32 p.m.
New subject: [JBoss JIRA] Commented: (ISPN-819) Deadlock detection seem to not detect deadlock case between two nodes on the same cache.
[
https://issues.jboss.org/browse/ISPN-819?page=com.atlassian.jira.plugin.s...
]
Thomas Jean commented on ISPN-819:
----------------------------------
Ok, you are right.
The DeadlockDetectionException is correctly thrown like expected, independently of the
number of nodes.
(In the last comment,i just made a confusion with ISPN-836 which is another case).
Deadlock detection seem to not detect deadlock case between two nodes
on the same cache.
----------------------------------------------------------------------------------------
Key: ISPN-819
URL: https://issues.jboss.org/browse/ISPN-819
Project: Infinispan
Issue Type: Bug
Components: Locking and Concurrency
Affects Versions: 4.2.0.CR3
Environment: Windows XP, 4.2.0-CR3
Reporter: Thomas Jean
Assignee: Mircea Markus
Priority: Critical
Fix For: 4.2.0.CR4, 4.2.0.Final
Attachments: AgentManager_InfiniSpan_Config_Generic.xml,
log4j_Generic.properties, nodeDeadLock-43429.log, nodeDeadLock-45547.log, testSetup.zip,
ThreeNodesDeadlocks.java, TwoNodesDeadlocks.java
Hello,
after activating deadlock detection with 500 ms of spin duration, i launch a classic
deadlock scenario.
Node 1 acquire local lock.
Node 1 try to acquire cluster wide lock.
Node 2 acquire local lock.
Node 2 try to acquire cluster wide lock.
Then deadlock is not detected and the TimeoutException occurs:
org.infinispan.util.concurrent.TimeoutException: Unable to acquire lock after [20
seconds] on key [counter_key] for requestor [DldGlobalTrans
action{coinToss=3136454835479173852, isMarkedForRollback=false,
lockIntention=counter_key, affectedKeys=[counter_key], locksAtOrigin=[]} Glob
alTransaction:<nodeDeadLock-11465>:5:local]! Lock held by
[DldGlobalTransaction{coinToss=3164596253027655679, isMarkedForRollback=false, lock
Intention=null, affectedKeys=[], locksAtOrigin=[]}
GlobalTransaction:<nodeDeadLock-16603>:1:remote].
I give my configuration file for ISPN and a Class TwoNodesDeadlocks.java to reproduce
the problem.
--
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
4:41 p.m.
New subject: [JBoss JIRA] Commented: (ISPN-819) Deadlock detection seem to not detect deadlock case between two nodes on the same cache.
[
https://issues.jboss.org/browse/ISPN-819?page=com.atlassian.jira.plugin.s...
]
Mircea Markus commented on ISPN-819:
------------------------------------
Thanks for the clarification.
Deadlock detection seem to not detect deadlock case between two nodes
on the same cache.
----------------------------------------------------------------------------------------
Key: ISPN-819
URL: https://issues.jboss.org/browse/ISPN-819
Project: Infinispan
Issue Type: Bug
Components: Locking and Concurrency
Affects Versions: 4.2.0.CR3
Environment: Windows XP, 4.2.0-CR3
Reporter: Thomas Jean
Assignee: Mircea Markus
Priority: Critical
Fix For: 4.2.0.CR4, 4.2.0.Final
Attachments: AgentManager_InfiniSpan_Config_Generic.xml,
log4j_Generic.properties, nodeDeadLock-43429.log, nodeDeadLock-45547.log, testSetup.zip,
ThreeNodesDeadlocks.java, TwoNodesDeadlocks.java
Hello,
after activating deadlock detection with 500 ms of spin duration, i launch a classic
deadlock scenario.
Node 1 acquire local lock.
Node 1 try to acquire cluster wide lock.
Node 2 acquire local lock.
Node 2 try to acquire cluster wide lock.
Then deadlock is not detected and the TimeoutException occurs:
org.infinispan.util.concurrent.TimeoutException: Unable to acquire lock after [20
seconds] on key [counter_key] for requestor [DldGlobalTrans
action{coinToss=3136454835479173852, isMarkedForRollback=false,
lockIntention=counter_key, affectedKeys=[counter_key], locksAtOrigin=[]} Glob
alTransaction:<nodeDeadLock-11465>:5:local]! Lock held by
[DldGlobalTransaction{coinToss=3164596253027655679, isMarkedForRollback=false, lock
Intention=null, affectedKeys=[], locksAtOrigin=[]}
GlobalTransaction:<nodeDeadLock-16603>:1:remote].
I give my configuration file for ISPN and a Class TwoNodesDeadlocks.java to reproduce
the problem.
--
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
4880
days inactive
4889
days old
infinispan-issues@lists.jboss.org
15 comments
3 participants
participants (3)
-
Manik Surtani (JIRA) -
Mircea Markus (JIRA)
-
Thomas Jean (JIRA)