[JBoss JIRA] Created: (JBAS-5095) Race condition between connection.close() and transaction.rollback() causes connection leak

[JBoss JIRA] Created:...

[JBoss JIRA] Created: (JGRP-817)...

Adrian Brock (JIRA)

Thursday, 20 December 2007 Thu, 20 Dec '07

11:01 a.m.

Race condition between connection.close() and transaction.rollback() causes connection leak ------------------------------------------------------------------------------------------- Key: JBAS-5095 URL: http://jira.jboss.com/jira/browse/JBAS-5095 Project: JBoss Application Server Issue Type: Bug Security Level: Public (Everyone can see) Components: JCA service Affects Versions: JBossAS-4.2.2.GA Reporter: Adrian Brock Assigned To: Adrian Brock Fix For: JBossAS-5.0.0.CR1, JBossAS-4.2.3.GA There is a race condition between the connection.close() and an asynchronous transaction completion (e.g. transaction timeout) that can lead to a leaked connection permit with eventual pool exhaustion when using track-connection-by-tx (i.e. all local resources). The problem is caused in the TxConnectionManager where the unregister assoication and the afterCompletion are not synchronized with each other. Both can see the connection as ready to return to the pool (one removes the handle, the other the synchronization at about the same time so they both check the same state) and both do it. The InternalManagedConnectionPool does not check for duplicate returns and uses an ArrayList to store connections (for the MRU behaviour). So the connection ends up in the pool twice This eventually causes problems like the leaked permit. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://jira.jboss.com/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

Show replies by date

Adrian Brock (JIRA)

Thursday, 20 December Thu, 20 Dec

11:03 a.m.

New subject: [JBoss JIRA] Commented: (JBAS-5095) Race condition between connection.close() and transaction.rollback() causes connection leak

[ http://jira.jboss.com/jira/browse/JBAS-5095?page=comments#action_12393156 ] Adrian Brock commented on JBAS-5095: ------------------------------------ I've committed a fix in 4.2.x that works around the problem in the InternalManagedConnectionPool by avoiding the duplciate return and logging a warning. But the underlying problem (the race in TxConnectionManager) needs fixing properly.

...

Race condition between connection.close() and transaction.rollback() causes connection leak ------------------------------------------------------------------------------------------- Key: JBAS-5095 URL: http://jira.jboss.com/jira/browse/JBAS-5095 Project: JBoss Application Server Issue Type: Bug Security Level: Public(Everyone can see) Components: JCA service Affects Versions: JBossAS-4.2.2.GA Reporter: Adrian Brock Assigned To: Adrian Brock Fix For: JBossAS-5.0.0.CR1, JBossAS-4.2.3.GA There is a race condition between the connection.close() and an asynchronous transaction completion (e.g. transaction timeout) that can lead to a leaked connection permit with eventual pool exhaustion when using track-connection-by-tx (i.e. all local resources). The problem is caused in the TxConnectionManager where the unregister assoication and the afterCompletion are not synchronized with each other. Both can see the connection as ready to return to the pool (one removes the handle, the other the synchronization at about the same time so they both check the same state) and both do it. The InternalManagedConnectionPool does not check for duplicate returns and uses an ArrayList to store connections (for the MRU behaviour). So the connection ends up in the pool twice This eventually causes problems like the leaked permit.

-- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://jira.jboss.com/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

Dimitris Andreadis (JIRA)

Wednesday, 6 February Wed, 6 Feb

8:19 a.m.

New subject: [JBoss JIRA] Updated: (JBAS-5095) Race condition between connection.close() and transaction.rollback() causes connection leak

...

Race condition between connection.close() and transaction.rollback() causes connection leak ------------------------------------------------------------------------------------------- Key: JBAS-5095 URL: http://jira.jboss.com/jira/browse/JBAS-5095 Project: JBoss Application Server Issue Type: Bug Security Level: Public(Everyone can see) Components: JCA service Affects Versions: JBossAS-4.2.2.GA Reporter: Adrian Brock Assigned To: Adrian Brock Priority: Critical Fix For: JBossAS-5.0.0.CR1 There is a race condition between the connection.close() and an asynchronous transaction completion (e.g. transaction timeout) that can lead to a leaked connection permit with eventual pool exhaustion when using track-connection-by-tx (i.e. all local resources). The problem is caused in the TxConnectionManager where the unregister assoication and the afterCompletion are not synchronized with each other. Both can see the connection as ready to return to the pool (one removes the handle, the other the synchronization at about the same time so they both check the same state) and both do it. The InternalManagedConnectionPool does not check for duplicate returns and uses an ArrayList to store connections (for the MRU behaviour). So the connection ends up in the pool twice This eventually causes problems like the leaked permit.

Dimitris Andreadis (JIRA)

Friday, 13 June Fri, 13 Jun

2:34 a.m.

New subject: [JBoss JIRA] Updated: (JBAS-5095) Race condition between connection.close() and transaction.rollback() causes connection leak

[ http://jira.jboss.com/jira/browse/JBAS-5095?page=all ] Dimitris Andreadis updated JBAS-5095: ------------------------------------- Fix Version/s: JBossAS-5.0.0.CR2 JBossAS-4.2.3.GA (was: JBossAS-5.0.0.CR1) Adding 4.2.3 fix version and rescheduling for AS5 CR2. Or the issue could be resolved since the workaround is there and a new one created for the final/proper fix.

...

Race condition between connection.close() and transaction.rollback() causes connection leak ------------------------------------------------------------------------------------------- Key: JBAS-5095 URL: http://jira.jboss.com/jira/browse/JBAS-5095 Project: JBoss Application Server Issue Type: Bug Security Level: Public(Everyone can see) Components: JCA service Affects Versions: JBossAS-4.2.2.GA Reporter: Adrian Brock Assigned To: Adrian Brock Priority: Critical Fix For: JBossAS-5.0.0.CR2, JBossAS-4.2.3.GA There is a race condition between the connection.close() and an asynchronous transaction completion (e.g. transaction timeout) that can lead to a leaked connection permit with eventual pool exhaustion when using track-connection-by-tx (i.e. all local resources). The problem is caused in the TxConnectionManager where the unregister assoication and the afterCompletion are not synchronized with each other. Both can see the connection as ready to return to the pool (one removes the handle, the other the synchronization at about the same time so they both check the same state) and both do it. The InternalManagedConnectionPool does not check for duplicate returns and uses an ArrayList to store connections (for the MRU behaviour). So the connection ends up in the pool twice This eventually causes problems like the leaked permit.

Dimitris Andreadis (JIRA)

Monday, 14 July Mon, 14 Jul

10:48 a.m.

New subject: [JBoss JIRA] Commented: (JBAS-5095) Race condition between connection.close() and transaction.rollback() causes connection leak

[ http://jira.jboss.com/jira/browse/JBAS-5095?page=comments#action_12421222 ] Dimitris Andreadis commented on JBAS-5095: ------------------------------------------ Adrian, can we close this for 4.2.3.GA and create a linked task for fixing it properly in a future version? I want to release 4.2.3.GA in a couple of days.

...

Race condition between connection.close() and transaction.rollback() causes connection leak ------------------------------------------------------------------------------------------- Key: JBAS-5095 URL: http://jira.jboss.com/jira/browse/JBAS-5095 Project: JBoss Application Server Issue Type: Bug Security Level: Public(Everyone can see) Components: JCA service Affects Versions: JBossAS-4.2.2.GA Reporter: Adrian Brock Assigned To: Adrian Brock Priority: Critical Fix For: JBossAS-5.0.0.CR2, JBossAS-4.2.3.GA There is a race condition between the connection.close() and an asynchronous transaction completion (e.g. transaction timeout) that can lead to a leaked connection permit with eventual pool exhaustion when using track-connection-by-tx (i.e. all local resources). The problem is caused in the TxConnectionManager where the unregister assoication and the afterCompletion are not synchronized with each other. Both can see the connection as ready to return to the pool (one removes the handle, the other the synchronization at about the same time so they both check the same state) and both do it. The InternalManagedConnectionPool does not check for duplicate returns and uses an ArrayList to store connections (for the MRU behaviour). So the connection ends up in the pool twice This eventually causes problems like the leaked permit.

Dimitris Andreadis (JIRA)

Tuesday, 15 July Tue, 15 Jul

9:39 a.m.

New subject: [JBoss JIRA] Assigned: (JBAS-5095) Race condition between connection.close() and transaction.rollback() causes connection leak

[ http://jira.jboss.com/jira/browse/JBAS-5095?page=all ] Dimitris Andreadis reassigned JBAS-5095: ---------------------------------------- Assignee: Jesper Pedersen (was: Adrian Brock) Give it to Jesper.

...

Race condition between connection.close() and transaction.rollback() causes connection leak ------------------------------------------------------------------------------------------- Key: JBAS-5095 URL: http://jira.jboss.com/jira/browse/JBAS-5095 Project: JBoss Application Server Issue Type: Bug Security Level: Public(Everyone can see) Components: JCA service Affects Versions: JBossAS-4.2.2.GA Reporter: Adrian Brock Assigned To: Jesper Pedersen Priority: Critical Fix For: JBossAS-5.0.0.CR2, JBossAS-4.2.3.GA There is a race condition between the connection.close() and an asynchronous transaction completion (e.g. transaction timeout) that can lead to a leaked connection permit with eventual pool exhaustion when using track-connection-by-tx (i.e. all local resources). The problem is caused in the TxConnectionManager where the unregister assoication and the afterCompletion are not synchronized with each other. Both can see the connection as ready to return to the pool (one removes the handle, the other the synchronization at about the same time so they both check the same state) and both do it. The InternalManagedConnectionPool does not check for duplicate returns and uses an ArrayList to store connections (for the MRU behaviour). So the connection ends up in the pool twice This eventually causes problems like the leaked permit.

Adrian Brock (JIRA)

Wednesday, 16 July Wed, 16 Jul

4:39 a.m.

New subject: [JBoss JIRA] Updated: (JBAS-5095) Race condition between connection.close() and transaction.rollback() causes connection leak

[ http://jira.jboss.com/jira/browse/JBAS-5095?page=com.atlassian.jira.plugi... ] Adrian Brock updated JBAS-5095: ------------------------------- Fix Version/s: JBossAS-4.2.4.GA (was: JBossAS-4.2.3.GA)

...

Race condition between connection.close() and transaction.rollback() causes connection leak ------------------------------------------------------------------------------------------- Key: JBAS-5095 URL: http://jira.jboss.com/jira/browse/JBAS-5095 Project: JBoss Application Server Issue Type: Bug Security Level: Public(Everyone can see) Components: JCA service Affects Versions: JBossAS-4.2.2.GA Reporter: Adrian Brock Assignee: Jesper Pedersen Priority: Critical Fix For: JBossAS-5.0.0.CR2, JBossAS-4.2.4.GA There is a race condition between the connection.close() and an asynchronous transaction completion (e.g. transaction timeout) that can lead to a leaked connection permit with eventual pool exhaustion when using track-connection-by-tx (i.e. all local resources). The problem is caused in the TxConnectionManager where the unregister assoication and the afterCompletion are not synchronized with each other. Both can see the connection as ready to return to the pool (one removes the handle, the other the synchronization at about the same time so they both check the same state) and both do it. The InternalManagedConnectionPool does not check for duplicate returns and uses an ArrayList to store connections (for the MRU behaviour). So the connection ends up in the pool twice This eventually causes problems like the leaked permit.

Adrian Brock (JIRA)

4:42 a.m.

New subject: [JBoss JIRA] Commented: (JBAS-5095) Race condition between connection.close() and transaction.rollback() causes connection leak

[ http://jira.jboss.com/jira/browse/JBAS-5095?page=com.atlassian.jira.plugi... ] Adrian Brock commented on JBAS-5095: ------------------------------------ I've moved this to 4.2.4 to fix properly. The problem is fixed in 4.2.3 but only with an ugly workaround that produces a WARNing when it occurs. The correct fix is to make the two checks number of handles=0 and connection is no longer track-by-tx atomic in the two different places where it occurs. This became an issue with JBossTM in 4.2.0+ where a connection.close() can race with a transaction.rollback() due to a transaction timeout.

...

Race condition between connection.close() and transaction.rollback() causes connection leak ------------------------------------------------------------------------------------------- Key: JBAS-5095 URL: http://jira.jboss.com/jira/browse/JBAS-5095 Project: JBoss Application Server Issue Type: Bug Security Level: Public(Everyone can see) Components: JCA service Affects Versions: JBossAS-4.2.2.GA Reporter: Adrian Brock Assignee: Jesper Pedersen Priority: Critical Fix For: JBossAS-5.0.0.CR2, JBossAS-4.2.4.GA There is a race condition between the connection.close() and an asynchronous transaction completion (e.g. transaction timeout) that can lead to a leaked connection permit with eventual pool exhaustion when using track-connection-by-tx (i.e. all local resources). The problem is caused in the TxConnectionManager where the unregister assoication and the afterCompletion are not synchronized with each other. Both can see the connection as ready to return to the pool (one removes the handle, the other the synchronization at about the same time so they both check the same state) and both do it. The InternalManagedConnectionPool does not check for duplicate returns and uses an ArrayList to store connections (for the MRU behaviour). So the connection ends up in the pool twice This eventually causes problems like the leaked permit.

Jesper Pedersen (JIRA)

Thursday, 24 July Thu, 24 Jul

6:31 a.m.

New subject: [JBoss JIRA] Updated: (JBAS-5095) Race condition between connection.close() and transaction.rollback() causes connection leak

[ https://jira.jboss.org/jira/browse/JBAS-5095?page=com.atlassian.jira.plug... ] Jesper Pedersen updated JBAS-5095: ---------------------------------- JBoss Forum Reference: http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4166381

...

Race condition between connection.close() and transaction.rollback() causes connection leak ------------------------------------------------------------------------------------------- Key: JBAS-5095 URL: https://jira.jboss.org/jira/browse/JBAS-5095 Project: JBoss Application Server Issue Type: Bug Security Level: Public(Everyone can see) Components: JCA service Affects Versions: JBossAS-4.2.2.GA Reporter: Adrian Brock Assignee: Jesper Pedersen Priority: Critical Fix For: JBossAS-5.0.0.CR2, JBossAS-4.2.4.GA There is a race condition between the connection.close() and an asynchronous transaction completion (e.g. transaction timeout) that can lead to a leaked connection permit with eventual pool exhaustion when using track-connection-by-tx (i.e. all local resources). The problem is caused in the TxConnectionManager where the unregister assoication and the afterCompletion are not synchronized with each other. Both can see the connection as ready to return to the pool (one removes the handle, the other the synchronization at about the same time so they both check the same state) and both do it. The InternalManagedConnectionPool does not check for duplicate returns and uses an ArrayList to store connections (for the MRU behaviour). So the connection ends up in the pool twice This eventually causes problems like the leaked permit.

-- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

Jesper Pedersen (JIRA)

6:33 a.m.

New subject: [JBoss JIRA] Updated: (JBAS-5095) Race condition between connection.close() and transaction.rollback() causes connection leak

[ https://jira.jboss.org/jira/browse/JBAS-5095?page=com.atlassian.jira.plug... ] Jesper Pedersen updated JBAS-5095: ---------------------------------- Attachment: jbas-5095.patch Patch against trunk

...

Race condition between connection.close() and transaction.rollback() causes connection leak ------------------------------------------------------------------------------------------- Key: JBAS-5095 URL: https://jira.jboss.org/jira/browse/JBAS-5095 Project: JBoss Application Server Issue Type: Bug Security Level: Public(Everyone can see) Components: JCA service Affects Versions: JBossAS-4.2.2.GA Reporter: Adrian Brock Assignee: Jesper Pedersen Priority: Critical Fix For: JBossAS-5.0.0.CR2, JBossAS-4.2.4.GA Attachments: jbas-5095.patch There is a race condition between the connection.close() and an asynchronous transaction completion (e.g. transaction timeout) that can lead to a leaked connection permit with eventual pool exhaustion when using track-connection-by-tx (i.e. all local resources). The problem is caused in the TxConnectionManager where the unregister assoication and the afterCompletion are not synchronized with each other. Both can see the connection as ready to return to the pool (one removes the handle, the other the synchronization at about the same time so they both check the same state) and both do it. The InternalManagedConnectionPool does not check for duplicate returns and uses an ArrayList to store connections (for the MRU behaviour). So the connection ends up in the pool twice This eventually causes problems like the leaked permit.

Jesper Pedersen (JIRA)

Tuesday, 26 August Tue, 26 Aug

7:40 a.m.

New subject: [JBoss JIRA] Updated: (JBAS-5095) Race condition between connection.close() and transaction.rollback() causes connection leak

[ https://jira.jboss.org/jira/browse/JBAS-5095?page=com.atlassian.jira.plug... ] Jesper Pedersen updated JBAS-5095: ---------------------------------- Attachment: (was: jbas-5095.patch)

...

Race condition between connection.close() and transaction.rollback() causes connection leak ------------------------------------------------------------------------------------------- Key: JBAS-5095 URL: https://jira.jboss.org/jira/browse/JBAS-5095 Project: JBoss Application Server Issue Type: Bug Security Level: Public(Everyone can see) Components: JCA service Affects Versions: JBossAS-4.2.2.GA Reporter: Adrian Brock Assignee: Jesper Pedersen Priority: Critical Fix For: JBossAS-5.0.0.CR2, JBossAS-4.2.4.GA Attachments: JBAS-5095.patch There is a race condition between the connection.close() and an asynchronous transaction completion (e.g. transaction timeout) that can lead to a leaked connection permit with eventual pool exhaustion when using track-connection-by-tx (i.e. all local resources). The problem is caused in the TxConnectionManager where the unregister assoication and the afterCompletion are not synchronized with each other. Both can see the connection as ready to return to the pool (one removes the handle, the other the synchronization at about the same time so they both check the same state) and both do it. The InternalManagedConnectionPool does not check for duplicate returns and uses an ArrayList to store connections (for the MRU behaviour). So the connection ends up in the pool twice This eventually causes problems like the leaked permit.

Jesper Pedersen (JIRA)

7:40 a.m.

New subject: [JBoss JIRA] Updated: (JBAS-5095) Race condition between connection.close() and transaction.rollback() causes connection leak

[ https://jira.jboss.org/jira/browse/JBAS-5095?page=com.atlassian.jira.plug... ] Jesper Pedersen updated JBAS-5095: ---------------------------------- Attachment: JBAS-5095.patch

...

Race condition between connection.close() and transaction.rollback() causes connection leak ------------------------------------------------------------------------------------------- Key: JBAS-5095 URL: https://jira.jboss.org/jira/browse/JBAS-5095 Project: JBoss Application Server Issue Type: Bug Security Level: Public(Everyone can see) Components: JCA service Affects Versions: JBossAS-4.2.2.GA Reporter: Adrian Brock Assignee: Jesper Pedersen Priority: Critical Fix For: JBossAS-5.0.0.CR2, JBossAS-4.2.4.GA Attachments: JBAS-5095.patch There is a race condition between the connection.close() and an asynchronous transaction completion (e.g. transaction timeout) that can lead to a leaked connection permit with eventual pool exhaustion when using track-connection-by-tx (i.e. all local resources). The problem is caused in the TxConnectionManager where the unregister assoication and the afterCompletion are not synchronized with each other. Both can see the connection as ready to return to the pool (one removes the handle, the other the synchronization at about the same time so they both check the same state) and both do it. The InternalManagedConnectionPool does not check for duplicate returns and uses an ArrayList to store connections (for the MRU behaviour). So the connection ends up in the pool twice This eventually causes problems like the leaked permit.

Jesper Pedersen (JIRA)

Wednesday, 27 August Wed, 27 Aug

3:52 a.m.

New subject: [JBoss JIRA] Updated: (JBAS-5095) Race condition between connection.close() and transaction.rollback() causes connection leak

[ https://jira.jboss.org/jira/browse/JBAS-5095?page=com.atlassian.jira.plug... ] Jesper Pedersen updated JBAS-5095: ---------------------------------- Attachment: (was: JBAS-5095.patch)

...

Race condition between connection.close() and transaction.rollback() causes connection leak ------------------------------------------------------------------------------------------- Key: JBAS-5095 URL: https://jira.jboss.org/jira/browse/JBAS-5095 Project: JBoss Application Server Issue Type: Bug Security Level: Public(Everyone can see) Components: JCA service Affects Versions: JBossAS-4.2.2.GA Reporter: Adrian Brock Assignee: Jesper Pedersen Priority: Critical Fix For: JBossAS-5.0.0.CR2, JBossAS-4.2.4.GA There is a race condition between the connection.close() and an asynchronous transaction completion (e.g. transaction timeout) that can lead to a leaked connection permit with eventual pool exhaustion when using track-connection-by-tx (i.e. all local resources). The problem is caused in the TxConnectionManager where the unregister assoication and the afterCompletion are not synchronized with each other. Both can see the connection as ready to return to the pool (one removes the handle, the other the synchronization at about the same time so they both check the same state) and both do it. The InternalManagedConnectionPool does not check for duplicate returns and uses an ArrayList to store connections (for the MRU behaviour). So the connection ends up in the pool twice This eventually causes problems like the leaked permit.

Jesper Pedersen (JIRA)

3:54 a.m.

New subject: [JBoss JIRA] Updated: (JBAS-5095) Race condition between connection.close() and transaction.rollback() causes connection leak

[ https://jira.jboss.org/jira/browse/JBAS-5095?page=com.atlassian.jira.plug... ] Jesper Pedersen updated JBAS-5095: ---------------------------------- Attachment: JBAS-5095.patch

...

Race condition between connection.close() and transaction.rollback() causes connection leak ------------------------------------------------------------------------------------------- Key: JBAS-5095 URL: https://jira.jboss.org/jira/browse/JBAS-5095 Project: JBoss Application Server Issue Type: Bug Security Level: Public(Everyone can see) Components: JCA service Affects Versions: JBossAS-4.2.2.GA Reporter: Adrian Brock Assignee: Jesper Pedersen Priority: Critical Fix For: JBossAS-5.0.0.CR2, JBossAS-4.2.4.GA Attachments: JBAS-5095.patch There is a race condition between the connection.close() and an asynchronous transaction completion (e.g. transaction timeout) that can lead to a leaked connection permit with eventual pool exhaustion when using track-connection-by-tx (i.e. all local resources). The problem is caused in the TxConnectionManager where the unregister assoication and the afterCompletion are not synchronized with each other. Both can see the connection as ready to return to the pool (one removes the handle, the other the synchronization at about the same time so they both check the same state) and both do it. The InternalManagedConnectionPool does not check for duplicate returns and uses an ArrayList to store connections (for the MRU behaviour). So the connection ends up in the pool twice This eventually causes problems like the leaked permit.

Jesper Pedersen (JIRA)

Thursday, 4 September Thu, 4 Sep

7:31 a.m.

New subject: [JBoss JIRA] Updated: (JBAS-5095) Race condition between connection.close() and transaction.rollback() causes connection leak

[ https://jira.jboss.org/jira/browse/JBAS-5095?page=com.atlassian.jira.plug... ] Jesper Pedersen updated JBAS-5095: ---------------------------------- Attachment: (was: JBAS-5095.patch)

...

Race condition between connection.close() and transaction.rollback() causes connection leak ------------------------------------------------------------------------------------------- Key: JBAS-5095 URL: https://jira.jboss.org/jira/browse/JBAS-5095 Project: JBoss Application Server Issue Type: Bug Security Level: Public(Everyone can see) Components: JCA service Affects Versions: JBossAS-4.2.2.GA Reporter: Adrian Brock Assignee: Jesper Pedersen Priority: Critical Fix For: JBossAS-5.0.0.CR2, JBossAS-4.2.4.GA There is a race condition between the connection.close() and an asynchronous transaction completion (e.g. transaction timeout) that can lead to a leaked connection permit with eventual pool exhaustion when using track-connection-by-tx (i.e. all local resources). The problem is caused in the TxConnectionManager where the unregister assoication and the afterCompletion are not synchronized with each other. Both can see the connection as ready to return to the pool (one removes the handle, the other the synchronization at about the same time so they both check the same state) and both do it. The InternalManagedConnectionPool does not check for duplicate returns and uses an ArrayList to store connections (for the MRU behaviour). So the connection ends up in the pool twice This eventually causes problems like the leaked permit.

Jesper Pedersen (JIRA)

7:31 a.m.

New subject: [JBoss JIRA] Resolved: (JBAS-5095) Race condition between connection.close() and transaction.rollback() causes connection leak

[ https://jira.jboss.org/jira/browse/JBAS-5095?page=com.atlassian.jira.plug... ] Jesper Pedersen resolved JBAS-5095. ----------------------------------- Resolution: Done

...

Race condition between connection.close() and transaction.rollback() causes connection leak ------------------------------------------------------------------------------------------- Key: JBAS-5095 URL: https://jira.jboss.org/jira/browse/JBAS-5095 Project: JBoss Application Server Issue Type: Bug Security Level: Public(Everyone can see) Components: JCA service Affects Versions: JBossAS-4.2.2.GA Reporter: Adrian Brock Assignee: Jesper Pedersen Priority: Critical Fix For: JBossAS-5.0.0.CR2, JBossAS-4.2.4.GA There is a race condition between the connection.close() and an asynchronous transaction completion (e.g. transaction timeout) that can lead to a leaked connection permit with eventual pool exhaustion when using track-connection-by-tx (i.e. all local resources). The problem is caused in the TxConnectionManager where the unregister assoication and the afterCompletion are not synchronized with each other. Both can see the connection as ready to return to the pool (one removes the handle, the other the synchronization at about the same time so they both check the same state) and both do it. The InternalManagedConnectionPool does not check for duplicate returns and uses an ArrayList to store connections (for the MRU behaviour). So the connection ends up in the pool twice This eventually causes problems like the leaked permit.

5976

days inactive

6235

days old

jboss-jira@lists.jboss.org

Manage subscription

15 comments

3 participants

tags (0)

participants (3)

Adrian Brock (JIRA)
Dimitris Andreadis (JIRA)
Jesper Pedersen (JIRA)

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

[JBoss JIRA] Created: (JBAS-5095) Race condition between connection.close() and transaction.rollback() causes connection leak