Re: Please test 0.3.x snapshots....

2012-05-24 Thread Guillaume Nodet
I actually just found a possible deadlock in the the transaction
manager, so I'll do a bit more testing with my local fix and commit it
later today.

On Tue, May 22, 2012 at 5:43 PM, Daniel Kulp dk...@apache.org wrote:

 Guillaume and I have gone through and pulled back several bug fixes for
 couple of the bundles based on the older 0.3.x releases and now have some
 new snapshots deployed.   It would be great if the various interested
 parties could take a quick look and test them out.   I'd like to start the
 release process for them possibly on Thursday.

 The svn branch for the changes is at:
 https://svn.apache.org/repos/asf/aries/branches/blueprint-0.3.2-fixes

 which is now a bit mis-named as we also have fixes for utils, jmx, and
 transaction manager in there.    Anyway, please speak up if there are
 additional issues or problems.

 --
 Daniel Kulp
 dk...@apache.org - http://dankulp.com/blog
 Talend Community Coder - http://coders.talend.com




-- 

Guillaume Nodet

Blog: http://gnodet.blogspot.com/

FuseSource, Integration everywhere
http://fusesource.com


[jira] [Resolved] (ARIES-856) Possible deadlock when starting the transaction manager

2012-05-24 Thread Guillaume Nodet (JIRA)

 [ 
https://issues.apache.org/jira/browse/ARIES-856?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Guillaume Nodet resolved ARIES-856.
---

Resolution: Fixed

 Possible deadlock when starting the transaction manager
 ---

 Key: ARIES-856
 URL: https://issues.apache.org/jira/browse/ARIES-856
 Project: Aries
  Issue Type: Bug
  Components: Transaction
Reporter: Guillaume Nodet
Assignee: Guillaume Nodet
 Fix For: transaction-manager-0.3.1, 1.0




--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: 
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira




Re: Jenkins build is still unstable: Aries #1462

2012-05-24 Thread Holly Cummins
Is anyone looking at these build failures? There seem to be 13 failing tests.

Holly

On Tue, May 22, 2012 at 5:35 PM, Apache Jenkins Server
jenk...@builds.apache.org wrote:
 See https://builds.apache.org/job/Aries/changes



[jira] [Reopened] (ARIES-851) BundleTrackerCustomizer in BundleHookBundleTracker is called multiple times for the same event.

2012-05-24 Thread Christoph Gritschenberger (JIRA)

 [ 
https://issues.apache.org/jira/browse/ARIES-851?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Christoph Gritschenberger reopened ARIES-851:
-


The initial fix caused several JPA-related tests to fail

 BundleTrackerCustomizer in BundleHookBundleTracker is called multiple times 
 for the same event.
 ---

 Key: ARIES-851
 URL: https://issues.apache.org/jira/browse/ARIES-851
 Project: Aries
  Issue Type: Bug
  Components: JPA, Util
 Environment: Linux-3.0.0, Karaf-3.0.0-SNAPSHOT with Felix, openjdk-6
Reporter: Christoph Gritschenberger
Assignee: Guillaume Nodet
 Fix For: 1.0

 Attachments: ARIES-851-stacktrace.TXT, 
 BundleHookBundleTracker-solution1.patch, 
 BundleHookBundleTracker-solution2.patch, bundletracker-test-src.zip, 
 jpa-test-bundle-src.zip, persistence-feature.xml, 
 test.bundletracker.bundle-0.1.0-SNAPSHOT.jar, 
 test.jpa.bundle-0.1.0-SNAPSHOT.jar


 Steps to reproduce
 * Compile or get a karaf-3.0.0-SNAPSHOT and unpack it
 * Install openjpa-2.2.0 (using the attached feature-file)
 * Deploy a bundle containing a persistence-unit (attached jar for example)
 get the following Exception 4 times:
 {code}
 ERROR: Bundle org.apache.aries.jpa.container [78] EventDispatcher: Error 
 during dispatch. (java.lang.IllegalStateException: Unable to create 
 persistence units for the bundle test.jpa.test.jpa.bundle/0.1.0.SNAPSHOT as 
 previous persistence units already exist.)
 java.lang.IllegalStateException: Unable to create persistence units for the 
 bundle test.jpa.test.jpa.bundle/0.1.0.SNAPSHOT as previous persistence units 
 already exist.
 {code}
 See attachment for full stacktrace
 In subsequent startups the Exception is still thrown but the stacktrace does 
 not show in the console or the log.
 Also when using equinox the Stacktrace does not appear in either the console 
 or the log ever, but it's still thrown.
 The problem is that the PersistenceBundleManager#addingBundle is called 
 multiple times for the newly installed bundle.
 The PersistenceBundleManager is passed as a BundleTrackerCustomizer to a 
 BundleHookBundleTracker.
 The BundleHookBundleTracker stores it in it's own private attribute and also 
 passes it to it's super-constructor.
 So the customizer is stored in the instance twice.
 It's called by both BundleTracker$Tracked.customizerAdding and 
 BundleHookBundleTracker$Tracked.customizerAdding.
 Both Tracked recognize the bundle as untracked and call the 
 addingBundle-method.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: 
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira