Re: [VOTE] subsystem-bundle 1.0.0 (2nd attempt)

2013-04-26 Thread David Bosschaert
+1 David On 25 April 2013 20:53, John W Ross jwr...@us.ibm.com wrote: This is the second vote for the release of subsystem-bundle 1.0.0, an uber bundle containing subsystem-api 1.0.0 and subsystem-core 1.0.0. It encompasses the same content as in the previous vote but for the

Re: [VOTE] subsystem-bundle 1.0.0 (2nd attempt)

2013-04-26 Thread Jeremy Hughes
+1 I used the verify_staged_release.sh script again. On 26 April 2013 09:03, David Bosschaert david.bosscha...@gmail.com wrote: +1 David On 25 April 2013 20:53, John W Ross jwr...@us.ibm.com wrote: This is the second vote for the release of subsystem-bundle 1.0.0, an uber bundle

BlueprintContainer deadlock

2013-04-26 Thread Paul Slater1
an ABBA deadlock is possible when the BP container locks a ServiceRecipe in notifySatisfaction(): Thread A: notifySatisfaction() locks ServiceRecipe - unregister() - AbstractServiceReferenceRecipe.serviceChanged - ReferenceRecipe.untrack (tries to lock 'monitor') Thread B holds

Re: [VOTE] subsystem-bundle 1.0.0 (2nd attempt)

2013-04-26 Thread Mark Nuttall
+1 Thank you John for all your hard work on Subsystems. Regards, Mark On 25 April 2013 20:53, John W Ross jwr...@us.ibm.com wrote: This is the second vote for the release of subsystem-bundle 1.0.0, an uber bundle containing subsystem-api 1.0.0 and subsystem-core 1.0.0. It encompasses the

Re: [VOTE] subsystem-bundle 1.0.0 (2nd attempt)

2013-04-26 Thread Stuart McCulloch
On 25 Apr 2013, at 20:53, John W Ross wrote: This is the second vote for the release of subsystem-bundle 1.0.0, an uber bundle containing subsystem-api 1.0.0 and subsystem-core 1.0.0. It encompasses the same content as in the previous vote but for the

Idea: Using capabilities to express dependency on blueprint extender and eventually blueprint namespaces

2013-04-26 Thread Christian Schneider
Hi all, there is one regular problem people face when using blueprint or even spring dm in the same way. The blueprint xml is basically only passive. So if there is no blueprint extender present then the bundle will go active but will never actually do something as the blueprint context is

RE: Idea: Using capabilities to express dependency on blueprint extender and eventually blueprint namespaces

2013-04-26 Thread Timothy Ward
Hi Christian, I think this is a good idea, and I was already planning to do this in the next few weeks. There's actually an OSGi standard namespace for extenders called osgi.extender which we should use, and chapter 135.2 of the R5 enterprise specification declares that the blueprint extender

Re: Generating source for uber bundles

2013-04-26 Thread David Jencks
+1 thanks :-) david jencks On Apr 25, 2013, at 9:12 AM, Jeremy Hughes jpjhug...@gmail.com wrote: +1 .. if it helps some people and doesn't hurt the rest then it must be good. On 25 April 2013 17:04, John W Ross jwr...@us.ibm.com wrote: I agree, and, based on Stuart's response to

[jira] [Commented] (ARIES-1002) Missing export-service directive for PlatformTransactionManager

2013-04-26 Thread Maurice Betzel (JIRA)
[ https://issues.apache.org/jira/browse/ARIES-1002?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13643062#comment-13643062 ] Maurice Betzel commented on ARIES-1002: --- Probably related to this issue is that the

Jenkins build is unstable: Aries #1851

2013-04-26 Thread Apache Jenkins Server
See https://builds.apache.org/job/Aries/1851/