Yes, the bullet in 104.5.3 regarding sending a CM_UPDATED event is incorrect and should not be in the spec. I have updated the spec errata at http://www2.osgi.org/Release4/Errata
-- BJ Hargrave Senior Technical Staff Member, IBM OSGi Fellow and CTO of the OSGi Alliance [EMAIL PROTECTED] office: +1 386 848 1781 mobile: +1 386 848 3788 "Simon Kaegi" <[EMAIL PROTECTED]> Sent by: [EMAIL PROTECTED] 2007-08-22 09:54 Please respond to OSGi Developer Mail List <[email protected]> To "OSGi Developer Mail List" <[email protected]> cc Subject Re: [osgi-dev] Configuration Admin Uncertainty Is there a bug in the spec here? Perhaps I'm missing something, but why is a CM_UPDATED event being created at all. In my mind the ManagedService is being updated. The configuration object is not being created or updated. -Simon ----- Original Message ----- From: "Felix Meschberger" <[EMAIL PROTECTED]> To: "OSGi Developer Mail List" <[email protected]> Sent: Wednesday, August 22, 2007 4:30 AM Subject: [osgi-dev] Configuration Admin Uncertainty > Hi all, > > While fixing an issue in the Apache Felix Configuration Admin Service > implementation, I ran accross two uncertainties, which I hope may be > solved: The Configuration Admin Service spec says in section 104.5.3 > (Configuring Managed Services), that a CM_UPDATED event has to be sent > out after a ManagedService has been registered. No such statement may be > found for Managed Service Factories. > > My first question is: I assume section 104.5.3 intends to send the > CM_UPDATED event after updating the Managed Service regardless of > whether a configuration exists for the Managed Service or not. > > My second question is: Is it correct, that after calling the > ManagedServiceFactory.updated method no CM_UPDATED event has to be sent > for a newly registered ManagedServiceFactory ? This would seem > asymetric, because this is the only time that an updated method is > called without a subsequent event. > > The Apache Felix implementation currently answers "yes" to question 1 > (send an event regardless of Configuration existence) and "yes" to the > second question (not sending a CM_UPDATED event). > > Thanks for any clarifications. > > Regards > Felix > > _______________________________________________ > OSGi Developer Mail List > [email protected] > http://www2.osgi.org/mailman/listinfo/osgi-dev _______________________________________________ OSGi Developer Mail List [email protected] http://www2.osgi.org/mailman/listinfo/osgi-dev _______________________________________________ OSGi Developer Mail List [email protected] http://www2.osgi.org/mailman/listinfo/osgi-dev
