Thanks Mark for the quick response: Opened JIRA: https://issues.apache.org/jira/browse/TUSCANY-2851
to remind us to address this as agreement here is reached. Scott On Tue, Feb 17, 2009 at 3:41 AM, Mark Combellack <mcombell...@apache.org>wrote: > Hi Scott, > > > > There are two outstanding OASIS issues regarding this problem. Basically, > the spec does not say what should happen but it should. > > > > The relevant issues are: > > > > http://www.osoa.org/jira/browse/JAVA-65 There is no lifecycle > defined for SCA Components > > http://www.osoa.org/jira/browse/JAVA-53 What happens if init() > throws a runtime exception > > > > JAVA-65 provides a proposal that will define the lifecycle for SCA > Components and explicitly details what to do with exceptions. The proposal > has not yet been accepted so may well be tweaked before it is included in > the specification. > > > > Thanks, > > > > Mark > ------------------------------ > > *From:* Scott Kurz [mailto:scottk...@gmail.com] > *Sent:* 16 February 2009 23:54 > *To:* dev@tuscany.apache.org > *Subject:* What to do with instance when RuntimeException thrown from > @Init? > > > > I feel like I remember seeing this conversation before but can't find it in > my mail search. > > Just noticing that if we throw a RuntimeException from an @Init method in a > composite-scoped component ... then the instance gets used anyway even > though it may have failed to initialize. > > Was thinking it would be a useful service for the container to provide to > discard the instance, under the assumption that the next time the service is > invoked, we'll retry the instance creation, since maybe later the @Init will > complete successfully? > > Don't see anything in a very quick scan of the specs but I didn't look > thoroughly. > > Any thoughts on this? > > Thanks, > Scott >