That I agree with which I believe is the aggregator issue u brought forward, correct?
If so why don't u go ahead and open a Jira ticket describing the issue so we can capture the relevant details. ses > On Oct 14, 2013, at 8:36 AM, flovansi <florent.vansilie...@swift.com> wrote: > > Yes I understand the purpose of the timeout but what if an exchange has just > been read by the SJMS consumer and the timeout is triggered? The processing > of the timeout commits the transaction while the exchange could still be > inflight and could fail. > > In general, the commit of the transaction triggered by the timeout should > know if an exchange is still inflight and only commit (or rollback in case > the exchange is failed) the transaction when the exchange is completed. > > I mean transaction are meant to not loose any messages. But in the SJMS > implementation, it seems that a message could be lost if the timeout is > triggered when an exchange is not completed. > > Am I right? > > > > -- > View this message in context: > http://camel.465427.n5.nabble.com/SJMS-implementation-Batch-Consumer-tp5741291p5741530.html > Sent from the Camel Development mailing list archive at Nabble.com.
