Hi The thoughs on this thread became much much wider for what I asked for that's:
How to better enlist the duplicate message count by the IdempotentConsumer on JMX? @Raul, your idea is for sure good (no doubt on that), but with 8600PS cowl under the cover you went too fast for my understanding! The scope of *this* discussion should be better kept in sync with the JIRA ticket already been mentioned. The Apache Camel team is currently working on the next 2.10 release which they like to provide that functionality with what we've got *today*. IMHO for now we better concentrate ourself not on the duplicate exchange IDs but only on the count of them (that's what the ticket description already says). Your proposal again would have another broader expectrum which could be also addressed by another ticket. What we intend to reach by this ticket is just counting of the duplicates, that's it. In this context there're two provided patches on the ticket: - Pros / Contras of the first patch: We don't count if JMX is enabled but there's a poor man Callback there - Pros / Contras of the second patch: We do count even if JMX is *not* enabled but the solution seems be more clear & precise Please note that no matter which approach, there's no danger of a possible memory-leak of List / Map etc. but just some more CPU cycles we would take while counting through a (thread-safe) AtomicLong counter. WDYT? Babak -- View this message in context: http://camel.465427.n5.nabble.com/DISCUSS-How-to-better-enlist-the-duplicate-message-count-by-the-IdempotentConsumer-on-JMX-tp5123118p5125297.html Sent from the Camel Development mailing list archive at Nabble.com.
