You are correct that elements are waiting until a watermark with a higher timestamp than theirs (or the patterns timeout) arrives.
Now for the Watermark emitter, 1) how do you measure the 10sec in processing time and ii) by how much do you advance the watermark. If you advance it by a lot, then the elements that may come later, they may be considered late. \ For the first question I suppose that you set the watermark interval to 10 sec and if you see that there were no elements in between, you consider it inactivity right? How do you estimate that there were no elements? You have a flag in the emitter? > On Jun 20, 2017, at 3:57 PM, Biplob Biswas <revolutioni...@gmail.com> wrote: > > Hi Kostas, > > I have out-of-orderness of around 5 seconds from what I have observed but > that too from events coming from a different topic. The initial topic > doesn't have out-of-order events still I have added a generous time bound of > 20 seconds. Still, I will try for a higher number just in order to check a > bit more. > > The second problem you suggested sounds more interesting because when I > print my events which have been ingested I see all the events. It's just > that those events are neither generating a match nor are generating an > anomaly which felt a bit weird for me. > > In what cases the elements could be buffered and waiting forever? I expected > that even if they are buffered when the timeout happens I would get all the > events which were waiting for a match, provided the watermark is reset and > that I am doing by setting the autoWatermarkinterval and increasing the > watermark when there are no new events and after every 10-second system > time. > > Regards, > Biplob > > > > -- > View this message in context: > http://apache-flink-user-mailing-list-archive.2336050.n4.nabble.com/Flink-CEP-not-emitting-timed-out-events-properly-tp13794p13857.html > Sent from the Apache Flink User Mailing List archive. mailing list archive at > Nabble.com.
