Hi Biplob, If you know what you want, you can always write your custom AssignerWithPeriodicWatermarks that does your job. If you want to just increase the watermark, you could simply check if you have received any elements and if not, emit a watermark with the timestamp of the previous watermark + X.
Kostas > On Jun 16, 2017, at 3:28 PM, Biplob Biswas <revolutioni...@gmail.com> wrote: > > Hi Kostas, > > Thanks for the reply, makes things a bit more clear. > > Also, I went through this link and it is something similar I am trying to > observe. > http://apache-flink-user-mailing-list-archive.2336050.n4.nabble.com/Listening-to-timed-out-patterns-in-Flink-CEP-td9371.html > > I am checking for timed out events and when I am using event time, its > behaviour is non-deterministic. For one pattern it's generating a few > 'matched events' and for a different pattern no 'matched events'. And almost > no timedout events in any of the patterns unless I run the series of mock > events a second time, during which I get a series of anomaly events. > > I had a topic created with this issue but I didn't get any satisfactory > solutions there, so was testing it with processing time whether it works > even or not. > > https://gist.github.com/revolutionisme/cf675ceee1492b93be020d4526bc9d38 > https://gist.github.com/revolutionisme/38578e631f7a15f02cb2488f9fe56c76 > > I would really like to know how to increment the watermark without any > events coming in, such that at least the timedout events are emitted by the > system. > > > > > -- > View this message in context: > http://apache-flink-user-mailing-list-archive.2336050.n4.nabble.com/Flink-CEP-not-emitting-timed-out-events-properly-tp13794p13800.html > Sent from the Apache Flink User Mailing List archive. mailing list archive at > Nabble.com.
