Hi,
1) By default, Flink's Kafka connector is polling data from Kafka every
100ms. There's a configuration key "flink.poll-timeout" to change the
frequency. I don't have experience with these internal log messages from
Kafka, but since they are on INFO level (and if you don't see any
unexpected da
Thanks Roberts,
About your questions, I don't have yet a real estimation regarding the
number of records received by the pipeline but I guess that the pipeline
could be idle for several minutes (I don't think that for hours).
My concern comes to me from two aspects:
1) I saw multiple lines in
Hi Andrés,
sorry for the late reply.
1. The slots are released, when the streaming pipeline ends. In principle,
it is not a problem when a slot is allocated, even when not processing any
incoming messages. So you are not doing something wrong. How many records
do you receive per pipeline? (are the