Yan, I've created https://issues.apache.org/jira/browse/FLINK-28975 to
track this.
Regards,
David
On Sun, Aug 14, 2022 at 6:38 PM Yan Shen wrote:
> Thanks David,
>
> I am working on a flink datastream job that does a temporal join of two
> kafka topics based on watermarks. The problem was quite
Thanks David,
I am working on a flink datastream job that does a temporal join of two
kafka topics based on watermarks. The problem was quite obvious when I
enabled idleness and data flowed through much faster with different results
even though the topics were not idle.
Regards.
On Mon, Aug 15,
Although I'm not very familiar with the design of the code involved, I also
looked at the code, and I'm inclined to agree with you that this is a bug.
Please do raise an issue.
I'm wondering how you noticed this. I was thinking about how to write a
failing test, and I'm wondering if this has some
Hi all,
After examining the source code further, I am quite sure
org.apache.flink.api.common.eventtime.WatermarksWithIdleness
does not work with FLIP-27 sources.
In org.apache.flink.streaming.api.operators.SourceOperator, there are
separate instances of WatermarksWithIdleness created for each spl
Hi,
I am using a org.apache.flink.connector.kafka.source.KafkaSource with a
watermark strategy like this:
WatermarkStrategy.forMonotonousTimestamps().withIdleness(Duration.ofSeconds(10))
I noticed that after a short while all the partitions seem to be marked as
idle even though there are message
