Re: About the retract of the calculation result of flink sql

Posted by Hequn Cheng on
URL: http://deprecated-apache-flink-user-mailing-list-archive.369.s1.nabble.com/About-the-retract-of-the-calculation-result-of-flink-sql-tp23503p23547.html

Hi clay,

Keyed group by: 

SELECT a, SUM(b) as d
FROM Orders
GROUP BY a

 

Non Keyed group by:

SELECT SUM(b) as d
FROM Orders

I would like to look into the problem. However, I can't find obvious problems from the sql. It would be great that can provide a minimal example to reproduce the issue. Also, use print sink to avoid sinking into multi kafka partitions, since it will also bring out of order problem.

Best, Hequn

On Mon, Oct 1, 2018 at 9:11 PM clay4444 <[hidden email]> wrote:

hi,Timo

I use env.setParallelism(1) in my code, I set the overall degree of
parallelism of the program to 1, so that some calculations will still be
parallelized?

clay,




--
Sent from: http://apache-flink-user-mailing-list-archive.2336050.n4.nabble.com/