same parallelism with different taskmanager and slots, skew occurs

Hi, there Recently I run streaming benchmark with flink 1.5.2 standalone on the cluster with 4 machines(1 as master and others as workers), it appears different result as below: (1). when I set the parallelism with 96, source, sink and middle operator parallelism all set to 96, start 3

Problem when use kafka sink with EXACTLY_ONCE in IDEA

Hi， I encountered an error while running the kafka sink demo in IDEA. This is the complete code: import java.util.Properties import org.apache.flink.api.common.serialization.SimpleStringSchema import org.apache.flink.runtime.state.filesystem.FsStateBackend import

Re: using updating shared data

Im trying to understand your use case. What is the source of the data ? FS ,KAFKA else ? On Tue, Jan 1, 2019 at 6:29 PM Avi Levi wrote: > Hi, > I have a list (couple of thousands text lines) that I need to use in my > map function. I read this article about broadcasting variables >

using updating shared data

Hi, I have a list (couple of thousands text lines) that I need to use in my map function. I read this article about broadcasting variables or using distributed cache

Re: RuntimeException with valve output watermark when using CoGroup

To return to this old thing, this was basically user error. The second of the transformations was keying by a field that was sometimes null after the first transformation. (This was supposed to never happen, but then it did happen in production.) The confusing part is where the exception occurs.