flink-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Tony Wei <tony19920...@gmail.com>
Subject Re: Consuming a Kafka topic with multiple partitions from Flink
Date Tue, 29 Aug 2017 07:00:14 GMT
Hi Isuru,

The way to assign partitions by FlinkKafkaConsumer can refer to this java
document. (
That means your partitions should have increasing id one by one, then each
subtask would be assigned equally.
If your partition ids are 0, 3, 6, then only one subtask was assigned with
parallelism set to 3.

In my case, I created a new topic with two partitions which id are 0 and 1,
and my consumer with 2 subtasks were assigned fairly with one to one
You can check your topic's  setup or create another topic to try this out.

Hope this will help you.

Best Regards,
Tony Wei

2017-08-29 12:26 GMT+08:00 Isuru Suriarachchi <isurues@gmail.com>:

> Hi all,
> I'm trying to implement a Flink consumer which consumes a Kafka topic with
> 3 partitions. I've set the parallelism of the execution environment to 3 as
> I want to make sure that each Kafka partition is consumed by a separate
> parallel task in Flink. My first question is whether it's always guaranteed
> to have a one-to-one mapping between Kafka partitions and Flink tasks in
> this setup?
> So far, I've just setup a single Kafka broker and created a topic with 3
> partitions and tried to consume it from my flink application with
> parallelism set to 3 (all on same machine). I see 3 parallel processes of
> each operation being created on Flink log. However, when I execute the
> Flink job, messages from all 3 Kafka partitions are consumed by a single
> task (Process (3/3)). Other two parallel tasks are idling. Am I mission
> something here? In addition to setting the parallelism, is there any other
> configuration that I have to do here?
> Here are the details about my setup.
> Kafka version:
> Flink version: 1.3.1
> Connector: FlinkKafkaConsumer010
> Thanks,
> Isuru

View raw message