Hi all, I've just come across a FlinkKafkaProducer misconfiguration issue especially when a FlinkKafkaProducer is created without specifying a kafka partitioner then a FlinkFixedPartitioner instance is used, and all messages end up in a single kafka partition (in case I have a single task manager instance). Just curious if it was done on purpose? Don't you consider a round robin strategy more suitable for this case? For now if I need a round-robin strategy I have to explicitly pass null value for it Thanks in advance. Kind Regards, Mike Pryakhin smime.p7s (2K) Download Attachment |
Hi Mike,
The rationale behind implementing the FlinkFixedPartitioner as the default is so that each Flink sink partition (i.e. one sink parallel subtask) maps to a single Kafka partition. One other thing to clarify: By setting the partitioner to null, the partitioning is based on a hash of the record's attached key (the key retrieved from the `SerializationSchema`), not round-robin. To use round-robin partitioning, a custom partitioner should be provided. Note however, a round-robin partitioner will create a lot of network connections to all Kafka brokers on all Flink sink parallel subtasks, which can be quite a lot. To conclude this, I think the appropriate partitioning scheme depends on the actual case. For example, for some simple Flink job that only does some filtering of data and has no aggregation within the pipeline, the key hash based partitioning would probably be more ideal. For more complex pipelines that partitions the computation by key already, it could make sense that a direct mapping of a Flink sink partition to Kafka partition would do. On the other hand, considering that the key for each record is always "re-calculated" by the `SerializationSchema` in each Flink Kafka Producer sink partition, it might make sense to actually make the key hash partitioner as the default instead. Cheers, Gordon -- Sent from: http://apache-flink-user-mailing-list-archive.2336050.n4.nabble.com/ |
Sounds like adding a round robin partitioner to the set of readily available partitioners would make sense.
On Fri, Dec 1, 2017 at 5:16 AM, Tzu-Li (Gordon) Tai <[hidden email]> wrote: Hi Mike, |
Exactly, at least it's worth mentioning the partitioner used by default in case none was specified in the javadoc, because the default behavior might not seem obvious. Kind Regards, Mike Pryakhin
smime.p7s (2K) Download Attachment |
Free forum by Nabble | Edit this page |