Hello all, I’d like to start discussion on KIP-708 [1] that aims to introduce rack aware standby task distribution in Kafka Streams. In addition to changes mentioned in the KIP, I’d like to get some ideas on additional change I have in mind. Assuming KIP moves forward, I was wondering if it makes sense to configure Kafka Streams consumer instances with the rack ID passed with the new StreamsConfig#RACK_ID_CONFIG property. In practice, that would mean that when “rack.id <http://rack.id/>” is configured in Kafka Streams, it will automatically translate into ConsumerConfig#CLIENT_RACK_ID config for all the KafkaConsumer clients that is used by Kafka Streams internally.
[1] https://cwiki.apache.org/confluence/display/KAFKA/KIP-708%3A+Rack+aware+Kafka+Streams+with+pluggable+StandbyTask+assignor <https://cwiki.apache.org/confluence/display/KAFKA/KIP-708:+Rack+aware+Kafka+Streams+with+pluggable+StandbyTask+assignor> P.S I have draft PR ready, if it helps the discussion moving forward, I can provide the draft PR link in this thread. Regards, Levani
