Yash Mayya created KAFKA-15888:
----------------------------------
Summary: DistributedHerder log context should not use the same
client ID for each Connect worker by default
Key: KAFKA-15888
URL: https://issues.apache.org/jira/browse/KAFKA-15888
Project: Kafka
Issue Type: Bug
Components: connect, KafkaConnect
Reporter: Yash Mayya
Assignee: Yash Mayya
By default, if there is no "{{{}client.id"{}}} configured on a Connect worker
running in distributed mode, the same client ID ("connect-1") will be used in
the log context for the DistributedHerder class in every single worker in the
Connect cluster. This default is quite confusing and obviously not very useful.
Further, based on how this default is configured
([ref|https://github.com/apache/kafka/blob/150b0e8290cda57df668ba89f6b422719866de5a/connect/runtime/src/main/java/org/apache/kafka/connect/runtime/distributed/DistributedHerder.java#L299]),
it seems like this might have been an unintentional bug. We could simply use
the workerId (the advertised host name and port of the worker) by default
instead, which should be unique for each worker in a cluster.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
