You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@kafka.apache.org by "Clemens Valiente (JIRA)" <ji...@apache.org> on 2017/03/01 15:30:45 UTC
[jira] [Assigned] (KAFKA-4824) add unique identifier to clientId
[ https://issues.apache.org/jira/browse/KAFKA-4824?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Clemens Valiente reassigned KAFKA-4824:
---------------------------------------
Assignee: Clemens Valiente
> add unique identifier to clientId
> ---------------------------------
>
> Key: KAFKA-4824
> URL: https://issues.apache.org/jira/browse/KAFKA-4824
> Project: Kafka
> Issue Type: Improvement
> Components: consumer, streams
> Affects Versions: 0.10.2.0
> Reporter: Clemens Valiente
> Assignee: Clemens Valiente
> Priority: Minor
>
> As discussed with [~damianguy]
> streams clientId is set to applicationId-threadId-consumer which of course wouldn’t be unique if running multiple instances per host.
> *Example use case:*
> deployment as docker container on marathon so it might end up multiple times on the same host and of course they're all running with the same configuration. All clients on the same host would have the same clientid
> *Problem*
> Consumers running with the same client id make it hard to identify lagging consumers correctly
> *Potential solution:*
> Internally we do have a processId. It might make sense if we attached that to the clientId. it would then be unique across instances
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)