You are viewing a plain text version of this content. The canonical link for it is here.

Posted to jira@kafka.apache.org by "Sandeep (Jira)" <ji...@apache.org> on 2020/05/15 11:10:00 UTC

[jira] [Commented] (KAFKA-9678) Introduce bounded exponential backoff in clients

    [ https://issues.apache.org/jira/browse/KAFKA-9678?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17108175#comment-17108175 ] 

Sandeep commented on KAFKA-9678:
--------------------------------

I would like to work on this but couldnt assign myself

> Introduce bounded exponential backoff in clients
> ------------------------------------------------
>
>                 Key: KAFKA-9678
>                 URL: https://issues.apache.org/jira/browse/KAFKA-9678
>             Project: Kafka
>          Issue Type: Improvement
>          Components: admin, consumer, producer 
>            Reporter: Guozhang Wang
>            Assignee: Sanjana Kaundinya
>            Priority: Major
>              Labels: needs-kip
>
> In all clients (consumer, producer, admin, and streams) we have retry mechanisms with fixed backoff to handle transient connection issues with brokers. However, with small backoff (many defaults to 100ms) we could send 10s of requests per second to the broker, and if the connection issue is prolonged it means a huge overhead.
> We should consider introducing upper-bounded exponential backoff universally in those clients to reduce the num of retry requests during the period of connection partitioning.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)