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

Posted to jira@kafka.apache.org by "Jun Rao (JIRA)" <ji...@apache.org> on 2018/10/23 01:52:00 UTC

[jira] [Updated] (KAFKA-7299) batch LeaderAndIsr requests during auto preferred leader election

     [ https://issues.apache.org/jira/browse/KAFKA-7299?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Jun Rao updated KAFKA-7299:
---------------------------
    Fix Version/s: 2.0.1
                   1.1.2

Kevin, since this is a small patch, I merged it to both 1.1 and 2.0 branch.

> batch LeaderAndIsr requests during auto preferred leader election
> -----------------------------------------------------------------
>
>                 Key: KAFKA-7299
>                 URL: https://issues.apache.org/jira/browse/KAFKA-7299
>             Project: Kafka
>          Issue Type: Sub-task
>          Components: core
>    Affects Versions: 2.0.0
>            Reporter: Jun Rao
>            Assignee: huxihx
>            Priority: Major
>             Fix For: 1.1.2, 2.0.1, 2.1.0
>
>
> Currently, in KafkaController.checkAndTriggerAutoLeaderRebalance(), we call onPreferredReplicaElection() one partition at a time. This means that the controller will be sending LeaderAndIsrRequest one partition at a time. It would be more efficient to call onPreferredReplicaElection() for a batch of partitions to reduce the number of LeaderAndIsrRequests.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)