[
https://issues.apache.org/jira/browse/KAFKA-10794?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
limeng updated KAFKA-10794:
---------------------------
Description:
There is more than 6000 topics and 300 brokers in my kafka cluster, and we
frequently run kafka-preferred-replica-election.sh to rebalance our cluster.
But the reblance process spendes too more time and cpu resource like the
picture blow.
We find that the function:'controllerContext.allPartitions' is invoked too many
times.
!截屏2020-12-02 上午11.43.48.png!
was:
There is more than 6000 topics and 300 brokers in my kafka cluster, and we set
the config:'auto.leader.rebalance.enable=true'.
The reblance process spendes too more time and cpu resource.
We find that the function:'controllerContext.allPartitions' is invoked too many
times.
!截屏2020-12-02 上午11.43.48.png!
> Replica leader election is too slow in the case of too many partitions
> ----------------------------------------------------------------------
>
> Key: KAFKA-10794
> URL: https://issues.apache.org/jira/browse/KAFKA-10794
> Project: Kafka
> Issue Type: Improvement
> Components: core
> Affects Versions: 2.6.0, 2.5.1
> Reporter: limeng
> Priority: Major
> Attachments: clipboard_image_1606806581629.png, 截屏2020-12-02
> 上午11.43.48.png
>
>
> There is more than 6000 topics and 300 brokers in my kafka cluster, and we
> frequently run kafka-preferred-replica-election.sh to rebalance our cluster.
> But the reblance process spendes too more time and cpu resource like the
> picture blow.
> We find that the function:'controllerContext.allPartitions' is invoked too
> many times.
> !截屏2020-12-02 上午11.43.48.png!
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
