You are viewing a plain text version of this content. The canonical link for it is here.
Posted to jira@kafka.apache.org by "Viktor Somogyi-Vass (Jira)" <ji...@apache.org> on 2022/07/21 15:44:00 UTC
[jira] [Updated] (KAFKA-13917) Avoid calling lookupCoordinator() in tight loop
[ https://issues.apache.org/jira/browse/KAFKA-13917?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Viktor Somogyi-Vass updated KAFKA-13917:
----------------------------------------
Affects Version/s: 3.3.0
3.2.1
> Avoid calling lookupCoordinator() in tight loop
> -----------------------------------------------
>
> Key: KAFKA-13917
> URL: https://issues.apache.org/jira/browse/KAFKA-13917
> Project: Kafka
> Issue Type: Improvement
> Components: consumer
> Affects Versions: 3.1.0, 3.1.1, 3.3.0, 3.1.2, 3.2.1
> Reporter: Viktor Somogyi-Vass
> Assignee: Viktor Somogyi-Vass
> Priority: Major
> Fix For: 3.3.0
>
>
> Currently the heartbeat thread's lookupCoordinator() is called in a tight loop if brokers crash and the consumer is left running. Besides that it floods the logs on debug level, it increases CPU usage as well.
> The fix is easy, just need to put a backoff call after coordinator lookup.
> Reproduction:
> # Start a few brokers
> # Create a topic and produce to it
> # Start consuming
> # Stop all brokers
> At this point lookupCoordinator() will be called in a tight loop.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)