You are viewing a plain text version of this content. The canonical link for it is here.
Posted to jira@kafka.apache.org by "Chia-Ping Tsai (Jira)" <ji...@apache.org> on 2026/07/06 14:15:00 UTC

[jira] [Assigned] (KAFKA-19117) Client Throttling Log messages should be of log level - WARN (Java client)

     [ https://issues.apache.org/jira/browse/KAFKA-19117?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Chia-Ping Tsai reassigned KAFKA-19117:
--------------------------------------

    Assignee: Siddhartha Devineni

> Client Throttling Log messages should be of log level - WARN (Java client)
> --------------------------------------------------------------------------
>
>                 Key: KAFKA-19117
>                 URL: https://issues.apache.org/jira/browse/KAFKA-19117
>             Project: Kafka
>          Issue Type: Improvement
>          Components: clients, logging
>    Affects Versions: 3.6.2
>            Reporter: Ravi Kalasapur
>            Assignee: Siddhartha Devineni
>            Priority: Minor
>             Fix For: 4.4.0
>
>
> We experienced an outage in our application due to a built up Kafka lag, which we eventually discovered to be the problem with Kafka Clients being throttled.
> Even when using the confluent cluster, the dashboard does point to exceeding the quota of the principal.
> It will be nice, if the log message on the client side is a WARNING. This helps those who don't have visibility to cluster, and just look at the problem from the client side, setup alerts etc. (instead of enabling TRACE to figure out where the problem is) 
>  
> NetworkClient#maybeThrottle
> {code:java}
>  log.trace("Connection to node {} is throttled for {} ms until timestamp {}", nodeId, throttleTimeMs, now + throttleTimeMs);  {code}
>  



--
This message was sent by Atlassian Jira
(v8.20.10#820010)