[jira] [Commented] (KAFKA-14672) Producer queue time does not reflect batches expired in the accumulator

2023-02-01 Thread Jason Gustafson (Jira)


[ 
https://issues.apache.org/jira/browse/KAFKA-14672?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17683183#comment-17683183
 ] 

Jason Gustafson commented on KAFKA-14672:
-

[~kirktrue] Nope. Please do.

> Producer queue time does not reflect batches expired in the accumulator
> ---
>
> Key: KAFKA-14672
> URL: https://issues.apache.org/jira/browse/KAFKA-14672
> Project: Kafka
>  Issue Type: Bug
>Reporter: Jason Gustafson
>Assignee: Kirk True
>Priority: Major
>
> The producer exposes two metrics for the time a record has spent in the 
> accumulator waiting to be drained:
>  * `record-queue-time-avg`
>  * `record-queue-time-max`
> The metric is only updated when a batch is ready to send to a broker. It is 
> also possible for a batch to be expired before it can be sent, but in this 
> case, the metric is not updated. This seems surprising and makes the queue 
> time misleading. The only metric I could find that does reflect batch 
> expirations in the accumulator is the generic `record-error-rate`. It would 
> make sense to let the queue-time metrics record the time spent in the queue 
> regardless of the outcome of the record send attempt.



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


[jira] [Commented] (KAFKA-14672) Producer queue time does not reflect batches expired in the accumulator

2023-02-01 Thread Kirk True (Jira)


[ 
https://issues.apache.org/jira/browse/KAFKA-14672?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17683182#comment-17683182
 ] 

Kirk True commented on KAFKA-14672:
---

Mind if I grab this, [~hachikuji] 

> Producer queue time does not reflect batches expired in the accumulator
> ---
>
> Key: KAFKA-14672
> URL: https://issues.apache.org/jira/browse/KAFKA-14672
> Project: Kafka
>  Issue Type: Bug
>Reporter: Jason Gustafson
>Assignee: Kirk True
>Priority: Major
>
> The producer exposes two metrics for the time a record has spent in the 
> accumulator waiting to be drained:
>  * `record-queue-time-avg`
>  * `record-queue-time-max`
> The metric is only updated when a batch is ready to send to a broker. It is 
> also possible for a batch to be expired before it can be sent, but in this 
> case, the metric is not updated. This seems surprising and makes the queue 
> time misleading. The only metric I could find that does reflect batch 
> expirations in the accumulator is the generic `record-error-rate`. It would 
> make sense to let the queue-time metrics record the time spent in the queue 
> regardless of the outcome of the record send attempt.



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