[
https://issues.apache.org/jira/browse/KAFKA-7791?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17031769#comment-17031769
]
Yaroslav Klymko edited comment on KAFKA-7791 at 2/6/20 5:12 PM:
----------------------------------------------------------------
[~Lothar] I think it depends on how "commit" being called, for instance if
method exposes error - it should not be logged by kafka client, as the caller
should handle it and decide what shall it do. However in case of such "commit"
called as part of automatic process or in a "fire and forget" manner - I
suppose we should log it.
The original issue I opened is for cases when I got error logs for calls that
expose errors to the caller, and the caller did no receive errors because kafka
client did retry and managed to retry successfully. And still the logs were
full of errors...
I'd personally decrease such log records to debug level. If for some reasons
that is not acceptable: info or war..., but never an error for successfully
retried operation and for method that exposes full result to the caller
was (Author: t3hnar):
[~Lothar] I think it depends on how "commit" being called, for instance of
method exposes error - it should not be logged be kafka client, as called
should handle it and decide what shell it do. However in case such "commit"
called as part of automatic process or in a "fire and forget" manner - I
suppose we should log it.
The original issue I opened is for cases when I got error logs for calls that
expose errors to caller, and called did no receive errors because kafka client
did retry and managed to retry successfully. And still the logs were full of
errors...
I'd personally decrease such log records to debug level. If for some reasons
that is not acceptable - info or max warn, but never and error for successful
retried operation and for method that exposes full result to the caller
> Not log retriable exceptions as errors
> --------------------------------------
>
> Key: KAFKA-7791
> URL: https://issues.apache.org/jira/browse/KAFKA-7791
> Project: Kafka
> Issue Type: Bug
> Components: clients
> Affects Versions: 2.1.0
> Reporter: Yaroslav Klymko
> Priority: Minor
>
> Background: I've spotted tons of kafka related errors in logs, after
> investigation I found out that those are harmless as being retried.
> Hence I propose to not log retriable exceptions as errors.
> Examples of what I've see in logs:
> * Offset commit failed on partition .. at offset ..: The request timed out.
> * Offset commit failed on partition .. at offset ..: The coordinator is
> loading and hence can't process requests.
> * Offset commit failed on partition .. at offset ..: This is not the correct
> coordinator.
> * Offset commit failed on partition .. at offset ..: This server does not
> host this topic-partition.
>
> Here is attempt to fix this: https://github.com/apache/kafka/pull/5904
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
