[
https://issues.apache.org/jira/browse/KAFKA-9909?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17106198#comment-17106198
]
Gopikrishna commented on KAFKA-9909:
------------------------------------
Appreciate [~guozhang] your response. There are two cases we are talking about.
# ill-formatted message: this can be handled very easily and pretty straight
forward. my example code just shown to replicate the issue but not having the
intention.
# Message received correctly, its good. but the application could not process
it due to various other dependencies on other DB/microservices. will retry
multiple times and also will process later by storing the partition and offset
details. that is the process i follow.
But checkpointing will help the second scenario as mentioned above. the code
shared will illustrate the first scenario but for second scenario also similar
to that as i am unable to the message currently, hence i want to have a
checkpoint to revisit later.
yes i agree, we can resend the message again if it is not processed, but
ideally it increases the count, and will be difficult to know the accurate
number of messages processed perfectly. Hope this is clear.
> Kafka Streams : offset control to Streams API
> ---------------------------------------------
>
> Key: KAFKA-9909
> URL: https://issues.apache.org/jira/browse/KAFKA-9909
> Project: Kafka
> Issue Type: Improvement
> Components: streams
> Affects Versions: 2.5.0
> Environment: All
> Reporter: Gopikrishna
> Priority: Minor
> Labels: Offset, commit
>
> Hello team, really inspired the way streams api is running today. I would
> like to have a feature to be flexible regarding the offset. when we write the
> processor api, processor context object can be used to commit the offset.
> this is not effective. but streams are controlling the offset. the moment the
> process method executed or scheduled window completed, the offset is
> committed automatically by streams internally.
> Like traditional kafka consumer, its better the context object should have
> complete control over the offset whether to commit or not. This will give
> more control to the api to handle failovers and especially when message
> cannot be processed, context should not commit the offset. Appreciate this
> can be implemented.
>
> h4. enable.auto.commit is by default false, but streams are committing
> automatically the offset.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
