[
https://issues.apache.org/jira/browse/FLUME-2988?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15726404#comment-15726404
]
Attila Simon commented on FLUME-2988:
-------------------------------------
Hi [~udai],
Sorry for not getting back to you earlier. The change looks good with two small
corrections:
- as [~denes] pointed out Long.valueOf is not needed
- I would recommend adding this to closer to the top of process() function eg:
where we know for sure that we have a reference to an event from the
channel.take() (== where flume first attempt to drain an event and knows that
the channel was not empty). More specifically in an else path at
https://github.com/apache/flume/blob/trunk/flume-ng-sinks/flume-ng-kafka-sink/src/main/java/org/apache/flume/sink/kafka/KafkaSink.java#L169
I think if you open a pull request for this on https://github.com/apache/flume
would help the reviewing process.
> Kafka Sink metrics missing eventDrainAttemptCount
> -------------------------------------------------
>
> Key: FLUME-2988
> URL: https://issues.apache.org/jira/browse/FLUME-2988
> Project: Flume
> Issue Type: Bug
> Components: Sinks+Sources
> Reporter: Denes Arvay
> Assignee: Udai Kiran Potluri
> Priority: Minor
> Attachments: FLUME-2988.1.patch
>
>
> {{eventDrainAttemptCount}} doesn't get incremented in Kafka Sink, only the
> {{eventDrainSuccessCount}}
> (https://github.com/apache/flume/blob/trunk/flume-ng-sinks/flume-ng-kafka-sink/src/main/java/org/apache/flume/sink/kafka/KafkaSink.java#L210)
> resulting in misleading statistics (e.g. when calculating {{attemptCount -
> successCount}})
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
