[
https://issues.apache.org/jira/browse/KAFKA-2747?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14994508#comment-14994508
]
Geoff Anderson commented on KAFKA-2747:
---------------------------------------
[~becket_qin] The group id is the same, but the committed offset was out of
range after mirror maker restarted due to the fact that no messages had been
committed yet.
[~ewencp] opened https://issues.apache.org/jira/browse/KAFKA-2759 to track
whether this is desirable behavior.
> Message loss if mirror maker is killed with hard kill and then restarted
> ------------------------------------------------------------------------
>
> Key: KAFKA-2747
> URL: https://issues.apache.org/jira/browse/KAFKA-2747
> Project: Kafka
> Issue Type: Bug
> Reporter: Geoff Anderson
>
> I recently added simple failover to the existing mirror maker test
> (https://github.com/apache/kafka/pull/427) and found that killing mirror
> maker process with a hard kill resulted in message loss.
> The test here has two single-node broker clusters, one producer producing to
> the source cluster, one consumer consuming from the target cluster, and a
> single mirror maker instance mirroring data between the two clusters.
> mirror maker is using old consumer, zookeeper for offset storage
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
