[
https://issues.apache.org/jira/browse/KAFKA-2017?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14964302#comment-14964302
]
Jason Gustafson commented on KAFKA-2017:
----------------------------------------
I'm hoping that we can modify the ConsumerMetadataRequest to return group
metadata (including member assignments) when it is received by the group's
coordinator. Admin tools would have to send one metadata request to discover
the coordinator and one to get the associated metadata, but I think that would
address the major operational concern. Perhaps if we have that, then we can
adopt [~junrao]'s suggestion and punt on persistence for now. However,
supporting this feature going forward would probably push us toward Kafka
persistence for group metadata, unless we're willing to store full assignments
(which can be quite large) in Zookeeper. Any thoughts?
> Persist Coordinator State for Coordinator Failover
> --------------------------------------------------
>
> Key: KAFKA-2017
> URL: https://issues.apache.org/jira/browse/KAFKA-2017
> Project: Kafka
> Issue Type: Sub-task
> Components: consumer
> Affects Versions: 0.9.0.0
> Reporter: Onur Karaman
> Assignee: Guozhang Wang
> Fix For: 0.9.0.0
>
> Attachments: KAFKA-2017.patch, KAFKA-2017_2015-05-20_09:13:39.patch,
> KAFKA-2017_2015-05-21_19:02:47.patch
>
>
> When a coordinator fails, the group membership protocol tries to failover to
> a new coordinator without forcing all the consumers rejoin their groups. This
> is possible if the coordinator persists its state so that the state can be
> transferred during coordinator failover. This state consists of most of the
> information in GroupRegistry and ConsumerRegistry.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
