[
https://issues.apache.org/jira/browse/FLUME-1541?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13449415#comment-13449415
]
Hari Shreedharan commented on FLUME-1541:
-----------------------------------------
Juhani,
This is excellent. Do you think you could add this backoff mechanism to the SDK
load balancing Rpc client also. It might just be copy paste code for now, but
it would be good to refactor the load balancing code in the RpcClient and Sinks
to use the same libraries.
> Implement a SinkSelector for LoadBalancingSinkProcessor that includes
> failover mechanics
> ----------------------------------------------------------------------------------------
>
> Key: FLUME-1541
> URL: https://issues.apache.org/jira/browse/FLUME-1541
> Project: Flume
> Issue Type: Bug
> Components: Sinks+Sources
> Affects Versions: v1.3.0
> Reporter: Juhani Connolly
> Fix For: v1.3.0
>
>
> Implement a new sink selector with the following behaviors:
> - Round robin usage of sinks
> - Temporarily removing sinks from the round robin pool when they fail
> - Increasing timeout mechanics on sequential failures
> - Spreading load equally over surviving sinks(current Round Robin sends all
> the failed sinks events to the next sink in line)
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
