[
https://issues.apache.org/jira/browse/FLUME-2298?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13872221#comment-13872221
]
Ted Malaska commented on FLUME-2298:
------------------------------------
1. The underlining channel will be an existing Flume channel like memory
channel. DC is just a wrapper that enabled that it is backed up in a
distributed way.
2.1 So a master is alway in lock step with it's slaves. If a slave has issues
taking put,takes, or commits it is dropped and a new one is request.
2.2 If will be configurable how many slaves a master will look for. I assume 1
will be the default
3 The selection process is documented in the design. The master will check for
open slaves in the ZK. Then it will ask the slave if it can have it. After
confirmation the status changes will be made in the ZK.
Let me know if you have any more questions
> Distributed Channel
> -------------------
>
> Key: FLUME-2298
> URL: https://issues.apache.org/jira/browse/FLUME-2298
> Project: Flume
> Issue Type: New Feature
> Components: Channel
> Reporter: Ted Malaska
> Assignee: Ted Malaska
> Attachments: FlumeDistributedChannelDesign.0.1.pdf
>
>
> This channel will allow for events to be persisted with a plugable method on
> more then one agent or node.
> The goal is to gain the following benefits:
> 1. Events will continue to flow to sinks with out loss or with out large
> delay even in the case of node failure.
> 2. Protect against loss in the case of a complete single node failure
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)
