You are viewing a plain text version of this content. The canonical link for it is here.
Posted to yarn-issues@hadoop.apache.org by "Karthik Kambatla (JIRA)" <ji...@apache.org> on 2013/08/27 22:32:55 UTC
[jira] [Commented] (YARN-1098) Separate out stateless services from
stateful services in the RM
[ https://issues.apache.org/jira/browse/YARN-1098?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13751687#comment-13751687 ]
Karthik Kambatla commented on YARN-1098:
----------------------------------------
Currently, RM is a {{CompositeService}} and its constituent services are started via {{CompositeService#serviceStart()}}. To separate out the services into two sets, and starting/stopping a single set, we should do one of the following:
# RM should start these services directly, and not call super.serviceStart().
# RM itself should be an abstract service with CompositeService fields - Stateless and Stateful services. These CompositeServices can then be started as and when we want them.
> Separate out stateless services from stateful services in the RM
> ----------------------------------------------------------------
>
> Key: YARN-1098
> URL: https://issues.apache.org/jira/browse/YARN-1098
> Project: Hadoop YARN
> Issue Type: Sub-task
> Components: resourcemanager
> Affects Versions: 2.1.0-beta
> Reporter: Karthik Kambatla
> Assignee: Karthik Kambatla
> Labels: ha
>
> From discussion on YARN-1027, it makes sense to separate out services that are stateful and stateless. The stateless services be HA-agnostic, while the stateful services need to be aware of HA.
> Separating them out and adding cleanup methods should help the cause.
--
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