[ https://issues.apache.org/jira/browse/YARN-1232?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13784465#comment-13784465 ]
Alejandro Abdelnur commented on YARN-1232: ------------------------------------------ IMO they are different: * the HA ids are to differentiate the RM instances within the same cluster. This is used by the failover logic only, a user does not need to be aware of them when creating applications. * the clusterID is to differentiate different clusters. This is used by the user to indicate against which cluster they want to work. Because all the configurations a client needs to be aware, as Bikas, Karthik and I chatted during the yarn meetup last friday, an easy way to handle this would be for a client to specify the yarn-site.xml of the cluster he wants to connect or have a nested configurations in the single yarn-site.xml, one per cluster. If we do something like that, then the ids being introduced here will only be used for HA. > Configuration to support multiple RMs > ------------------------------------- > > Key: YARN-1232 > URL: https://issues.apache.org/jira/browse/YARN-1232 > Project: Hadoop YARN > Issue Type: Sub-task > Components: resourcemanager > Reporter: Karthik Kambatla > Assignee: Karthik Kambatla > Labels: ha > Attachments: yarn-1232-1.patch, yarn-1232-2.patch, yarn-1232-3.patch, > yarn-1232-4.patch, yarn-1232-5.patch, yarn-1232-6.patch > > > We should augment the configuration to allow users specify two RMs and the > individual RPC addresses for them. -- This message was sent by Atlassian JIRA (v6.1#6144)