[
https://issues.apache.org/jira/browse/GRIFFIN-253?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16879033#comment-16879033
]
Liao, Xiaoge commented on GRIFFIN-253:
--------------------------------------
It's not namenode ip, and it should be the ip of resource manager
> How to implement the HA in apache griffin?
> ------------------------------------------
>
> Key: GRIFFIN-253
> URL: https://issues.apache.org/jira/browse/GRIFFIN-253
> Project: Griffin
> Issue Type: Improvement
> Reporter: Jay Wong
> Priority: Critical
>
> I follow the official guide to deploy griffin in cluster mode. here is the
> link :
> [https://github.com/apache/griffin/blob/master/griffin-doc/deploy/deploy-guide.md]
> When I try to modify the conf properties, example:
> service/src/main/resources/application.properties
> #yarn url
> yarn.uri=[http://namenode1:8088|http://namenode1:8088/]
> is it will fail to run well if active-standby switch happen? (the case
> :namenode1 go down and namenode2 raise to the new master)
> How can I avoid the impact on Griffin when the switch happen?
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
