[ 
https://issues.apache.org/jira/browse/AMBARI-2852?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15503777#comment-15503777
 ] 

Gervais Mickaƫl commented on AMBARI-2852:
-----------------------------------------

Does somebody has any feedback or updates on this issue?
Thanks

> Ambari installation and configuration to manage an existing deployed hadoop 
> Cluster
> -----------------------------------------------------------------------------------
>
>                 Key: AMBARI-2852
>                 URL: https://issues.apache.org/jira/browse/AMBARI-2852
>             Project: Ambari
>          Issue Type: New Feature
>          Components: ambari-web
>    Affects Versions: 1.4.0
>            Reporter: Anisha Agarwal
>
> In case you want to use Ambari for management of a live Hadoop cluster 
> (including monitoring and eco-system services), without deployment. Ambari 
> assumes it has the control of Hadoop installation. Suppose there is an 
> existing HDP Hadoop cluster and one wants to install Ambari to start 
> managing/monitoring Hadoop. The proposal is to allow Ambari installation on 
> the cluster but bypass the Hadoop (and eco-system) deployment. We should 
> provide the ability to import the existing configuration without over writing 
> existing setup. 
>  
> Proposed phases:
> 1. Provide on Ambari wiki documentation, leveraging Ambari API, all manual 
> steps to configure Ambari without redeploying Hadoop and its eco-system
> 2. Provide a template configuration file and one Utility (leaning toward 
> Python script) that automates the process of important an existing 
> configuration
> 3. Provide User interface to allow step above via UI



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Reply via email to