incubator-ambari-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Yusaku Sako (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (AMBARI-2466) Hive/Oozie database settings should accept custom JDBC URLs
Date Fri, 21 Jun 2013 23:47:19 GMT

    [ https://issues.apache.org/jira/browse/AMBARI-2466?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13690905#comment-13690905
] 

Yusaku Sako commented on AMBARI-2466:
-------------------------------------

Also, while implementing, I needed a way to check to see whether the cluster is already installed
or not.
I noticed that App.clusterStatus object is never loaded from the server upon page load.  When
I tried to access it, it returned incorrect status ("cluster not created"), rather than erroring
out; this is misleading and can potentially cause issues if we make assumptions about it being
correct, so I've added that to the page loading mechanism in cluster_controller.js.
I've also modified App.clusterStatus.updateFromServer so that the API call can happen asynchronously.


                
> Hive/Oozie database settings should accept custom JDBC URLs
> -----------------------------------------------------------
>
>                 Key: AMBARI-2466
>                 URL: https://issues.apache.org/jira/browse/AMBARI-2466
>             Project: Ambari
>          Issue Type: Bug
>          Components: client
>    Affects Versions: 1.2.4
>            Reporter: Yusaku Sako
>            Assignee: Yusaku Sako
>            Priority: Critical
>             Fix For: 1.2.4
>
>         Attachments: AMBARI-2466_branch-1.2.4.patch, AMBARI-2466.patch
>
>
> Ran into issues setting up Hive and Oozie with an Oracle database.
> 1. We are hard-coding port 1521 for the JDBC URL.  
> 2. There are two types of JDBC URLs for Oracle:
>  * jdbc:oracle:thin:@[HOST][:PORT]:SID
>  * jdbc:oracle:thin:@//[HOST][:PORT]/SERVICE
> We are making the assumption that it is the latter, but this may not work depending on
how Oracle is set up.
> 3. We prompt for the "Database Name".  In Oracle context, this could be the SID or SERVICE
NAME, but it's not clear what this is.
> As a solution to all of the above, we will construct the JDBC URL based on the database
type, host, and name for Hive/Oozie and present it to the user as an editable text field during
install.
> Post-install, the JDBC URL remains editable, but does not change automatically as changes
other database-related parameters.

--
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

Mime
View raw message