You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@ambari.apache.org by "Jayush Luniya (JIRA)" <ji...@apache.org> on 2015/03/31 09:48:53 UTC
[jira] [Commented] (AMBARI-10023) Ambari framework to support
adding any new hadoop compatible file system at ease
[ https://issues.apache.org/jira/browse/AMBARI-10023?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14388195#comment-14388195 ]
Jayush Luniya commented on AMBARI-10023:
----------------------------------------
[~vijikarthi]
Thanks for taking this initiative.
I just wanted to comment on point 2.a in the design document
{noformat}
All file system service implementation should extend from common/HCFS service layer.
a) HDFS service implementation should extend from common services/HCFS service.
(Note, here the common/hdfs service is extending from common/hcfs service. This is a new
capability that is not implemented in current Ambari design.)
{noformat}
The stack inheritence actually does support a service in common-services to inherit from another service in common services. So you can have common-services/HDFS/1.0 inherit from common-services/HCFS/1.0. In common-services/HDFS/1.0/metainfo.xml you need to add
{noformat}
<extends>common-services/HCFS/1.0</extends>
{noformat}
> Ambari framework to support adding any new hadoop compatible file system at ease
> --------------------------------------------------------------------------------
>
> Key: AMBARI-10023
> URL: https://issues.apache.org/jira/browse/AMBARI-10023
> Project: Ambari
> Issue Type: Epic
> Components: ambari-server, ambari-web
> Reporter: Vijay Srinivasaraghavan
> Attachments: HCFS-Integration-Design Draft 1.pdf, Service Type-Reference Implementation-Patch.txt
>
>
> 1) Align Ambari’s design with Hadoop’s design, supporting pluggable filesystems.
> 2) Eliminate GlusterFS as a special case.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)