[jira] [Commented] (MRM-1963) S3 Blob OAK Support for Data Storage

["Martin Stockhammer (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/MRM-1963?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16178824#comment-16178824 ] 

Martin Stockhammer commented on MRM-1963:
-----------------------------------------

This is not as simple as it looks. 
Currently the Repository files are all stored in the filesystem and the Maven libraries, used to manage the artifacts and index files, use the filesystem API. JCR Oak is only used for storing metadata, not the artifacts.
I do not think there is a way to implement other storage types in the near future. 

> S3 Blob OAK Support for Data Storage
> ------------------------------------
>
>                 Key: MRM-1963
>                 URL: https://issues.apache.org/jira/browse/MRM-1963
>             Project: Archiva
>          Issue Type: Wish
>          Components: Design
>            Reporter: David W Klotz
>
> It would be really cool if I could utilize S3 as the storage backend for archiva. It looks like JCR-OAK supports S3 Blob storage through a plugin:
> https://jackrabbit.apache.org/oak/docs/plugins/blobstore.html
> I assume the change would be adding to the RepositoryFactory:
> https://github.com/apache/archiva/blob/master/archiva-modules/plugins/metadata-store-jcr/src/main/java/org/apache/archiva/metadata/repository/jcr/RepositoryFactory.java
> This would enable a nice distributed setup, and would allow multiple archiva front-ends to connect to s3 to store and pull files.



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)