[ 
https://issues.apache.org/jira/browse/CLOUDSTACK-4103?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Animesh Chaturvedi updated CLOUDSTACK-4103:
-------------------------------------------


ACS 4.3,0 feature freeze is Nov 8th. I will cut the 4.3 branch from master on 
that day. Please provide an update on your issue as a comment. 

> S3 reduced redundancy option should be per s3 resource and not a global 
> setting.
> --------------------------------------------------------------------------------
>
>                 Key: CLOUDSTACK-4103
>                 URL: https://issues.apache.org/jira/browse/CLOUDSTACK-4103
>             Project: CloudStack
>          Issue Type: Bug
>      Security Level: Public(Anyone can view this level - this is the 
> default.) 
>    Affects Versions: 4.2.0
>            Reporter: Thomas O'Dowd
>
> Issue 3028 (https://issues.apache.org/jira/browse/CLOUDSTACK-3028) added a 
> global option to control the setting of reduced redundancy (s3.rrs.enable). 
> The default of the option is false.
> The reduced redundancy is normally an object level setting on S3 storage. For 
> example, snapshots could be uploaded using reduced redundancy storage while 
> templates are stored on normal storage. Currently we put everything in one 
> bucket. Different S3 servers/buckets can also be added. The reduced 
> redundancy could be configured as per bucket so that different objects going 
> to that bucket can take up less storage resources. Eventually, we could also 
> have configuration to split different types of objects out to different 
> buckets but that needs more thought.
> John also mentioned in 3028:
>        I don't feel that a global option is the correct functionality. In my 
> opinion,
>        it should be a flag on the object store instance.
>        In general, I think we should employ global options sparingly due to 
> their
>        broad impact and the complexity they add to the operation of the 
> system.
> This likely won't be done for 4.2 so I'm going to mark 3028 as closed as it 
> solves the original issue where reduced redundancy was always being selected.



--
This message was sent by Atlassian JIRA
(v6.1#6144)

Reply via email to