[
https://issues.apache.org/jira/browse/STRATOS-717?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14075907#comment-14075907
]
Akila commented on STRATOS-717:
-------------------------------
Hi Nirmal,
Do you have any ideas on how to achieve this? As I see it, there are 2
approaches that we can use.
1. Let Cartridge Agent handle the replication configuration
Most of the DBMS's support replication/clustering for eg - MySQL,
MongoDB. But they need some additional configuration to be applied.
These configuration includes IP addresses of other nodes so Stratos
will have to handle that dynamically when instances spawned/deleted.
But we will face a big problem when we have a large deployment with
many manager nodes, network partitions and partitions. IMO, we cannot
use cartridge agent to do this configuration at large scale.
2. Cartridge Manager
I would like to propose that we introduce a component to handle
cartridge instance specific configuration. This might look like an
overkill but all the other alternatives seem to be less reliable. I
haven't gone through it in detail yet. But I do like the idea of
having a central component to facilitate dynamic configuration of
cartridge instances. We should be able to use Puppet to apply the
configuration.
WDYT?
Thanks.
--
Akila Ravihansa Perera
Software Engineer
WSO2 Inc.
http://wso2.com
Phone: +94 77 64 154 38
Blog: http://ravihansa3000.blogspot.com
> Data replication of data Cartridges
> -----------------------------------
>
> Key: STRATOS-717
> URL: https://issues.apache.org/jira/browse/STRATOS-717
> Project: Stratos
> Issue Type: New Feature
> Reporter: Nirmal Fernando
> Fix For: 4.1.0
>
>
> Currently, Stratos Data Cartridges cannot be auto-scaled due to the fact that
> the data cannot be replicated among two or more instances. We need to come up
> with a design which is reliable, faster etc.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
