maintaining many branches is definitely not what we want.

Further research is required to answer how to solve this problem

On Mon, Aug 3, 2015 at 8:11 PM, Luke Han <[email protected]> wrote:

> Question: can higher version Hadoop still comptabile with lower one?
>
> If yes, let's make it happen in 0.7.3 release.
>
> if no, I would like to say we should release with majority version and have
> guide for such difference,
> or we should release different version to support different hadoop, like
> 0.7.3 for Hadoop 2.4 and 0.7.4 for 2.6 ...
>
> Manage too many branches is not good idea,
>
> Thanks.
>
>
> Best Regards!
> ---------------------
>
> Luke Han
>
> On Fri, Jul 31, 2015 at 5:57 PM, hongbin ma (JIRA) <[email protected]>
> wrote:
>
> > hongbin ma created KYLIN-920:
> > --------------------------------
> >
> >              Summary: Support Kylin in higher version of hadoop stack
> >                  Key: KYLIN-920
> >                  URL: https://issues.apache.org/jira/browse/KYLIN-920
> >              Project: Kylin
> >           Issue Type: Improvement
> >     Affects Versions: v0.7.1
> >             Reporter: hongbin ma
> >             Assignee: hongbin ma
> >
> >
> > most of our dev/production environments are aligned with hdp 2.2 (
> >
> http://docs.hortonworks.com/HDPDocuments/HDP2/HDP-2.2.4/HDP_RelNotes_v224/index.html
> > )
> >
> > Some users are failing to build their cubes with higher versions of
> > hbase/haodop/hive, we should either:
> > 1. guide users to use specific hadoop versions for specific Kylin branch
> > 2. for each Kylin branch, fork sub-branches (not necessary if
> > configuration change will do it) that will work with different hadoop
> > versions.
> >
> >
> >
> > --
> > This message was sent by Atlassian JIRA
> > (v6.3.4#6332)
> >
>



-- 
Regards,

*Bin Mahone | 马洪宾*
Apache Kylin: http://kylin.io
Github: https://github.com/binmahone

Reply via email to