[Gluster-infra] Plugin Downstream / Upstream CI

Shubhendu Tripathi shtripat at redhat.com
Mon Feb 2 04:07:22 UTC 2015


On 02/01/2015 10:00 PM, Vijay Bellur wrote:
> On 01/29/2015 05:32 PM, Shubhendu Tripathi wrote:
>> Hi,
>>
>> As mentioned below, there is a suggestion to make glusterfs-hadoop part
>> of main gluster build/distribution.
>> Kindly guide us on the steps/guidelines to be followed for the same.
>>
>
> Just to be clear, do you want glusterfs-hadoop to be part of the CI 
> infrastructure hosted by gluster.org? Or have simultaneous releases of 
> glusterfs-hadoop along with glusterfs releases?
>
> Regards,
> Vijay

Hi Vijay,

I had a small talk with Bala on this.
I understand that glusterfs-hadoop plugin can be another Java project  
under CI infrastructure hosted by gluster.org.

Still need to understand the downstream mechanism. I understand it would 
be similar to oVirt downstream (RHSC) and we can build downstream on 
need basis and would be easier to align with RHS release.

Brad/Steve, please add your points.

Thanks and Regards,
Shubhendu

>
>> Thanks and Regards,
>> Shubhendu Tripathi
>>
>> On 01/27/2015 10:36 PM, Brad Childs wrote:
>>> Since glusterfs-hadoop is fairly slow and steady state I think we
>>> should look at removing our own CI and becoming part of the main
>>> gluster build/distribution:
>>>
>>> http://download.gluster.org/pub/gluster/glusterfs/LATEST/Fedora/fedora-22/x86_64/ 
>>>
>>>
>>>
>>> If we all agree I can locate the appropriate contacts to start the
>>> ball rolling.
>>>
>>> -bc
>>>
>>>
>>> ----- Original Message -----
>>>> From: "Jon Cope" <jcope at redhat.com>
>>>> To: "Brad Childs" <bdc at redhat.com>
>>>> Cc: "Stephen Watt" <swatt at redhat.com>
>>>> Sent: Tuesday, January 27, 2015 10:51:21 AM
>>>> Subject: Plugin Downstream / Upstream CI
>>>>
>>>> Hey Brad,
>>>>
>>>> I'd like to give Hank and Diane a green light to appropriate the
>>>> downstream
>>>> CI build/test cluster on rack08.  Before that, I wanted to check with
>>>> you
>>>> first. Is there any real value in maintaining them?
>>>>
>>>> Secondly,  the upstream CI is going to require a bit of work.  The
>>>> change in
>>>> the pom.xml file affecting the maven deploy causes a big fail
>>>> post-tests and
>>>> will require some re-working to sort out.  Again though, is the
>>>> system of
>>>> value any more?  How do you feel on instead pointing users to the
>>>> readme in
>>>> the github repo?
>>>>
>>>> Regards,
>>>> Jon
>>>>
>>
>> _______________________________________________
>> Gluster-infra mailing list
>> Gluster-infra at gluster.org
>> http://www.gluster.org/mailman/listinfo/gluster-infra
>>
>>
>



More information about the Gluster-infra mailing list