hbase-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From bo zhaobo <bzhaojyathousa...@gmail.com>
Subject Re: DISCUSS: Is there any idea about ARM CI for Hbase?
Date Thu, 27 Jun 2019 03:04:29 GMT
Hi Sean,

Thank you very much for the nice suggestions, you showed us a good way to
try, and the goal we should focus on.

I'm so glad to hear that you could help about this if possible. Yeah, "do
it in existing CI for ARM" that's what we want to do. That's very great if
you could give some details about how to add the resources and test into
the existing CI. Thank you, Sean.

Also, that's good if there will also be other contributors want to take
part in this work. ;-)

Thank you

Best Regards,

ZhaoBo



[image: Mailtrack]
<https://mailtrack.io?utm_source=gmail&utm_medium=signature&utm_campaign=signaturevirality5&>
Sender
notified by
Mailtrack
<https://mailtrack.io?utm_source=gmail&utm_medium=signature&utm_campaign=signaturevirality5&>
19/06/27
上午10:55:04

Sean Busbey <busbey@apache.org> 于2019年6月26日周三 下午7:16写道:

> Hi ZhaoBo!
>
> As an ASF project HBase is generally a "do-acracy", meaning generally folks
> who are willing to show up and do the work determine the direction we take.
> I'd say so long as you're invested in making ARM work you can push forward
> barring any specific objections.
>
> I'd thus presume the current consensus from this thread is you're good to
> go.
>
> I personally think it'd be best if you could integrate any needed CI
> testing into our existing processes on builds.a.o. To do that you'd have to
> first work with ASF Infra folks to get the appropriate VMs added as Jenkins
> workers. If you like I can give you specific pointers on this.
>
>
>
> On Wed, Jun 26, 2019, 04:14 bo zhaobo <bzhaojyathousandy@gmail.com> wrote:
>
> > Hi team,
> >
> > Sorry for disturbing, but we really need your kind suggestion. We want to
> > know the appropriate way that Hbase community can accept. Which I
> mentioned
> > the options in the past email for HBASE ARM support are:
> > 1. Introducing an new ARM CI(3rd party CI), which provided by OpenLab,
> > OpenLab will provide the ARM resources and people resources to develop
> and
> > maintain the CI status.
> >
> > 2. Still using the old CI(Apache Jenkins(https://builds.apache.org)),
> > OpenLab will provide more ARM resources to help test and build hbase on
> > ARM in
> > the old development process. That means we can just provide the ARM
> > resources
> > into your exist ARM test.
> > Notes: the ARM resources provided by OpenLab are Virtual Machines.
> >
> > Also we will provide the human resources(ME- ZhaoBo) to fix the ARM
> related
> > issues for Hbase, and maintain the ARM CI status is good, including
> option
> > 1 and 2. And I'm glad to do this and maintain in the future, if there are
> > some issues break Hbase on ARM, I will also try to report to JIRA and fix
> > in Hbase.
> >
> > We not only aim to just build Hbase success on ARM, but also plan to add
> > more test cases on ARM, such as the same with Hbase already done on X86.
> >
> > So I'm looking forward to get your(Hbase team) choice. Thank you very
> much.
> >
> > Best Regards,
> >
> > ZhaoBo
> >
> >
> > [image: Mailtrack]
> > <
> >
> https://mailtrack.io?utm_source=gmail&utm_medium=signature&utm_campaign=signaturevirality5&
> > >
> > Sender
> > notified by
> > Mailtrack
> > <
> >
> https://mailtrack.io?utm_source=gmail&utm_medium=signature&utm_campaign=signaturevirality5&
> > >
> > 19/06/26
> > 下午5:12:38
> >
> > bo zhaobo <bzhaojyathousandy@gmail.com> 于2019年6月23日周日 下午7:00写道:
> >
> > > Any updates from team?  Thanks you
> > >
> > >>
> > > [image: Mailtrack]
> > > <
> >
> https://mailtrack.io?utm_source=gmail&utm_medium=signature&utm_campaign=signaturevirality5&
> >
> > Sender
> > > notified by
> > > Mailtrack
> > > <
> >
> https://mailtrack.io?utm_source=gmail&utm_medium=signature&utm_campaign=signaturevirality5&
> >
> > 19/06/21
> > > 下午3:37:23
> > > *×*REMOVE <#m_-6684894170990945484_m_9060835149257889588_>
> > >
> >
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message