cloudstack-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Sheng Yang <sh...@yasker.org>
Subject Re: [Review Request] Re-enabling baremetal on master
Date Wed, 19 Jun 2013 18:36:11 GMT
On Wed, Jun 19, 2013 at 11:24 AM, Chip Childers
<chip.childers@sungard.com>wrote:

> On Wed, Jun 19, 2013 at 11:00:33AM -0700, Sheng Yang wrote:
> > Hi,
> >
> > I've created the https://reviews.apache.org/r/11977/  for review. The
> > branch re-enabled the baremetal for master. And all major bugs are
> cleaned.
> >
> > https://issues.apache.org/jira/browse/CLOUDSTACK-1610
> > https://issues.apache.org/jira/browse/CLOUDSTACK-1618
> > https://issues.apache.org/jira/browse/CLOUDSTACK-1614
> > https://issues.apache.org/jira/browse/CLOUDSTACK-1440
> >
> > In fact it's not a feature merge, because the code is already in MASTER
> > ready. We just disable it due to stability problem of 4.1 release. Now
> I've
> > tried to enable it, and the changeset is very small, mostly just revert
> the
> > old disabling baremetal codes, and fix some issues with introducing other
> > new features. Here is the summary:
>
> [snip]
>
> So David's standing veto was because of this comment (from him):
>
> "Baremetal seems to be suffering from a significant lack of unit tests
> and integration tests for marvin to consume. Let's get those in place
> before we consider re-enabling this."
>
> If I remember correctly, the reason that master has the code in it, is
> specifically because we decided that disabling the feature was easier to
> honor the veto than reverting all of the changes.
>
> That being said, have we addressed the original veto's concerns?
>

Not yet. I didn't realize it's vetoed due to this. Let me see what can I do
about it.

In fact the above bugs cannot be detected for unit test or marvin test(I
even not sure if they're valid bugs or not, but at that time Frank is on
vacation and nobody took a look at these then decided disable the feature,
and after I re-enabled them, everything works fine for me).

--Sheng

>
> -chip
>

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