openstack team mailing list archive

[<masumotok@xxxxxxxxxxxxx>]

Hello,

We made a decision that our live-migration branch make proceed to Cactus(not Bexar), 
since we considered that being rushed to Bexar is not good choice from QA point of view.
In addition, being rushed to Bexar is surely a burden to especially core developers, that is not our true will.

Regarding to the DB issues that core developers pointed out to me, 
we are still keep going to correct them to be merged at early Cactus, 
I kindly ask all of you to support us.

Thanks for your attention.
Kei Masumoto

-----Original Message-----
From: openstack-bounces+masumotok=nttdata.co.jp@xxxxxxxxxxxxxxxxxxx [mailto:openstack-bounces+masumotok=nttdata.co.jp@xxxxxxxxxxxxxxxxxxx] On Behalf Of Thierry Carrez
Sent: Wednesday, January 19, 2011 5:27 PM
To: openstack@xxxxxxxxxxxxxxxxxxx
Subject: [Openstack] Live migration branch status (and other Feature Freeze info)

As promised during yesterdays meeting, a quick summary of the situation
around feature freeze (FF) and the live migration branch.

All the major feature branches now landed with the notable exception of
the live-migration branch. There might still be a few "default behavior
change" branches that also need FF exceptions, but that means you can
safely switch to high QA gear and fix as many bugs as you can.

Mostly due to collective failure to sufficiently review the
live-migration branch in time, we had to revert the live migration
branch changes in trunk. The branch still has a standing FF exception to
be reproposed and merged in Bexar. However the time is running very
short: if a new branch fixing all the issues raised by termie (including
the lack of unit tests) can be proposed by Thursday and if it passes all
reviews by Friday, we will accept it for merging before the end of the week.

If it misses any of those (UTC) deadlines, then it's off the table and
deferred to Cactus. We can't afford to merge a database schema change
later without seriously jeopardizing Bexar QA.

Personally I'd prefer if the whole feature was calmly prepared and baked
in a branch over the next weeks, proposed when Cactus opens after
February 4, and reviewed/merged early then. It sounds much saner than
rushing the writing of feature test code two weeks before a release. But
I also understand that it would be a bit unfair to masumotok and NTT
Data, who proposed their branch early and worked a lot recently on
fixing the issues we raised. So I'll leave the choice between "still
trying" and "deferring" to them.

Regards,

-- 
Thierry Carrez (ttx)
Release Manager, OpenStack

_______________________________________________
Mailing list: https://launchpad.net/~openstack
Post to     : openstack@xxxxxxxxxxxxxxxxxxx
Unsubscribe : https://launchpad.net/~openstack
More help   : https://help.launchpad.net/ListHelp