automatic promotion of freeze exceptions?

[Chris Murphy]

On Nov 6, 2013, at 12:23 PM, Andre Robatino <robatino at fedoraproject.org> wrote:

> Adam Williamson <awilliam <at> redhat.com> writes:
> 
>> On Sat, 2013-11-02 at 14:52 +0000, Andre Robatino wrote:
>>> Do we or should we have a policy of automatically promoting unfixed freeze
>>> exceptions from Alpha to Beta, or Beta to Final? The concept doesn't make
>>> sense for blockers, since those have to be fixed at each stage before going
>>> to the next, but FEs don't.
>> 
>> This comes up every so often, but there isn't an explicit policy either
>> way. In practice it does not happen because there is no process or
>> automated mechanism to make it happen.
> 
> I wasn't thinking of automation, just a documented process that would give
> an individual permission to manually move an Accepted Beta FE to an Accepted
> Final FE (for example), without having to have a dedicated discussion in a
> meeting. I brought this up in the last Blocker Bug Review, and other
> people's feeling was that there was a danger in making them Accepted without
> discussion, since there may have been a lot of changes in the meantime. But
> isn't it standard practice to periodically reevaluate all Accepted Blockers
> and FEs anyway? If so, why single out these particular FEs for an additional
> evaluation, on top of the ones they would already be getting?

Alpha FE's becoming Beta FE's seems uncontroversial. For Beta FE's to become Final FE's may have unacceptable risk for final, that's acceptable for beta, therefore need re-evaluation.

I suggest spending an extra 20 seconds on this question during Beta FE review, rather than having to re-evaluate them again from scratch for Final. Alternatively enable someone in QA to reject FE's as too risky, without formal review.


Chris Murphy