On Fri, Jan 8, 2010 at 1:07 PM, Bruce Momjian <bruce(at)momjian(dot)us> wrote:
> Robert Haas wrote:
>> > You can't move from commitfest to beta until all _known_ bugs are
>> > fixed/addressed, and you can't move from beta to RC using the same
>> > criteria.
>> Hmm. For 8.4, I don't think we actually fixed "all" known bugs - I
>> think we made a decision about which ones had to be fixed and which
>> ones we were going to push off, and then did so.
> Right, we have to decide which ones get pushed to the TODO list, but
> also consider that several obvious bugs got into the 8.4.0 release,
> which is something we are going to try to avoid this time.
True. It's worth being clear, though I'm sure we're on the same
wavelength here, that those bugs didn't come from the open items list.
They came from stabilization issues related to large patches
committed in that release cycle. That's why it seems to me that
pushing off some of the large patches that were not submitted until
the final CommitFest is likely to speed up the release.
We discussed doing this at the very beginning of 8.4 release cycle
and, the more I think about it, the more I think it's not fair not to
go ahead and do it. Otherwise, we're rewarding people for ignoring a
guideline that was discussed, and punishing (1) the people who have
refrained from submitting large patches at the last minute, (2) people
who would like to see their already-committed patches released on a
reasonable time frame, and (3) people who don't want the tree to be
frozen for a near-eternity while we shake out all the bugs that these
large, last-minute patches introduce. We're also increasing the
chances the the final release will contain undiscovered bugs, since
they will have had ONLY the beta period, and no part of the
development cycle, to shake out.
In response to
pgsql-hackers by date
|Next:||From: Bruce Momjian||Date: 2010-01-08 18:29:01|
|Subject: Re: damage control mode|
|Previous:||From: Tom Lane||Date: 2010-01-08 18:24:45|
|Subject: Re: Listen / Notify - what to do when the queue is full |