what's a blocker bug? (was: [Release Planning 3.4] v3-4-test will be frozen tomorrow)

[Björn Jacke]

Hi,

On 2009-10-14 at 11:48 +0200 Karolin Seeger sent off:
> The following blockers exist right now:

How about renaming "blocker bugs" to "regression bugs"? A regression bug means
a bug that was introduced inside a stable release branch.

Taking back a look at the 3.4.1 release: There had been a lot of bugs being
fixed after 3.4.0 but the 3.4.1 release took quite a while to come because not
*all* important bugs had been fixed. Most people being bitten by bugs would
have been happy with a 3.4.1 release with the most important bugs being fixed.
We had lot's of duplicate bug reports in bugzilla because people had been
hitting the same bugs again and again - while there was already a fix for it.

As we have stable release branaches we shouldn't introduce new bugs by
definition. And by the way: IMHO "by definition" works quite well thanks to
bugzilla and patch reviews and last but not least thanks to Karolin's work.

As long as we do not introduce new bugs inside a release branch, an existing
(old) bug should not be able to *block* an outstanding bugfix relrease. If some
bug, whether it's important or not is not yet fixed, it will be fixed with the
next release.

For today's situation: I think it's not optimal if people don't get a 3.4.3
release which already contains a bunch of important fixes just because we are
waiting for some more fixeѕ to come.

I'd like to hear what others think about this. And sorry I didn't come up with
this proposal earlier during the policy discussion.

Cheers
Björn
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://lists.samba.org/pipermail/samba-technical/attachments/20091015/0b77018a/attachment.pgp>