FreeBSD 6.0 and onwards

[Scott Long]

Richard Bejtlich wrote:
> Scott Long wrote:
> 
> So the current plan is to branch RELENG_6 (aka 6-STABLE) sometime
> around May or June 2005. That will begin a 1-3 month freeze and
> stabilization process for the 6.0 release. After that is released, we
> will do 6.1, 6.2 and onwards at likely 4 month intervals. In May/June
> 2006 we'll look at doing RELENG_7, or we might wait until Nov/Dec 2006
> (12 months vs 18 months). The 5.4 release will likely be in Feb/March
> 2005, with a 5.5 release possibly in June/July, depending on where 6.0
> is. There may be 5.x releases after 6.0 if 6.0 turns out to not be as
> stable as needed (as is often the case with and .0 release).
> 
> --
> 
> Hi Scott,
> 
> I like the idea of time-oriented vs. feature-oriented releases,
> especially for planning projects.  This reminds me of the OpenBSD
> 6-month schedule.
> 
> I may be a little confused by your description of 6.0.  Do you mean to
> have 6.0 itself assume the STABLE title, and not a later 6.x version
> like 6.1, 6.2, or even 6.3?  As you know 5.0 appeared in January 2003
> but we are just now making 5.3 be STABLE.

5.x has been quite an anomoly.  We held back the '5-STABLE' designation
while we did work that we felt was needed to give the 5.x series 
long-term viability.  6.0 will benefit from the work that has been
done so far, and with the shorter scheduled time left on it, it's less
likely to get sidetracked by major changes.  We all know that any
".0" release also has some rough edges, but you don't identify and fix
those issues unless you bite the bullet and actually release it.

> 
> Looking at www.freebsd.org/releases/ it seems the 4.x tree has had
> almost a five year run, and 5.x nearly 3 years.  Only now however is
> 5.3 becoming a version more people will run on production systems,
> thanks to the STABLE title.  If 6.0 is truly going to be STABLE by
> this time next year, I wouldn't be surprised if people skip future 5.x
> releases until 6.0 arrives.

5.x has had slighly less that 2 years at this point.  If you count the
proposed 5.4 and possible 5.5, it will still come in at under three
years.  However, it will still be supported by the security team for
quite a while after that.  We've also introduced the concept of errata
fixes that will fix high-profile, non-security bugs for a particular
release.  So 5.3, 5.4, etc will still stay viable for quite some time
to come.  Just because 6-STABLE gets branched on June 1 doesn't mean
that 5.x is dead or that 6.0 is immediately suitable for enterprise
work.  It's just a stepping stone to having a 6.x release that is fully
capable.

> 
> Maybe 5.x should hold the STABLE title a little longer than you predict?
> 
> Just a few thoughts from a satisfied FreeBSD user.

One of the problems of holding a branch for too long is that you run
into support issues that become hard to maintain.  A good example is 4.x
and PERL.  We don't feel like we can update from 5.003 in 4.x because of
the API stability issues, but it's incredibly hard to maintain a ports
system that can deal with such an ancient version of PERL.  But again,
just because a new branch is made doesn't mean that the old branch
suddenly ceases to exist.  I would expect there to continue to be errata
and security fixes going into 5.x at the time that RELENG_7 is created
several years from now.

Scott