ARRRRGH! Guys, who's breaking -STABLE's GMIRROR code?!

Jamie Bowden ragnar at sysabend.org
Thu Sep 14 08:44:14 PDT 2006


On 9/9/06, Mark Andrews <Mark_Andrews at isc.org> wrote:

> > Yeah, -STABLE is what you should run if you want stable code, right?

>         No. STABLE means STABLE API.

>         If you want stable code you run releases.  Between releases
>         stable can become unstable.  Think of stable as permanent
>         BETA code.  Changes have passed the first level of testing
>         in current which is permanent ALPHA code.

No, this is what it means now.  I've been running FreeBSD since 1.1,
and -STABLE used to mean exactly that.  The developement branch was
-C, and -S was where things went after extensive testing.  You were
not allowed to break -S or Jordan would rip your fingers off.  This
change to the current structure wasn't meant to be permanent when it
was done (between 4 and 5, IIRC), and was only done out of necessity
because the changes across that major release were huge.

FreeBSD needs an interim track that mirrors what -STABLE used to be,
which is a track between point releases that can be relied upon (and
RELEASE_x_y doesn't work, since it only addresses security and bugs
deemed worthy, which most aren't).

-- 
Jamie Bowden
-- 
"It was half way to Rivendell when the drugs began to take hold"
Hunter S Tolkien "Fear and Loathing in Barad Dur"
Iain Bowen <alaric at alaric.org.uk>


More information about the freebsd-stable mailing list