Date: Fri, 05 Jun 1998 08:39:14 -0700 From: "Jordan K. Hubbard" <jkh@time.cdrom.com> To: Glen Foster <gfoster@gfoster.com> Cc: stable@FreeBSD.ORG Subject: Re: future of -STABLE (was: kernel compile problem) Message-ID: <29666.897061154@time.cdrom.com> In-Reply-To: Your message of "Fri, 05 Jun 1998 10:32:27 EDT." <199806051432.KAA19172@gfoster.intr.net>
next in thread | previous in thread | raw e-mail | index | archive | help
> What are the plans for a 3.0-STABLE branch? Would this be a natural > aftermath of a 3.0-RELEASE? Or, perhaps it is contemplated that > another mechanism for reconciling the sometimes-at-odds interests of > the development and production camps is in the works? I suppose it would be correct to say that it will be a natural aftermath of 3.0 going to release status, but not necessarily right after 3.0.0 is released, if you get my meaning. I'd expect the -stable designator to make the jump only after 3.0 has been in "mainstream" use for awhile and the early-adopters have come back with their early reports. Depending on how things look in the 2-6 months following the release of 3.0.0, we'll move the -stable tag over when we feel the time is right. > The 2.2 branch has been good to me but I am realizing the need to get > acquainted with 3.0 well before it is in my client's best interest to > move to it. For my purposes, some semblance of "stability," in the > predictability sense as well as the code quality/reliability sense is > necessary first. Yes, well, 2.2 will also hopefully continue to be good to its users until at least the end of this year, after which it will go into maintenance mode and an eventual state of complete stasis. > If it isn't asking for too much of a crystal ball, what can we expect > for the future of 3.0 and 2.2? Here's something I recently posted to comp.unix.freebsd.misc. It may prove instructive: From: jkh@time.cdrom.com (Jordan K. Hubbard) Newsgroups: comp.unix.bsd.freebsd.misc Subject: General release schedule for next 6 months... Date: 04 Jun 1998 08:38:08 -0700 There have been some questions about this lately, and it's probably not a bad time to issue a general status report, so here it is. <disclaimer> All dates given are subject to change without notice and really only constitute my best guesses given things as they stand now. Should something really pathological raise its head before release time (though I doubt it), all bets are off until the issue is dealt with. </disclaimer> The 2.2-stable branch, the most recent release for which is 2.2.6, will live on for probably two more releases: 2.2.7, scheduled for a mid-July release, and 2.2.8 probably following in early November. I also expect 2.2.8 to be the last major release on the 2.2-stable branch, it basically going into end-of-life status at that point (some support, but no significant development). The -current branch should spawn the first official 3.0 release in mid-October, hopefully no later than the 20th. I also do not expect the -stable designator to pass over to the -current branch until the release or two following that one, probably 3.0.5. This means that "mainstream" customers are still encouraged to stick with 2.2-stable until at least the end of the year, early spring of '99 probably being a fine time to start contemplating the new vistas of 3.0.x :-) I should also probably note that while it's our policy to be extremely conservative about recommending that anyone deploy commercial services on one of our "dot-zero" releases, -current has been really coming along rather nicely and there are actually a growing number of people running it in production who swear by it. Your Mileage May Vary, of course, and if you have even the slightest doubts about your ability to venture into new territory then you are encouraged to stick with 2.2.x, but there are probably many who'll have good success with adopting 3.0 even as early as Q4 '98. I've certainly been running it for some time on a number of my own servers and it hasn't mangled me yet. Also, for the record, 3.0 will definitely support both SMP and ELF (as well as a number of nifty features like kernel threads, POSIX ASYNC I/O, VM86 mode, etc etc). If things continue to go at the pace they're going, it may even be available for the DEC ALPHA architecture as well as our traditional x86 distribution - it's really too early to say, but initial signs are at least promising. I think that's everything, but if you have any more questions, you know where to find me. :) -- - Jordan Hubbard Co-founder/Release Manager, The FreeBSD Project Walnut Creek CDROM To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?29666.897061154>