From owner-freebsd-hackers Thu Nov 27 14:40:12 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id OAA28123 for hackers-outgoing; Thu, 27 Nov 1997 14:40:12 -0800 (PST) (envelope-from owner-freebsd-hackers) Received: from smtp03.primenet.com (smtp03.primenet.com [206.165.6.133]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id OAA28117; Thu, 27 Nov 1997 14:40:04 -0800 (PST) (envelope-from tlambert@usr07.primenet.com) Received: (from daemon@localhost) by smtp03.primenet.com (8.8.8/8.8.8) id PAA00322; Thu, 27 Nov 1997 15:40:29 -0700 (MST) Received: from usr07.primenet.com(206.165.6.207) via SMTP by smtp03.primenet.com, id smtpd000298; Thu Nov 27 15:40:22 1997 Received: (from tlambert@localhost) by usr07.primenet.com (8.8.5/8.8.5) id PAA12847; Thu, 27 Nov 1997 15:38:07 -0700 (MST) From: Terry Lambert Message-Id: <199711272238.PAA12847@usr07.primenet.com> Subject: Re: BIND 8.1.1 To: peter@netplex.com.au (Peter Wemm) Date: Thu, 27 Nov 1997 22:38:06 +0000 (GMT) Cc: tlambert@primenet.com, shimon@simon-shapiro.org, peter@freebsd.org, hackers@freebsd.org, julian@whistle.com, steve@visint.co.uk, nate@mt.sri.com In-Reply-To: <199711270204.KAA01487@spinner.netplex.com.au> from "Peter Wemm" at Nov 27, 97 10:04:50 am X-Mailer: ELM [version 2.4 PL23] Content-Type: text Sender: owner-freebsd-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > The main "problem" is that 8.1.1 has been out for a while and a fair few > significant bugs have been found and fixed - but I at least am missing some > of the fixes that I've seen followup discussions about on the lists. If > the ISC collected them together and released 8.1.2 that'd be a big help. This is what "vendor branches" are for... 8-). > The libc resolver interface is a seperate animal to the named server. We > don't have to change libc just to get bind 8.1.x up. I'm not entirely > sure that I'm completely in favour of the way that the irs resolver is > done, I'd like to see room for adding new types at run time, perhaps via > .so objects etc. Some of the basic types would have to be permanently > present though so that static binaries (ugh!) are still useable. Me too. I'd like to see the same for address families so I can dynamically put the ISO and XNS networking back in if I want to (which I do). But at least the new framework is not fundamentally opposed to the idea (ie: it would fit pretty well, but require an irs.conf file change to implement). It's not like trying to dynamically add VOP's.. ;-). Actually, right now, I'm more worried about 2.2.5 going the way of 4.9.6 (ie: no one upgrades to 3.0 jut like 8.1.1). A bit of a stabilization problem, if nothing else. I would say "put it in 3.0, since the upgrade is going to be an expensive one no matter how you slice it, and you might as well hide the cost there". Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers.