From owner-freebsd-hackers Mon Jul 13 07:42:49 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id HAA09932 for freebsd-hackers-outgoing; Mon, 13 Jul 1998 07:42:49 -0700 (PDT) (envelope-from owner-freebsd-hackers@FreeBSD.ORG) Received: from pau-amma.whistle.com (s205m64.whistle.com [207.76.205.64]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id HAA09926 for ; Mon, 13 Jul 1998 07:42:47 -0700 (PDT) (envelope-from dhw@whistle.com) Received: (from dhw@localhost) by pau-amma.whistle.com (8.8.8/8.8.7) id HAA08166 for hackers@freebsd.org; Mon, 13 Jul 1998 07:42:07 -0700 (PDT) (envelope-from dhw) Date: Mon, 13 Jul 1998 07:42:07 -0700 (PDT) From: David Wolfskill Message-Id: <199807131442.HAA08166@pau-amma.whistle.com> To: hackers@FreeBSD.ORG Subject: Re: Changes to login.conf In-Reply-To: <199807130639.XAA00272@usr07.primenet.com> Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG >From: Terry Lambert >Date: Mon, 13 Jul 1998 06:39:40 +0000 (GMT) >Well, this is more along the lines of a schema based system, where you >say "here is the proposed changes, are they consistent?". >It's more useful that they be globally consistent than that they be >merely internally consistent. Quite. >This moves into the real of LDAP and subschema enforcement, I think. That's a possible implementation, sure. In case it isn't Real Obvious(!) yet, I try to avoid jumping to implementation concerns too quickly.... :-} >You could also consider the case of a modification to the bind >configuration information that would result in both upper and lower >case versions of the same host name. Yes, though I'd class the nameserver "reality checking" as rather distinct from /etc/rc*-checking (though the rc-checking probably ought to do a reality check with the nameservers the host in question uses). >There should be a method for proposing a group of changes as an atomic >unit, and having them vetted (or rejected) as a unit. Yes -- something that is capable of eating the global config stuff without getting too much indigestion. :-} It could start fairly limited, and if it were well-designed, could expand -- direction of expansion being toward areas where more analgesics are required (typically, from self-inflicted injuries).... And if it weren't well-designed, well, then, it would either stagnate (& be forgotten), or change: if for the better, great; if not, it could die. And since it need not be an integral part of anything, its fate would have minimal impact on anyone or -thing else.... unless, of course, it became so useful for enough folks that they kept working on it.... :-} david -- David Wolfskill UNIX System Administrator dhw@whistle.com voice: (650) 577-7158 pager: (650) 371-4621 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message