From owner-freebsd-hackers  Sun Jul  6 16:51:54 1997
Return-Path: <owner-hackers>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.5/8.8.5) id QAA12096
          for hackers-outgoing; Sun, 6 Jul 1997 16:51:54 -0700 (PDT)
Received: from whistle.com (s205m131.whistle.com [207.76.205.131])
          by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA12087;
          Sun, 6 Jul 1997 16:51:43 -0700 (PDT)
Received: (from smap@localhost) by whistle.com (8.7.5/8.6.12) id QAA18549; Sun, 6 Jul 1997 16:51:13 -0700 (PDT)
Received: from bubba.whistle.com(207.76.205.7) by whistle.com via smap (V1.3)
	id sma018547; Sun Jul  6 16:51:01 1997
Received: (from archie@localhost) by bubba.whistle.com (8.8.5/8.6.12) id QAA13723; Sun, 6 Jul 1997 16:51:01 -0700 (PDT)
From: Archie Cobbs <archie@whistle.com>
Message-Id: <199707062351.QAA13723@bubba.whistle.com>
Subject: Re: kern/3446
In-Reply-To: <199707062239.PAA26655@quack.kfu.com> from "nsayer@quack.kfu.com" at "Jul 6, 97 03:39:44 pm"
To: nsayer@quack.kfu.com
Date: Sun, 6 Jul 1997 16:51:01 -0700 (PDT)
Cc: fenner@FreeBSD.ORG, joerg@FreeBSD.ORG, jkh@FreeBSD.ORG,
        hackers@FreeBSD.ORG
X-Mailer: ELM [version 2.4ME+ PL31 (25)]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-hackers@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk


> > Synopsis: IPFIREWALL reject returns port unreachable, not host
> 
> > State-Changed-From-To: open-closed
> > State-Changed-By: fenner
> > State-Changed-When: Sun Jul 6 12:42:34 PDT 1997
> > State-Changed-Why: 
> > Turns out this is yet another duplicate, for kern/3452.
> > I missed that one because it's closed.
> 
> I don't know how so many duplicates got made. I believe I sent this
> in a total of twice.
> 
> I must protest in the strongest possible terms the closure without
> action of this PR.
> 
> The language given in the closure of 3452 suggests that the PR
> should be dismissed because FreeBSD is acting correctly
> according to the RFCs. That is not the issue here. The issue
> here is that behavior that is correct according to the RFC
> breaks what is perhaps the most populous unix implementation
> that the world has ever known. I feel that that is worth at
> _least_ of a sysctl variable (as exists for TCP extensions,
> for exmaple), if not an outright substitution of behavior that
> actually works for behavior that is theoretically correct.

This bug is made obsolete by the new ipfw changes, which allow you
to specify as part of the rule itself what type of ICMP error code
is returned... so you can now choose either way :-)

-Archie

___________________________________________________________________________
Archie Cobbs   *   Whistle Communications, Inc.  *   http://www.whistle.com