Re: [dhcwg] draft-ietf-dhc-dhcpv6-opt-netboot-05 - how to proceed?

"H. Peter Anvin" <hpa@zytor.com> Fri, 06 November 2009 02:39 UTC

Return-Path: <hpa@zytor.com>
X-Original-To: dhcwg@core3.amsl.com
Delivered-To: dhcwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 41A203A6867 for <dhcwg@core3.amsl.com>; Thu, 5 Nov 2009 18:39:48 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.599
X-Spam-Level:
X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id LQwZHg1tMEn6 for <dhcwg@core3.amsl.com>; Thu, 5 Nov 2009 18:39:47 -0800 (PST)
Received: from terminus.zytor.com (terminus.zytor.com [198.137.202.10]) by core3.amsl.com (Postfix) with ESMTP id 590363A6A2E for <dhcwg@ietf.org>; Thu, 5 Nov 2009 18:39:27 -0800 (PST)
Received: from anacreon.sc.intel.com (hpa@localhost [127.0.0.1]) (authenticated bits=0) by terminus.zytor.com (8.14.3/8.14.3) with ESMTP id nA62aXnv000503 (version=TLSv1/SSLv3 cipher=DHE-RSA-CAMELLIA256-SHA bits=256 verify=NO); Thu, 5 Nov 2009 18:36:34 -0800
Message-ID: <4AF38BB1.5060908@zytor.com>
Date: Thu, 05 Nov 2009 18:36:33 -0800
From: "H. Peter Anvin" <hpa@zytor.com>
User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.4pre) Gecko/20091014 Fedora/3.0-2.8.b4.fc11 Thunderbird/3.0b4
MIME-Version: 1.0
To: Ted Lemon <Ted.Lemon@nominum.com>
References: <200910062234.AAA07005@TR-Sys.de><AD61797F-DB2D-4889-8888-5339521ADE8F@nominum.com><OF029AB4A7.6EEA954A-ONC125764B.004E95CE-C125764B.00536F30@de.ibm.com><FBD14811-7C9A-4FCB-8707-071A7CA12B96@nominum.com><OF73909874.9B93FADE-ONC1257653.002E96C8-C1257653.00302A9F@de.ibm.com><589F59F0-3E17-44FF-8918-FED51F03EE4D@nominum.com><4ADC156D.1070709@ucd.ie><669B95F2-FA8F-47AA-AB8A-4292DDB1A4EF@nominum.com><4ADD70CB.2080409@ucd.ie><OF2692FAA8.2F4AF98C-ONC1257657.002954C1-C1257657.00298474@de.ibm.com> <200910221814.n9MIETQR002070@cichlid.raleigh.ibm.com> <8A82D1BFEDDE7E4597978355239BBBCB4C204C@PACDCEXCMB06.cable.comcast.com> <200910221921.n9MJLJgW010502@cichlid.raleigh.ibm.com> <8A82D1BFEDDE7E4597978355239BBBCB4C2063@PACDCEXCMB06.cable.comcast.com> <200910221941.n9MJfSV7019188@cichlid.raleigh.ibm.com> <82D0F823-61A5-45F9-A87E-21720DF221F9@nominum.com>
In-Reply-To: <82D0F823-61A5-45F9-A87E-21720DF221F9@nominum.com>
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: 7bit
Cc: Thomas Narten <narten@us.ibm.com>, "dhcwg@ietf.org" <dhcwg@ietf.org>, "Woundy, Richard" <Richard_Woundy@cable.comcast.com>
Subject: Re: [dhcwg] draft-ietf-dhc-dhcpv6-opt-netboot-05 - how to proceed?
X-BeenThere: dhcwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: <dhcwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/dhcwg>, <mailto:dhcwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/dhcwg>
List-Post: <mailto:dhcwg@ietf.org>
List-Help: <mailto:dhcwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/dhcwg>, <mailto:dhcwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 06 Nov 2009 02:39:48 -0000

On 10/23/2009 09:08 AM, Ted Lemon wrote:
> On Oct 22, 2009, at 12:41 PM, Thomas Narten wrote:
>> Well, I wasn't all that clear in my first note either! I'm really
>> hoping for the same. I'd be happy to support load balancing, but only
>> if was a real requirement coming from the folk who are using netboot
>> and have experience with it. Otherwise, I'd prefer that we keep things
>> simple.
> 
> I may be speaking out of turn here, but I think that the fact that this
> draft explicitly recommends http over tftp for downloading the boot file
> will solve a lot of the load problems that exist with the current
> DHCP/tftp boot mechanism in DHCPv4.   As I understand it, the primary
> problem with netbooting in IPv4 is that tftp is such a low-performance
> solution (due to the lock-step nature of the protocol), not that tftp
> boot servers get overloaded during power events.
> 

[Sorry for the late reply]

TFTP boot servers, because of the lockstep protocol and generally very
simple implementation of clients, is subject to a failure-capture effect
where a "losing" client ends up continuing to "lose" and thus end up
with total boot failures.  This is one of many reasons why TCP/HTTP is a
much better option.

I'm more than willing to list others ;)

	-hpa