Re: [dhcwg] draft-jiang-dhc-stateless-reconfiguration-01

Sheng Jiang <jiangsheng@huawei.com> Mon, 17 February 2014 05:43 UTC

From: Sheng Jiang <jiangsheng@huawei.com>
To: Yusuke DOI <yusuke.doi@toshiba.co.jp>, "<dhcwg@ietf.org>" <dhcwg@ietf.org>
Thread-Topic: [dhcwg] draft-jiang-dhc-stateless-reconfiguration-01
Thread-Index: AQHPK4ZD1LIqFKk32E6YYvetgI1Xj5q4tc1Q
Date: Mon, 17 Feb 2014 05:42:35 +0000
Message-ID: <5D36713D8A4E7348A7E10DF7437A4B923ADE4D33@nkgeml512-mbx.china.huawei.com>
References: <530170E0.3040004@toshiba.co.jp>
In-Reply-To: <530170E0.3040004@toshiba.co.jp>
Accept-Language: en-GB, zh-CN, en-US
Content-Language: zh-CN
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
MIME-Version: 1.0
Archived-At: http://mailarchive.ietf.org/arch/msg/dhcwg/KZEUhVvOY_btYmUn1R8zCxqILIg
Subject: Re: [dhcwg] draft-jiang-dhc-stateless-reconfiguration-01
Precedence: list

Hi, Yusuke,

Thanks for your reply and support. Some explanation below.

>>    {Question to WG No.2} There could be two kind of stateless DHCPv6
>>    reconfiguration modes as the following, which one is proper?  Or we
>>    should support both?
>
>Is there any difficulty to support both modes? I think these two modes should
>have different use cases. 'Push mode' will fit to our use case because the
>configuration is network-wide and push-mode (with multicast target -- it can
>be broadcasted on a link) is much much more efficient than trigger mode in
>LLN. But some administrator may want to have per-client configuration and
>that is the nature of DHCP, as described.

The difficulty is no technical aspect. However, there are two considerations for push mode:

a) the push mode is different from traditional DHC behavior in that clients is the initiator. So, this would need the agreement of DHC WG However, even the existing reconfiguration in stateful DHCPv6 is initiated by server side. So, this may not be too controversial. It is still a question requiring confirmation for now.

b) "push mode' may have more security issues that trigger mode. How to prevent a fake DHCPv6 server push malicious configuration is a difficulty, although DHCPv6 snooping may help. "Trigger mode" is safer for that client would contact original server for update configuration.

Regards,

Sheng

>Regards,
>
>// Yusuke DOI <yusuke.doi@toshiba.co.jp>
>
>_______________________________________________
>dhcwg mailing list
>dhcwg@ietf.org
>https://www.ietf.org/mailman/listinfo/dhcwg

[dhcwg] draft-jiang-dhc-stateless-reconfiguration… Yusuke DOI
Re: [dhcwg] draft-jiang-dhc-stateless-reconfigura… Sheng Jiang