[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Stateless DHCP options "authorized" in RAs, in a standard-format container
- To: "STARK, BARBARA H (ATTLABS)" <bs7652@att.com>
- Subject: Stateless DHCP options "authorized" in RAs, in a standard-format container
- From: Rémi Després <remi.despres@free.fr>
- Date: Fri, 19 Mar 2010 15:47:29 +0100
- Cc: "Syam Madanapalli" <smadanapalli@gmail.com>, "Fred Baker" <fred@cisco.com>, "IPv6 Operations" <v6ops@ops.ietf.org>, "Lindqvist Kurt Erik" <kurtis@kurtis.pp.se>, "Ralph Droms" <rdroms@cisco.com>, "6man Chairs" <6man-chairs@tools.ietf.org>, "Dave Thaler" <dthaler@wollive.windowsmedia.com.akadns.net>, "Jari Arkko" <jari.arkko@piuha.net>, <jjeong@cs.umn.edu>, <luc.beloeil@orange-ftgroup.com>, "Daniel Park" <soohong.park@samsung.com>, "Suresh Krishnan" <suresh.krishnan@ericsson.com>, "Dan Wing" <dwing@cisco.com>
- In-reply-to: <750BF7861EBBE048B3E648B4BB6E8F4F12257DF6@crexc50p>
- References: <4B9DFC7D.3070704@piuha.net> <4B9E96E2.10108@piuha.net> <1315FBDA-12A2-4C16-B66F-CBD4802E6766@cisco.com> <4BA089F9.5010006@piuha.net> <65B6B54D-98AD-4772-B2E0-6E2CA8DE76C0@cisco.com> <419DB14D-BFDC-4118-BB3E-F4D9570D927D@kurtis.pp.se> <A5AB4B97-11BA-4D27-860C-3811D075BFC6@cisco.com> <4BA0EC66.3010403@piuha.net> <308C7176-40DF-4F82-AC2D-A4EAC6E2766B@cisco.com> <818AC122-7E68-45F7-A167-672F7DE47207@free.fr> <10e14db21003190405p323e9ddaoab4aab650971ae3e@mail.gmail.com> <FE0A52AB-91C7-4E88-B055-F95A439B40AE@free.fr> <750BF7861EBBE048B3E648B4BB6E8F4F12257DF6@crexc50p>
Le 19 mars 2010 à 15:21, STARK, BARBARA H (ATTLABS) a écrit :
>> The standard needs only to say that:
>> - Routers MAY send RAs containing stateless DNS options
>> - Hosts MAY be ready to receive stateless DNS options in RAs
>> - IF THEY DO, it MUST be with the approved format.
>> (And, for this, the format proposed by Suresh is so obvious and simple
>> that I doubt there would be any competitive proposals)
>>
>> This being done, one can expect that all vendors of products that send
>> and/or receive RAs will eventually add this simple capability to their
>> products, making things more efficient and simpler in many
>> circumstances.
>
> Thus doubling the complexity of the devices that now have to support *both* DHCPv6 stateless and RAs for receiving/sending that *identical* info,
No.
Again, no one is obliged to implement it in any device.
But those who do it (to increase efficiency), have a standard format to do it.
Deployment is incremental and can remain partial.
The point is that authorizing some DHCP options in RAs is more powerful, and simpler, than RFC 5006.
This way should therefore, IMHO, be preferred.
RD