Re: [Softwires] [Softwire] draft-ietf-softwire-map-00 does NOT reflect the consensus from the WG

[Peng Wu <[e-mail]>]

Satoru,

On Mon, Jun 25, 2012 at 5:05 PM, Satoru Matsushima
<[e-mail]> wrote:
> Hi Peng,
>
> On 2012/06/25, at 17:50, Peng Wu wrote:
>
>>> Hmm, I've read 'draft-cui-softwire-b4-translated-ds-lite' as you called 'lightweight 4over6'. LW46 for short, it looks me that MAP just provides LW46 a provisioning means which would be described in the section 5, or appendix section because following text described in section 5:
>>>
>>> "Other optional alternatives to retrieve the public address and port-
>>>  set also exist.  The specific protocol extensions are out of scope in
>>>  this document, however some alternatives are mentioned in the Appendix
>>>  section."
>> Let me take a step further following your direction:
>> You are saying MAP is a provisioning mean in lw4over6. Then, what exactly?
>> With IPv4-IPv6 addressing independency, you proivde an IPv4 address
>> and port set, within another DHCPv6 option?
>> No BMR, No FMR, No embedding address. Look at what's left. how is this
>> even similar to MAP?
>
> Let's think that a CE provisioned with following BMR comes from MAP DHCPv6 options.
>
> BMR:
>  o Rule-ipv6-prefix  : {exact matched with CE's delegated prefix}
>  o Rule-ipv4-prefix  : x.x.x.x/32
>  o EA-length         : 0
>  o Port-param option : {PSID/length}
>
> This BMR could be a LW46 provisioning means.

Again, all the information needed is the IPv4 address and port set.

1) The item like rule-ipv6-prefix is not needed at all.
2) Port set or PSID still needs extra provisioning (while in regular
MAP it's embedded in IPv6 address)

So why make it so difficult and obscure