Re: [ipwave] WGLC for draft-ietf-ipwave-ipv6-over-80211ocb-08 - multicast request

["Sri Gundavelli (sgundave)" <[e-mail]>]

I will check draft-perkins to see what I am missing. But given the nature
of this 802-11-OCB-link, I do not understand why all OBU/RSU nodes in
vicinity will not receive a packet sent to ALL_NODES_MULTIGROUP.

I realize, you are dealing with too many questions and so we can discuss
this after you resolve your other issues. But, I do not believe we need to
register a new group, but we can discuss that later.


Sri



On 10/6/17, 9:56 AM, "Alexandre Petrescu" <[e-mail]>
wrote:

>
>
>Le 06/10/2017 à 18:50, Sri Gundavelli (sgundave) a écrit :
>> Hi Alex,
>> 
>> I remember asking one question on why we need this?  Sorry, I may have
>> missed the response.
>
>BEcause it is one possible way of implementing IP multicast.  IP
>multicast is absolutely necessary for IP.
>
>> If there is already ALL_NODES_MULTICAST GROUP (FF02::1), which all OBY¹s
>> and RSU¹s participate, why do we need a new group?
>
>Because the subnet structure in WiFi is different than on Ethernet.
>There is a draft explaining that I-D.perkins-intarea-multicast-ieee802.
>
>Further, because the subnet structure in OCB mode is different than the
>subnet structure in WiFi.
>
>In an ALL_NODES group ff02::1 one is sure that whoever subscribed to
>that group will receive whatever message is sent to that group.  By
>extension, an ALL_NODES group in WiFi is relatively similar, although
>even there there are adifferences.
>
>In OCB mode that is not assured - there is no link-layer mechanism to
>make sure that whoever joined that subnet is still there.
>
>Can I ask you in return: why does the Babel routing protocol have its
>own multicast group(s)?
>
>Alex
>
>> 
>> 
>> 
>>> A Group ID named TBD, of length 112bits is requested to IANA; this
>>>Group
>>> ID signifies ³All 80211OCB Interfaces Address".
>> 
>> 
>> 
>> 
>> 
>> 
>> Regards
>> Sri
>> 
>> 
>> 
>> 
>> On 10/6/17, 9:40 AM, "its on behalf of Alexandre Petrescu"
>> <[e-mail] on behalf of [e-mail]> wrote:
>> 
>>>
>>>
>>> Le 02/10/2017 à 18:50, Margaret Cullen a écrit :
>>> [...]
>>>
>>>> In Section 7:  I think you are attempting to allocate a ³Link-Local
>>>> Scope Multicast Address² from the ³IPv6 Multicast Address Space
>>>> Registry², which you can find here:
>>>>
>>>> 
>>>>https://www.iana.org/assignments/ipv6-multicast-addresses/ipv6-multicas
>>>>t-
>>>> addresses.xhtml#link-local.
>>>> If so, you should say that :-).  Even if I am wrong about what you
>>>> are trying to allocate, though, you need to list a specific IANA
>>>> registry in this section, and be clearer about exactly what you are
>>>> asking them to allocate.
>>>
>>> We are asking a Group ID, see the IANA Considerations section.
>>>
>>> This Group ID can then be used to form multicast addresses of various
>>> scopes, including the Link-Local scope.
>>>
>>> I prefer to be silent at this time about which scope will be used,
>>> unless you disagree.
>>>
>>> Alex
>>>
>>> _______________________________________________
>>> its mailing list
>>> [e-mail]
>>> https://www.ietf.org/mailman/listinfo/its
>> 
>>