Re: AW: I-D Action:draft-gundavelli-v6ops-l2-unicast-00.txt

[Sri Gundavelli <sgundave@cisco.com>]

Thanks to Stig for his review of the draft. Will reword the below text,
should be in -01 version.

Sri




------ Forwarded Message
From: Stig Venaas <stig@venaas.com>
Date: Mon, 15 Feb 2010 13:03:29 -0800
To: Sri Gundavelli <sgundave@cisco.com>
Subject: Re: L2 Unicast of multicast messages - ID

>> I think the document is fine. I've followed some of the previous
>> discussion on this topic. Just one comment.
>>
>> In section 3 it says:
>>
>>        address in the link-layer header will be an unicast address.  It
>>        is up to to the system architecture as when to transmit a IPv6
>>        multicast message as an link-layer unicast message, as long as
>>        there is no real impact to the multicast communication.
>>
>> This sentence is pretty vague. Especially "no real impact", not sure
>> what that means. And what do you mean by "multicast communication".
>>
>> Also, the reason you may want the system architecture to transmit
>> unicast, is that it has a positive impact on the communication, right?
>> If whether you use unicast or not has no impact, then this would be
>> pointless ;)
>>
> 
> :) My point is that, if the usage of the semantic is used in the right away,
> there should be any impact to the multicast communication. If I'm hosting
> two IPv6 VLAN's on the same 802.11 link, if the AP ensures the RA's are
> segregated and sent to the right groups, its to me no impact to multicast
> communication.  But, I see your point, this needs to worded correctly. I
> will fix this in -01 version, posted it a hour back.

Sorry I was a bit late. I think the draft is good. Just trying to be a
bit difficult here :) But I think it can be made clearer.

Stig