|
Comments below. > Date: Tue, 4 Nov 2008 18:11:43 -0500 > From: suresh.krishnan@ericsson.com > To: Bernard_Aboba@hotmail.com > CC: dna@ietf.org > Subject: Re: [dna] #9: Detailed description of ND packets > > Hi Bernard, > > Bernard Aboba wrote: > > There is an issue in Section 4.5.1. This section should also describe the > > link layer addresses in the NS packet, to make clear that the NS is > > sent to the previously cached link layer address of the router (NOT to > > a multicast address). > > Will do. > > > > > " The probing node SHOULD NOT include a Source link-layer address > > option if it has not performed duplicate address detection [RFC4862], > > for the source address of the NS, on the newly attached link." > > > > In DNAv4 (RFC 4436), the goal was not only to detect network attachment, but > > also to establish bi-directional reachability. By not including the Source > > link-layer address, Simple DNA does not achieve this goal. > > I think this comes back to the issue on when DAD is required. > I do understand your point, but I still see an issue with putting the SLLAO > in the packet when we are not reasonably sure that the address is not > duplicated. From your mail, I see that you consider the chance of > the address being taken by some other node as very low. I do not know > how to quantify this in text. If I can quantify when DAD is not > required, I think it will be very straight forward to address your > concern. Do you have any suggestions? A potential suggestion: The probing node SHOULD include a Source link-layer address option if either: a. the probing node has performed duplicate address detection [RFC4862] after attachment; a. the probing node has previously performed duplicate address detection [RFC4862] for the source address included within the NS, and the conditions described in Section X are satisfied, so that the results of the prior DAD are still considered valid. |
_______________________________________________ dna mailing list dna@ietf.org https://www.ietf.org/mailman/listinfo/dna