Answering Tassos' question specifically, if the use of the flow label for ND/DAD requires a signalling phase to tell the other end the flow label value in use, it is clearly out of scope for 3697bis. (Not forbidden, just not described.)
However, I can't imagine how you would signal anything before performing neighbour discovery. I thnk it's just orthogonal to 3697bis. The usage of the flow label in link-local scope before completing address assignment doesn't seem to me to be constrained by 2460, 3697 or 3697bis. However, I agree that 20 bits is small for a nonce. It puts you into birthday-paradox territory on a LAN with a few hundred nodes. Brian On 2011-10-16 01:55, Hemant Singh (shemant) wrote: > I certainly don’t believe so. The usual use of Flow Label seems to assume > the labeled packet traverses more than one hop. After all, the typical use > of the Flow Label is to let each hop know what processing to do based on the > Flow Label. However, a link-local domain whose traffic is restricted to > within one link with no traversal across a router/hop, the Flow Label does > not have a hop-by-hop use but certainly if some information needs to be > provided to all nodes the flow label can be used. But such information is > best provided by the control protocol of ND used in the link. So why > overload the Flow Label. Anyway, some folks may not like the small size of > 20 bits to be used in a nonce . > > > > Hemant > > > > From: Tassos Chatzithomaoglou [mailto:[email protected]] > Sent: Saturday, October 15, 2011 7:50 AM > To: Brian E Carpenter > Cc: Hemant Singh (shemant); IPv6 WG Mailing List > Subject: Re: FW: New Version Notification for > draft-hsingh-6man-enhanced-dad-01.txt > > > > Also, looking at draft-ietf-6man-flow-3697bis-07, i can see the following > under introduction: > > A stateful > scenario is one where a node that processes the flow label value > needs to store information about the flow, including the flow label > value. > > ... > > The flow label can be used most simply in stateless scenarios. This > specification concentrates on the stateless model and how it can be > used as a default mechanism. Details of stateful models, signaling, > specific flow state establishment methods and their related service > models are out of scope for this specification. > > > Does this mean that the usage of flow label on ND/DAD is out of scope of > 3697bis? > > -- > Tassos > > > > > -------------------------------------------------------------------- IETF IPv6 working group mailing list [email protected] Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6 --------------------------------------------------------------------
