Pekka Savola wrote:

     Inbound load balancing - Nodes with replicated interfaces may want
           to load balance the reception of incoming packets across
           multiple network interfaces on the same link.  Such nodes
           have multiple link-layer addresses assigned to the same
           interface.  For example, a single network driver could
           represent multiple network interface cards as a single
           logical interface having multiple link-layer addresses.

           Load balancing is handled by allowing routers to omit the
           source link-layer address from Router Advertisement packets,
           thereby forcing neighbors to use Neighbor Solicitation
           messages to learn link-layer addresses of routers.  Returned
           Neighbor Advertisement messages can then contain link-layer
           addresses that differ depending on who issued the
           solicitation.

Maybe just add a sentence at the start of the second paragraph:

Hosts cannot perform inbound load balancing.

I don't think that is true. What is true is that 2461 has no explicit support for hosts to do inbound load balancing, and the ability for routers to omit the SLLA in the RAs is explicit support for routers doing so.


The example of how a host could do this:
When sending a RS use a SLLAO which is a random function.
When sending a NS use the peer's (target) address in some hash to determine which SLLAO to use.
When responding to a NS, use a TSLLA based on the same hash.
When sending an unsolicited NS, pick a random one of the link layer addresses the host has (just as for RS).


Clearly this isn't perfect, since LLAO in the RS and unsolicited NA
might later be replaced by another one. But it shows it isn't impossible.

So I'd suggest adding a sentence at the start of the second paragraph:
        This specification has no explicit support for hosts to perform
        inbound load balancing.
and rewording the rest to start with
        Routers can perform inbound load balancing by omitting ...

   Erik

--------------------------------------------------------------------
IETF IPv6 working group mailing list
ipv6@ietf.org
Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------

Reply via email to