[pim] Re: Erik Kline's Discuss on draft-ietf-pim-zeroconf- mcast-addr-alloc-ps-07: (with DISCUSS)

>
>
> > Please provide an example of how multicast addresses with T=1 overlap
> with ff02::1:ffxx:xxxx addresses.
>
>
>
> Sure, I will base this example off of the example in RFC 4489 section 4.
>
>
>
> Let’s say that a device’s IPv6 link local address is
> *fe80::a12:34ff:fe56:7890*. It will use a Solicited-Node multicast
> address of *ff02::1:ff56:7890*. It uses RFC 4489 to calculate a
> link-scoped multicast prefix of *ff32:00ff:a12:34ff:fe56:7890::/96*. If
> it chooses *0xff567890* as its group ID (which is compliant with host
> allocation described in RFC 3307 section 4.3.2), then it will use the IPv6
> multicast address *ff32:00ff:a12:34ff:fe56:7890:ff56:7890*. Both of these
> IPv6 multicast addresses use Ethernet multicast address
> *33:33:ff:56:78:90*.
>
>
>
> If you think this example would be useful to include in the document, how
> would you feel about updating
> draft-ietf-pim-updt-ipv6-dyn-mcast-addr-grp-id
> <https://datatracker.ietf.org/doc/draft-ietf-pim-updt-ipv6-dyn-mcast-addr-grp-id/>
> to include it? That document is focused specifically on improving the IPv6
> multicast address architecture to better accommodate multicast address
> allocation protocols.
>

Thank you; it's overall more clear that the "collision" that is concerned
here is (effectively) the resulting MAC address.

I think in this case what triggered me was "this dynamic range overlaps",
which confused me as the IPv6 prefixes do not overlap.  I think if you just
clarify that the group ID/MAC address is the thing that overlaps (or
collides) it would be helpfully more accurate.

Thanks for bearing with me,
-Erik

_______________________________________________
pim mailing list -- [email protected]
To unsubscribe send an email to [email protected]