On 5/17/2016 8:54 AM, Joe Touch wrote: > >> On May 17, 2016, at 8:31 AM, Juliusz Chroboczek >> <j...@pps.univ-paris-diderot.fr> wrote: >> >> Hi, >> >> I just received a bug report for shncpd from somebody who noted that >> shncpd's DHCPv4 server will happily assign addresses a.b.c.0 and a.b.c.255 >> to clients. That's obviously broken. > Not since 1993 and RFC 1519, depending on the subnet mask. > > See also RFC 3021. > >> Which addresses must be avoided, for both v4 and v6? > All 0's and all 1's within a subnet range except for /31 in IPv4. > > I don't think there are prohibited or reserved interface or group IDs in > IPv6.
I stand corrected: On 5/17/2016 9:01 AM, Bless, Roland (TM) wrote: > https://tools.ietf.org/html/rfc5453 > provides an answer w.r.t. IPv6. > > https://tools.ietf.org/html/rfc7136 > is probably also of interest. > > Regards, > Roland That's a useful pointer, but those two docs aren't the whole story apparently. The full registry, as created by RFC5354, is here: http://www.iana.org/assignments/ipv6-interface-ids/ipv6-interface-ids.xhtml#ipv6-interface-ids-1 It's more than I had originally expected and includes all 0s, but not all 1's. Joe > > Specific addresses and ranges for both, but I think that's a different issue. > > Joe > >> I'm already setting >> the g and u bits in modified EUI-64 to 0 (silly as it is), what else >> should I take care of? >> >> -- Juliusz >> >> _______________________________________________ >> homenet mailing list >> homenet@ietf.org >> https://www.ietf.org/mailman/listinfo/homenet _______________________________________________ homenet mailing list homenet@ietf.org https://www.ietf.org/mailman/listinfo/homenet
