DHCPv6 and MAC Address inclusion

[Ted Lemon]

On Jan 23, 2012, at 10:34 PM, "scott_stone at trendmicro.com<mailto:scott_stone at trendmicro.com>" <scott_stone at trendmicro.com<mailto:scott_stone at trendmicro.com>> wrote:
I've done some reading on this - the intent here was for laptops/etc that have wired and wireless cards, to get the same IP regardless.  Makes sense if you think of it that way but it destroys my datacenter use-case.  DHCPv6 was likely designed 15+ years ago like IPv6 was and by the time anyone got around to actually using IPv6 en masse (ie, 2011), it was long obsolete and hadn't had the updates/attention that IPv4/DHCPv4 got over the same time period.

No, that's really not the intent. If that were the intent, the DHCP client would not send an IAID. I'm one of the authors of the spec, FWIW, and it's not fifteen years old–I haven't been going to the IETF that long.   I'm interested in constructive comments, but you're just making stuff up here.

 Still scares me that there is no "private nonroutable" block of IPv6 space.  Someone screws up an ACL somewhere and bam, private/firewalled machines (DB hosts, etc) are now exposed to the internet.

Again, you might want to do a bit more research before making statements like these. There are in fact address ranges that do precisely what you want here.   They are called unique local addresses, and the advantage they have is that a routing mistake doesn't break your walled garden.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.isc.org/pipermail/dhcp-users/attachments/20120124/8bf549c8/attachment.html>