Re: [v6ops] New Version Notification for draft-hilliard-v6ops-host-addr-update-00.txt

[Tore Anderson <[e-mail]>]

* Gert Doering <[e-mail]>

> On Thu, Jul 20, 2017 at 10:21:03AM +0200, Lorenzo Colitti wrote:
> > That's not true at all. There are (tens of) millions of home networks that
> > do both IA_NA and SLAAC.  
> 
> That's an interesting statement.  Which products do it that way today,
> out of the box?

LEDE/OpenWrt/HomeWrt at least.

From what I've heard the DHCPv6 stuff is there mostly to support
automatic host name discovery, but it also has the effect of
facilitating prefixes longer than /64. At least HomeWrt will do that if
there are too few /64s to number all the links in the Homenet. There
are Norwegian ISPs that hand out /62s and /60s to their subscribers, so
this can be a real concern.

That said, for me having the additional DHCPv6-assigned address in
addition to the SLAAC one has been a net negative, since it doesn't
reconfigure along with the link prefix following a PD change (but
nevertheless tends to be preferred for outbound traffic). Maybe this
would have been better if it was a ULA rather than a GUA though, I
don't know.

When it comes to the enterprise data centre networks I operate, on the
other hand, I do absolutely need DHCPv6 for network booting. However I
do not really need it once the system has booted, SLAAC serves my needs
fine then. So I'm actually considering setting up my DHCPv6 server so
that (by default) only the PXE client arch types will get addresses
assigned through DHCPv6.

Tore