Re: problem statement [was Re: New Version Notification for draft-hinden-ipv4flag-00.txt]

[Jen Linkova <[e-mail]>]

On Mon, Nov 20, 2017 at 1:25 AM, Nick Hilliard <[e-mail]> wrote:
> One relatively straightforward way of dealing with extraneous dhcpv4
> packets would be to create a new dhcpv4 reply option hinting to the
> requester to cease DHCPv4 requests on the interface in question, or to
> slow down the request rate from one every two minutes to one every X
> minutes where X is network defined.

The idea of using a protocol X to tell device 'there is no protocol X
on this link' sounds very entertaining ;)
If I run a single-stack network I do not want to configure IPv4
addresses, dhcp relays etc on v6-only interfaces
and most likely I have no desire to run DHCPv4 for those segments. I
might not even allow v4 traffic on the link at all
(as it reduces the attack surface).

There are two unrelated things which could be done:
A) To protect the network from excessive amount of undesirable (v4)
traffic some actions need to be taken on the network level. Those
actions
could not be completely delegated to hosts as most likely we are
talking about large number of devices, uncontrolled, potentially
outdated etc.
However the network indeed could provide some hints to hosts in case
those hosts are willing to listen. Here comes B)
B) To improve the device performance at v6-only network, the device
can do smth to detect the lack of IPv4 on the link (either explicitly,
like the RA flag we are discussing) or implicitly (no DHCPv4 replies
-> back-off etc).


-- 
SY, Jen Linkova aka Furry