Re: Confirmation to advance: draft-ietf-6man-ipv6only-flag-05

[Nick Hilliard <[e-mail]>]

David Farmer wrote on 23/05/2019 23:13:
> What he is saying is a dual-stack host that has functioning IPv6
> should not enable IPv4 LL. And functioning IPv6 means at a minimum it
> has learned an IPv6 GUA and probably a default router as well, you
> could also test for connectivity beyond the default router, but that
> 
> The risk this creates is that on a dual-stack network if a there is
> an issue with the DHCPv4 server, such as it crashed or was
> misconfigured then the dual-stack host cannot fall back to IPv4 LL
> and will not be able to perform Link-Local communications with
> IPv4-Only devices on the common link between them.
>
> Personally, I think this is an acceptable risk, but I'm not convinced
> that there is a consensus that this is an acceptable risk. Therefore,
> I'm not ok with abandoning the IPv6-Only flag until there is a
> consensus that this is an acceptable risk.. Furthermore, we need more
> than 6man's consensus on this risk.

the usual ietf position is that operational misconfig - including
daemons crashing - is out of scope for protocol development.  If it's 
out of scope, there's no real issue regarding consensus.

Obviously, from an operational point of view, if DHCP service is 
important on a network, the operator is at liberty to install redundant 
servers / relays.

Nick