Re: Issue with some unnamed ISP's in northern north america, RFC 1918 Violation

[Matthew Petach via NANOG <nanog () lists nanog org>]

That's not at all uncommon.
ATT fiber works that way as well.  The static IP block comes out on the
same L2 port as the dynamic NAT pool.
Devices which ask for an address via DHCP get NAT'd v4 space, devices which
are configured manually use the static IP address and gateway.  Works just
fine.
Doesn't seem like much of a violation to me.

Perhaps you could clarify why you think it's a violation of the RFC to have
both a dynamic DHCP block and a static block existing on the same layer 2
segment?

Thanks!

Matt



On Tue, Mar 10, 2026 at 2:19 PM wolf1098--- via NANOG <nanog () lists nanog org>
wrote:

> One of the fibre providers, they are issuing static ips, which is good,
> but the way the statics are handled. The modem/router, if you want static
> ips, requires it to be in router mode, you can't bridge. it seems the
> router's static ip gateway does some bgp towards the isp to route the
> static ip's block assigned to the customer to it, via its dhcp established
> route??? idk. ------ THE big issue, which is, since it has to be in router
> mode. they end up with a dynamic ip, that is the gateway for dhcp RFC1918
> addresses, that are on its lan, sounds reasonable, but as a tertiary use,
> the static ip WAN gw stack, LOADS IN ON the same L2 lan bridge, side by
> side with the RFC 1918 space, without even attempting to put it in a tagged
> vlan. don't do this.... without using a custom linux router, good luck
> using both the static ips block, and the double 'natted' dynamic ip.
> _______________________________________________
> NANOG mailing list
>
> https://lists.nanog.org/archives/list/nanog () lists nanog org/message/CWEOU5LH5DZ4OISZZ6DC4OZELSZXNZN4/
_______________________________________________
NANOG mailing list 
https://lists.nanog.org/archives/list/nanog () lists nanog org/message/UXOSQPTNY3IAVKENG4V7BPZPAY7GOUKM/