Re: [v6ops] I-D Action: draft-palet-v6ops-rfc6177-bis-00.txt

["Mudric, Dusan (Dusan)" <[e-mail]>]

Hi David,

>RFC7084;
>   WPD-2:  The IPv6 CE router MAY indicate as a hint to the delegating
>           router the size of the prefix it requires.  If so, it MUST
>           ask for a prefix large enough to assign one /64 for each of
>           its interfaces, rounded up to the nearest nibble, and SHOULD
>           be configurable to ask for more.
>Maybe we need "MUST be configurable to ask for more.", which is the IETF's fault!

I believe this the right approach. We cannot have IPv6 network where device and application vendors need first to go around the world and negotiate with all ISPs to provide /48, before they start building new products that require /48. The network MUST be flexible, by default.

Thanks,
Dusan.

From: David Farmer <[e-mail]>
Sent: Tuesday, August 28, 2018 8:04 PM
To: Fred Baker <[e-mail]>
Cc: Mudric, Dusan (Dusan) <[e-mail]>; [e-mail]; IPv6 Operations <[e-mail]>; JORDI PALET MARTINEZ <jordi.palet=[e-mail]>
Subject: Re: [v6ops] I-D Action: draft-palet-v6ops-rfc6177-bis-00.txt


On Tue, Aug 28, 2018 at 5:32 PM, Fred Baker <[e-mail]<mailto:[e-mail]>> wrote:


On Aug 28, 2018, at 8:06 AM, Mudric, Dusan (Dusan) <[e-mail]<mailto:[e-mail]>> wrote:
> There must a way to define how prefixes are assigned to always allow /48, if needed. If not in rfc6177-bis than in another RFC. We have already one limitation for SLAAC that makes sure every home gets at least one /64. We can add another one, in another RFC, to make sure /48 is given on a demand. That will insure future flexibility in any network and make the prefix assignment non-political process.

My understanding, no hats and potentially incorrect, is that any ISP I know of will respond to a DHCPv6 IA_PD with a prefix at least as large as requested, up to /48.

Fred, the statement you are making I believe is true for many or most ISP's providing IPv6 today, but not all. However, way too many ISPs are still not yet providing IPv6. Further, many of the early IPv6 capable CPE or terminal devices only ask for a /64 and can't be told otherwise. So when I look at the current state of things, I'm both optimistic and pessimistic at the same time.
Dusan, it's not as simple as the ISP are not providing /48s, as Fred says many, if not most, doing IPv6 are, a big part of the issue is too many CPEs only ask for a /64, and don't allow a different option. But from the perspective of a normal end-user they are being limited, and who is at fault is complicated.  There is plenty of blame to go around for everyone, the ISPs, the CPE manufacturers, the IETF, and even a little for the end-users themselves.

RFC7084;
   WPD-2:  The IPv6 CE router MAY indicate as a hint to the delegating
           router the size of the prefix it requires.  If so, it MUST
           ask for a prefix large enough to assign one /64 for each of
           its interfaces, rounded up to the nearest nibble, and SHOULD
           be configurable to ask for more.

I think at least part of the problem is a common one, too many inexpensive CPE do not implement optional features. Because both ISP's and end-users have economic incentives to use inexpensive CPE, this creates an incentive for CPE manufacturers to make inexpensive CPE do not implement optional features.  Around and around we go.

It is a complicated systemic failure mode we have going here.

Maybe we need "MUST be configurable to ask for more.", which is the IETF's fault!

Thanks.


--
===============================================
David Farmer               Email:[e-mail]<mailto:[e-mail]>
Networking & Telecommunication Services
Office of Information Technology
University of Minnesota
2218 University Ave SE        Phone: [phone]
Minneapolis, MN 55414-3029   Cell: 612-812-9952
===============================================