Re: I-D Action: draft-ietf-6man-ipv6-alt-mark-07.txt

[Brian E Carpenter <[e-mail]>]

Amplifying my comment now that I'm on a computer not a phone:

For example, https://www.ietf.org/archive/id/draft-li-native-short-addresses-01.html

Regards
   Brian
On 29-Jul-21 22:32, Brian Carpenter wrote:
> I think IPv6 with smaller addresses would be a better approach, and there have been various proposals that amount to that.
> 
> Regards,
>     Brian Carpenter
>     (via tiny screen & keyboard)
> 
> On Thu, 29 Jul 2021, 20:57 Mark Smith, <[e-mail] <mailto:[e-mail]>> wrote:
> 
>     Hi Stewart,
> 
>     On Sat, 24 Jul 2021 at 19:39, Stewart Bryant <[e-mail] <mailto:[e-mail]>> wrote:
>     >
>     > All of which is why MPLS or some evolution therefore is a better approach to providing a packet transport network (or other controlled domain). The operator of an MPLS network has complete control over the separation of the user traffic and their infrastructure traffic.
>     >
>     > It is going to be really interesting to see whether SRv6 triumphs 
or crashes out because of the difficulty of providing the degree of traffic separation that is intrinsic to  MPLS.
>     >
> 
>     It has seemed to me that what is really missing is a general purpose
>     local network limited protocol, "larger" than a link-layer protocol,
>     yet "smaller" than a global internetworking protocol.
> 
>     IPv6 is really too "big" for SR. The IPv6 overhead is because of the
>     128 bit addresses, and they're that large because it is a global
>     internetworking protocol that has to be able to uniquely address every
>     node on the internetwork.
> 
>     It makes sense to try to use IPv6 for SR, since IPv6 is a future
>     commodity protocol, however as it is "too large", there are then hacks
>     like EH insertion or the SID compression proposals to try to get
>     around the fundamental problem of using a protocol that isn't really a
>     good enough fit for a local network problem and a solution like SR.
> 
>     MPLS is a local network protocol. Fine for use "inside the network"
>     and something like SR.
> 
>     However I think it would be useful if we had a more general purpose
>     local network protocol that is also well suited for use by hosts, and
>     that, for example, transport layer protocols like UDP or TCP could be
>     placed directly inside.
> 
>     I've idly wondered if we could repurpose IPv4 for that by giving it 
a
>     new version number. It's now too small to be a global internetworking
>     protocol, however would still be large enough to solve a local network
>     problem. It just needs to be distinguished from the legacy use of IPv4
>     as a global Internet protocol.
> 
>     It would be preferable though for a local network protocol to have
>     addresses large enough to be so that there can be likely globally
>     unique subnets, so that merging networks doesn't require renumbering
>     or NAT. I wonder if 64 bit addresses, 32 bits Global ID (similar to
>     ULA 48 bit ID), 16 bits subnets and 16 bits host addresses would be
>     good enough.
> 
>     Regards,
>     Mark.
> 
>     > - Stewart
>     >
>     >
>     > On 24 Jul 2021, at 00:00, Mark Smith <[e-mail] <mailto:[e-mail]>> wrote:
>     >
>     >
>     >
>     > On Fri, 23 Jul 2021, 18:20 Giuseppe Fioccola, <[e-mail] <mailto:[e-mail]>> wrote:
>     >>
>     >> Hi Mike,
>     >>
>     >> To avoid misunderstanding, the precondition of controlled domain 
may be kept as MUST. We can further specify that authentication MUST be used if, for specific scenarios, it is applied outside a controlled domain.
>     >
>     >
>     > Realise that a "MUST be limited to a controlled domain" in an RFC 
is nothing more than an aspiration. It's theory rather than reality.
>     >
>     > Packets are encouraged to try to exit "controlled" domains attached to the Internet due to the domain's default route, and then can leave the controlled domain ("leak") due failure of the controlling boundary because of implementation bugs, operator configuration error or partial node failure.
>     >
>     > Authentication must be a MUST for anything that is designed for a 
controlled domain if the controlled domain may be attached to the Internet, which is a possibility for any of them if they use IPv6.
>     >
>     > Packets getting to where they shouldn't would be one of the motivations of Postel's "Be conservative with what you send".
>     >
>     >
>     > Regards,
>     > Mark.
>     >
>     >
>     >>
>     >>
>     >> Regards,
>     >>
>     >>
>     >>
>     >> Giuseppe
>     >>
>     >>
>     >>
>     >>
>     >>
>     >> From: Mike Simpson <[e-mail] <mailto:[e-mail]>>
>     >> Sent: Friday, July 23, 2021 9:36 AM
>     >> To: Giuseppe Fioccola <[e-mail] <mailto:[e-mail]>>
>     >> Cc: Erik Kline <[e-mail] <mailto:[e-mail]>>; Yoshifumi Nishida <[e-mail] <mailto:[e-mail]>>; [e-mail] <mailto:[e-mail]>; Christopher Wood <[e-mail] <mailto:[e-mail]>>; [e-mail] <mailto:[e-mail]>
>     >> Subject: Re: I-D Action: draft-ietf-6man-ipv6-alt-mark-07.txt
>     >>
>     >>
>     >>
>     >> Why not just keep it at MUST so that you don’t pollute the internets.
>     >>
>     >>
>     >>
>     >> We will end up having to filter for it anyway as always but it seems foolhardy and unpleasant to intentionally weaken the language.
>     >>
>     >>
>     >>
>     >> Your new hotness belongs in your controlled domain. If you are going to try and force it onto networks you don’t control then it’s not going to work and you will end up having to tunnel it anyways.
>     >>
>     >>
>     >>
>     >> Why is this so hard to understand?
>     >>
>     >>
>     >>
>     >> On 22 Jul 2021, at 15:09, Giuseppe Fioccola <[e-mail] <mailto:[e-mail]>> wrote:
>     >>
>     >> 
>     >>
>     >> Hi Erik,
>     >>
>     >> Thanks for the input.
>     >>
>     >> I tend to agree that the condition “MUST” can be 
changed to “SHOULD”. I can address your comments in the -08 version.
>     >>
>     >>
>     >>
>     >> Regards,
>     >>
>     >>
>     >>
>     >> Giuseppe
>     >>
>     >>
>     >>
>     >> From: Erik Kline <[e-mail] <mailto:[e-mail]>>
>     >> Sent: Wednesday, July 21, 2021 11:15 PM
>     >> To: Giuseppe Fioccola <[e-mail] <mailto:[e-mail]>>
>     >> Cc: Stewart Bryant <[e-mail] <mailto:[e-mail]>>; Christopher Wood <[e-mail] <mailto:[e-mail]>>; Yoshifumi Nishida <[e-mail] <mailto:[e-mail]>>; [e-mail] <mailto:[e-mail]>; [e-mail] <mailto:[e-mail]>
>     >> Subject: Re: FW: I-D Action: draft-ietf-6man-ipv6-alt-mark-07.txt
>     >>
>     >>
>     >>
>     >> Giuseppe,
>     >>
>     >>
>     >>
>     >> I think in S2.1 "MUST NOT" be used outside a "controlled domain" 
is perhaps a bit too strong.  Similarly in S6, "MUST be applied in...controlled domains" might be moderated down to "SHOULD only be applied...".
>     >>
>     >>
>     >>
>     >> I'll note that it is possible for an AH option to be used to ensure the DstOpt variant is unmodified en route, and these two in conjunction can be used wherever desired to send such packets outside the given domain (subject, of course, to all the middlebox interference any such packet would inevitably receive -- but that's a separate issue).
>     >>
>     >>
>     >>
>     >> On Tue, Jun 22, 2021 at 11:27 AM Giuseppe Fioccola <[e-mail] <mailto:[e-mail]>> wrote:
>     >>
>     >> Dear Stewart, Christopher, Yoshi, All,
>     >> Please note that I just submitted a new version of the draft. It 
has been thoroughly reviewed to address the comments received during the Last Call.
>     >>
>     >> Your inputs are always welcome.
>     >>
>     >> Regards,
>     >>
>     >> Giuseppe
>     >>
>     >> -----Original Message-----
>     >> From: ipv6 <[e-mail] <mailto:[e-mail]>> On Behalf Of [e-mail] <mailto:[e-mail]>
>     >> Sent: Tuesday, June 22, 2021 8:13 PM
>     >> To: [e-mail] <mailto:[e-mail]>
>     >> Cc: [e-mail] <mailto:[e-mail]>
>     >> Subject: I-D Action: draft-ietf-6man-ipv6-alt-mark-07.txt
>     >>
>     >>
>     >> A New Internet-Draft is available from the on-line Internet-Drafts directories.
>     >> This draft is a work item of the IPv6 Maintenance WG of the IETF.
>     >>
>     >>         Title        
   : IPv6 Application of the Alternate Marking Method
>     >>         Authors         : Giuseppe Fioccola
>     >>                           Tianran Zhou
>     >>                           Mauro Cociglio
>     >>                           Fengwei Qin
>     >>                           Ran Pang
>     >>         Filename        : draft-ietf-6man-ipv6-alt-mark-07.txt
>     >>         Pages        
   : 21
>     >>         Date        
    : 2021-06-22
>     >>
>     >> Abstract:
>     >>    This document describes how the Alternate Marking Method can be used
>     >>    as a passive performance measurement tool in an IPv6 domain.  It
>     >>    defines a new Extension Header Option to encode Alternate Marking
>     >>    information in both the Hop-by-Hop Options Header and Destination
>     >>    Options Header.
>     >>
>     >>
>     >> The IETF datatracker status page for this draft is:
>     >> https://datatracker.ietf.org/doc/draft-ietf-6man-ipv6-alt-mark/ <https://datatracker.ietf.org/doc/draft-ietf-6man-ipv6-alt-mark/>
>     >>
>     >> There is also an htmlized version available at:
>     >> https://datatracker.ietf.org/doc/html/draft-ietf-6man-ipv6-alt-mark-07 <https://datatracker.ietf.org/doc/html/draft-ietf-6man-ipv6-alt-mark-07>
>     >>
>     >> A diff from the previous version is available at:
>     >> https://www.ietf.org/rfcdiff?url2=draft-ietf-6man-ipv6-alt-mark-07 <https://www.ietf.org/rfcdiff?url2=draft-ietf-6man-ipv6-alt-mark-07>
>     >>
>     >>
>     >> Internet-Drafts are also available by anonymous FTP at:
>     >> ftp://ftp.ietf.org/internet-drafts/ <ftp://ftp.ietf.org/internet-drafts/>
>     >>
>     >>
>     >> --------------------------------------------------------------------
>     >> IETF IPv6 working group mailing list
>     >> [e-mail] <mailto:[e-mail]>
>     >> Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6 <https://www.ietf.org/mailman/listinfo/ipv6>
>     >> --------------------------------------------------------------------
>     >>
>     >> --------------------------------------------------------------------
>     >> IETF IPv6 working group mailing list
>     >> [e-mail] <mailto:[e-mail]>
>     >> Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6 <https://www.ietf.org/mailman/listinfo/ipv6>
>     >> --------------------------------------------------------------------
>     >>
>     >> --------------------------------------------------------------------
>     >> IETF IPv6 working group mailing list
>     >> [e-mail] <mailto:[e-mail]>
>     >> Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6 <https://www.ietf.org/mailman/listinfo/ipv6>
>     >> --------------------------------------------------------------------
>     >
>     > --------------------------------------------------------------------
>     > IETF IPv6 working group mailing list
>     > [e-mail] <mailto:[e-mail]>
>     > Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6 <https://www.ietf.org/mailman/listinfo/ipv6>
>     > --------------------------------------------------------------------
>     >
>     >
> 
>     --------------------------------------------------------------------
>     IETF IPv6 working group mailing list
>     [e-mail] <mailto:[e-mail]>
>     Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6 
<https://www.ietf.org/mailman/listinfo/ipv6>
>     --------------------------------------------------------------------
>