[prev in list] [next in list] [prev in thread] [next in thread] 

List:       ipng
Subject:    Re: "In-Flight IPv6 Extension Header Insertion Considered Harmful" (draft-smith-6man-in-flight-eh-in
From:       Erik Nordmark <erik () zededa ! com>
Date:       2019-10-10 13:59:16
Message-ID: d68b6625-7583-c0cb-b396-1b728fa41390 () zededa ! com
[Download RAW message or body]

On 10/9/19 7:37 AM, Mark Smith wrote:
> Hi All,
> 
> Despite what was clarified in RFC 8200, it seems this ID is needed.

Mark,

I think there are critical technical aspects which are missing.
The key one to me is that ICMP error messages generated they would 
include the inserted headers, which unless removed at the boundary would 
confuse the original sender since the sender didn't include that 
extension header.

Thus if the goal for in-flight EH insertion is to be transparent to the 
rest of the network, the solution needs to rewrite the inside of any 
ICMP error that has been generated between the ingress and egress of the 
EH insertion.
If PMTUd is to work, that includes calculating the MTU to report to the 
originator.

There are some other issues if the path MTU between the ingress and 
egress gets close to 1280, because in the worst case that means that a 
1280 byte original can't make it across the in-flight path without 
someone fragmenting it, and PMTUd can't be used to get an IPv6 sender to 
fragment something which is 1280 bytes or less.

I can't tell from the draft whether you want to expand and focus on the 
technical description of the issues with in-flight insertion and the 
requirements if this is supposed to be transparent to the rest of the 
Internet, or if you want to expand and focus on the more philosophical 
arguments.

Regards,
   Erik

> 
> Looking for co-authors.
> 
> Thanks very much,
> Mark.
> 
> 
> ---------- Forwarded message ---------
> From: <internet-drafts@ietf.org>
> Date: Wed, 9 Oct 2019 at 16:30
> Subject: New Version Notification for
> draft-smith-6man-in-flight-eh-insertion-harmful-00.txt
> To: Mark Smith <markzzzsmith@gmail.com>
> 
> 
> 
> A new version of I-D, draft-smith-6man-in-flight-eh-insertion-harmful-00.txt
> has been successfully submitted by Mark Smith and posted to the
> IETF repository.
> 
> Name:           draft-smith-6man-in-flight-eh-insertion-harmful
> Revision:       00
> Title:          In-Flight IPv6 Extension Header Insertion Considered Harmful
> Document date:  2019-10-08
> Group:          Individual Submission
> Pages:          10
> URL:
> https://www.ietf.org/internet-drafts/draft-smith-6man-in-flight-eh-insertion-harmful-00.txt
> Status:
> https://datatracker.ietf.org/doc/draft-smith-6man-in-flight-eh-insertion-harmful/
> Htmlized:
> https://tools.ietf.org/html/draft-smith-6man-in-flight-eh-insertion-harmful-00
> Htmlized:
> https://datatracker.ietf.org/doc/html/draft-smith-6man-in-flight-eh-insertion-harmful
> 
> 
> Abstract:
>     In the past few years, as well as currently, there have and are a
>     number of proposals to insert IPv6 Extension Headers into existing
>     IPv6 packets while in flight.  This contradicts explicit prohibition
>     of this type of IPv6 packet proccessing in the IPv6 standard.  This
>     memo describes the possible failures that can occur with EH
>     insertion, the harm they can cause, and the existing model that is
>     and should continue to be used to add new information to an existing
>     IPv6 and other packets.
> 
> 
> 
> 
> Please note that it may take a couple of minutes from the time of submission
> until the htmlized version and diff are available at tools.ietf.org.
> 
> The IETF Secretariat
> 
> --------------------------------------------------------------------
> IETF IPv6 working group mailing list
> ipv6@ietf.org
> Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
> --------------------------------------------------------------------
> 

--------------------------------------------------------------------
IETF IPv6 working group mailing list
ipv6@ietf.org
Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------
[prev in list] [next in list] [prev in thread] [next in thread]