Re: I-D Action: draft-gont-6man-address-usage-recommendations-00.txt

Brian E Carpenter <brian.e.carpenter@gmail.com> Sun, 29 May 2016 21:01 UTC

Subject: Re: I-D Action: draft-gont-6man-address-usage-recommendations-00.txt
References: <20160527233307.11106.7797.idtracker@ietfa.amsl.com>
From: Brian E Carpenter <brian.e.carpenter@gmail.com>
Organization: University of Auckland
To: 6man <ipv6@ietf.org>
Message-ID: <17696295-c5b3-b7f7-fb20-f07da2030df2@gmail.com>
Date: Mon, 30 May 2016 09:01:06 +1200
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:45.0) Gecko/20100101 Thunderbird/45.1.0
MIME-Version: 1.0
In-Reply-To: <20160527233307.11106.7797.idtracker@ietfa.amsl.com>
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
Archived-At: <http://mailarchive.ietf.org/arch/msg/ipv6/E8q2HmGQDfNzA0vY38WrwbE-Pxs>
Precedence: list

Hi,

> 
> 4.  Address Stability Considerations
...
>    Enforcing a maximum lifetime on IPv6 addresses may cause long-lived
>    TCP connections to fail.

That is only one example. Any scenario where a peer expects an address to
remain valid will fail. For example, any p2p application where some kind
of rendezvous or discovery process has occurred, such that the peers know
each other's addresses. Unstable addresses require significant complications
in such applications. The only good news is that mobility generally creates
the same problem, so many applications have already been fixed. (Related
problems were explored in 6renum, of course.)

> 6.  Advice on IPv6 Address Configuration
> 
>    [TBD]
> 
> 7.  Advice on IPv6 Address Usage
> 
>    [TBD]

I think we need to delete those sections. Stating the problem is about all we
can do today. Best, or least-bad, practice will emerge over a number of years.

   Brian

Re: I-D Action: draft-gont-6man-address-usage-rec… Brian E Carpenter
Re: I-D Action: draft-gont-6man-address-usage-rec… Fernando Gont
Re: I-D Action: draft-gont-6man-address-usage-rec… Brian E Carpenter