Re: 6man w.g. last call for <draft-ietf-6man-default-iids-11.txt>

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

On 22/05/2016 13:48, Mark Smith wrote:
> On 22 May 2016 at 01:16, Fernando Gont <[e-mail]> wrote:
>> On 05/19/2016 10:18 PM, Brian E Carpenter wrote:
>>> On 20/05/2016 12:48, Lorenzo Colitti wrote:
>>> ...
>>>> Let's all always keep in mind that the current text of the draft says that
>>>> on every network, hosts SHOULD configure an IPv6 address that never changes
>>>> until the end of time. I think that recommendation is irresponsible.
>>>
>>> Thus, I'm continuing to run my laptop with only an RFC4941 address for a few
>>> days to see whether it has any negative impact at all. So far, so good.
>>
>> This is a key test: Open a bunch of ssh sessions, and check if they
>> break overnight.
>>
>> RFC4941 would make them break. That's an interoperability problem that,
>> in the general case, you don't want to have, I think.

Well, my SSH sessions seem to time out anyway, because the server end
is unhappy about long sessions with no traffic. Look, it all depends on
the user scenario. It's pretty clear that for casual (coffee shop or
such like) use stable addresses are pointless. But for any scenario
with connections that might last for days, it's different.

Anyway, my point is that this is an issue with RFC4941. Since your draft
is an if statement with no else clause:
 if stable_address_wanted
   then RFC7217
issues in RFC4941 are out of scope.

>>
> 
> I think there would be a lot of very persistent connections (e.g.,
> last weeks and months) within Enterprise networks that aren't typical
> on the Internet e.g., long lived connections between database clients
> and servers, where the database client in this example is part of the
> application middle tier.

Yes. In that case stable_address_wanted == True.

    Brian