[OPSAWG] Fwd: I-D Action:draft-davies-reusable-ipv4-address-block-00.txt

[Christopher LILJENSTOLPE <[e-mail]>]

Begin forwarded message:

> From: Christopher LILJENSTOLPE <[e-mail]>
> Date: November 13, 2009 4:42:27 PM GMT+11:00
> To: Brian E Carpenter <[e-mail]>
> Cc: "Davies, Greg J" <[e-mail]>,  
> "[e-mail]" <[e-mail]>
> Subject: Re: [OPSAWG] I-D Action:draft-davies-reusable-ipv4-address- 
> block-00.txt
>
> Greetings Brian,
>
> 	I've talked to the AD's, and I am going to do another draft that  
> states this as a specific problem statement.  The Davies draft will  
> then refer to that as a informative reference.
>
> 	Chris
>
> On Nov 13, 2009, at 4:26 PM, Brian E Carpenter wrote:
>
>> Greg,
>>
>> This is exactly what needs to be stated clearly and simply
>> in the draft, IMHO.
>>
>>  Brian
>>
>> On 2009-11-13 18:16, Davies, Greg J wrote:
>>> Brian,
>>>      I agree that deploying IPv6 as soon as possible is very  
>>> important, but unfortunately it is only one part of the  
>>> transition.  Even if a service provider had IPv6 deployed today,  
>>> in a dual-stack approach, they are still likely to run out of  
>>> public IPv4 addresses following the global run-out.  The provider  
>>> will need to continue supporting IPv4 because customers are not  
>>> likely to have a complete IPv6 environment for many years (e.g.  
>>> they will have IPv4 only games consoles, Windows XP PCs without  
>>> IPv6 enabled, IPv4-only apps, other IPv4-only devices, etc.).
>>>
>>> The issue is not one of whether you can force service providers to  
>>> adopt IPv6 more quickly it is that you can't force customers to  
>>> completely overhaul their environement to be fully IPv6 in a short  
>>> timeframe.  We have to give them time to make the transition in  
>>> line with lifecycle replacement of their equipment.
>>>
>>> Providers thus have to do two things:
>>> 1. Introduce IPv6 as the long term solution
>>> 2. Implement a solution that will maintain IPv4 connectivity for  
>>> customers once public IPv4 addresses have been exhausted.
>>>
>>> Regards,
>>> Greg
>>>
>>>
>>> -----Original Message-----
>>> From: Brian E Carpenter [mailto:[e-mail]]
>>> Sent: Friday, 13 November 2009 3:50 PM
>>> To: Davies, Greg J
>>> Cc: [e-mail]
>>> Subject: Re: [OPSAWG] I-D Action:draft-davies-reusable-ipv4- 
>>> address-block-00.txt
>>>
>>> Greg,
>>>
>>> But that doesn't motivate either
>>>
>>> a) why the ISP would want to use private space
>>> or
>>> b) how this make IPv6 deployment happen sooner.
>>>
>>> Let me be clear about the counter-argument: if we
>>> refuse to allocate this new private space, ISPs will
>>> run out of IPv4 addresses sooner and will deploy IPv6
>>> faster. You may not like this but it's an argument
>>> I've heard.
>>>
>>>  Brian
>>>
>>> On 2009-11-13 17:35, Davies, Greg J wrote:
>>>> Thanks Brian for the feedback, perhaps we can bring out the  
>>>> motivation more strongly.
>>>>
>>>> The key motivation is to to avoid address conflicts with  
>>>> customers when the service provider is using private IPv4  
>>>> addresses (e.g. as in the NAT444 or the incremental CGN solutions).
>>>>
>>>> Many customers use the 10 range in their LAN evironments and in  
>>>> some cases it is as bad as having the subnet set as the whole /8  
>>>> (this may even have been the default CPE configuration).  Where  
>>>> there is an overlap between the LAN and WAN addresses the  
>>>> customer is likely to have problems and this will result in  
>>>> frustration and support calls.  Given most consumer customers  
>>>> don't need to know about IP addresses and subnets the best  
>>>> approach is to avoid any potential for address conflict in the  
>>>> first place.  Hence the need for an alternative address block to  
>>>> be used during the IPv4 to IPv6 transition phase.
>>>>
>>>> Regards
>>>> Greg Davies
>>>>
>>>>
>>>> -----Original Message-----
>>>> From: [e-mail] [mailto:[e-mail]] On  
>>>> Behalf Of Brian E Carpenter
>>>> Sent: Friday, 13 November 2009 2:59 PM
>>>> To: [e-mail]
>>>> Subject: Re: [OPSAWG] I-D Action:draft-davies-reusable-ipv4- 
>>>> address-block-00.txt
>>>>
>>>> Hi,
>>>>
>>>> As a straightforward request to the IANA, there isn't much to say
>>>> about this draft. However, it contains no motivation. I understand
>>>> that some ISPs believe this approach to be essential as part of an
>>>> IPv6 deployment strategy, but so far the draft-shirasaki-*
>>>> documents haven't (apparently) convinced the community that
>>>> the motivation is sufficient. In fact there is some fear that
>>>> allocating more ambiguous private address space will only make
>>>> things move more slowly.
>>>>
>>>> So, I believe that either this draft needs a convincing
>>>> motivation section, or a companion document, so that the
>>>> community can reach a considered consensus. I believe interest
>>>> in this will be extremely wide once it gets to a last call,
>>>> so the motivation needs to be very solid.
>>>>
>>>> IMHO.
>>>>
>>>>  Brian
>>>> _______________________________________________
>>>> OPSAWG mailing list
>>>> [e-mail]
>>>> https://www.ietf.org/mailman/listinfo/opsawg
>>>>
>>>
>> _______________________________________________
>> OPSAWG mailing list
>> [e-mail]
>> https://www.ietf.org/mailman/listinfo/opsawg
>>
>
> ---
> 李柯睿
> Check my PGP key here:
> https://www.asgaard.org/~cdl/cdl.asc
>

---
李柯睿
Check my PGP key here:
https://www.asgaard.org/~cdl/cdl.asc