Re: Heads up - Plea for allocating a /8 to ISPs

Chris,

Given the way the IETF "works" and based on the quantity of response this topic has generated, it would seem unlikely that a consensus can be reached before the IPv4 free pool is exhausted, regardless of the merits of the idea.  That would appear to leave folks deploying NAT444 with the option of each going to their respective RIRs and getting space (accelerating IPv4 free pool exhaustion) or everybody squatting on a hopefully allocated-but-forever-unused prefix.

With respect to the latter, I suspect the last damaging approach would be using something in 44/8. It appears _very_ sparsely populated but given its nature, it's unlikely IANA would be able to recover all of that particular block. Since IANA can't allocate anything but /8s, it is likely forever locked up... 

Regards,
-drc


On Nov 11, 2010, at 10:43 PM, Christopher LILJENSTOLPE wrote:
> It's not being used for internal devices, please re-read the draft.  It is to enable NAT444 to allow the delivery of IPv4 services, after run-out, using NAT444, to act as the v4 substrate between a CPE supporting 1918 to the customer, and the public internet.
> 
> 	Chris
> 
> On 12Nov2010, at 17.24, Sam Silvester wrote:
> 
>> On Thu, Nov 11, 2010 at 2:48 PM, Christopher LILJENSTOLPE
>> <[email protected]> wrote:
>>> -----BEGIN PGP SIGNED MESSAGE-----
>>> Hash: SHA1
>>> 
>>> A nit here, but the ask is for a /10.
>> 
>> I've had a read of the draft. Why a /10 (or whatever length?) - where
>> did this come from?
>> 
>> Assuming you're going to use this space for "internal devices" that
>> aren't capable of IPv6, it seems arbitrary.
>> 
>> I've heard comments that many ISPs are using many instances of RFC1918
>> address space for this purpose. You state the goal is to avoid clashes
>> with RFC1918 addressing on the customer side of the CPE - why not just
>> use your already allocated public space, using multiple instances and
>> the like to re-use a smaller block to scale to the required number of
>> addresses between your CPE WAN interfaces and the Big Ugly NAT Box?
>> 
>> Basically, it seems like a /10 might be fine for a number of carriers,
>> but not for others - as such it seems a hack to assign it and use a
>> "one size fits most" approach.
>> 
>> As such, I don't support this.
>> _______________________________________________
>> OPSAWG mailing list
>> [email protected]
>> https://www.ietf.org/mailman/listinfo/opsawg
>> 
> 
> ---
> 李柯睿
> Check my PGP key here:
> https://www.asgaard.org/~cdl/cdl.asc
> 
> _______________________________________________
> v6ops mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/v6ops

_______________________________________________
v6ops mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/v6ops