| | | | | | | | |
|
|
|
You're here: Home |
On Feb 26, 2007, at 10:54 PM, Randy Bush wrote:
The point of this example was to make it self-adjusting. As the free pool gets smaller, the restrictions increase automatically. By definition, it would extend the free pool as far as it needs to gothe problem with this is, in operational reality, when i can not get the allocation i operationally need, the ipv4 free pool may be there in name but it is useless because i will be forced to do something strange such as nat or ipv6.
Well, yes. Sometime in relatively near future, ISPs are going to get to the point where they have the following choice:
a) say to the customer "Sorry, we don't want your money" b) ask the customer "What do you _REALLY_ need?"c) say to the customer "Here is one (1) IPv4 address. Go NAT your enterprise"
d) say to the customer "Here is a /48 for IPv6. Go NAT your enterprise" e) wave money around and say to the black market "name your price!"regardless of whether the IPv4 countdown proposal goes for quick trauma or chronic pain. I guess I'm just skeptical that the address consuming community is going to be able to have will power to not immediately dig into the reserve on a case-by-case basis without any sort of policy framework to base requests on.
But as I've said, perhaps I'm too cynical. Rgds, -drc
