Re: [v6ops] [E] New Version Notification for draft-mishra-v6ops-variable-slaac-problem-stmt-01.txt

[Philip Homburg <[e-mail]>]

>While the draft has a particular solution in mind, I wouldn't mind discussing 
>the problems the draft enumerates.

I'm curious why existing mechanisms for distributing prefixes are not enough.

On landlines we have a very successful deployment of DHCPv6 IA_PD. We have
homenet for distributing prefixes within a site (and for simple cases, nested
DHCPv6 IUA_PD works as well).

We have RIR policies that allow sites to get a /48, also for individual
consumers.

So why then have people problems giving each subnet its own /64?

- Is this a protocol issue?
- Is a technical issue? (equipment that doesn't support prefix delegation)
- Is it economical and if so where? RIR fees, cost of equipment, support, etc.

For example, if customers of 5G mobile need to have multiple subnets behind
the mobile link, why do the mobile telcos only provide a /64?

Why does the draft list a home network as a problem (Section 4.4)? Is this
for the rare case that the upstream ISP doesn't support DHCPv6 IA_PD?

Somehow the draft completely fails to mention DHCP prefix delegation. Which
is weird because it is the main mechanism for provides sites with enough
space to number the sites' subnets.