On Tue, May 23, 2017 at 12:41 PM, The IESG <iesg-secretary@xxxxxxxx> wrote:
The IESG has received a request from the IPv6 Operations WG (v6ops) to
consider the following document:
- 'Unique IPv6 Prefix Per Host'
<draft-ietf-v6ops-unique-ipv6-prefix-per-host-03.txt> as Best Current
Practice
The IESG plans to make a decision in the next few weeks, and solicits
final comments on this action.
I think this is a desirable way to configure public IPv6 network access and support the publication of this document as BCP. I see a couple issues that probably need to be resolved before publication:
1. Off-link flag.
If the UE/
subscriber desires to send anything external including other UE/
subscriber devices (assuming device to device communications is
enabled and supported), then, due to the L-bit set, it SHOULD send
this traffic to the First Hop Provider Router.
I think the document might need to say what the First Hop Provider Router needs to do in the case when it the host sends the router a packet for a destination in the same prefix as the host. (This can happen since L=0.) It seems that the router could simply turn around and send an NS for that IPv6 address, which would likely just end up going back to the host. Seems like we might want to avoid that.
2. If this document is to be a BCP, then the RA timers should probably not be a part of this document. They are over-prescriptive, and they conflict with the suggestions in RFC 7772, which says:
In order to limit the amount of power used to receive Router
Advertisements to, say, 2% of idle power (i.e., to impact idle
battery life by no more than 2%), the average power budget for
receiving RAs must be no more than 0.1 mA, or approximately 7 RAs
per hour.
That argues for an RA interval of > 500 seconds, which is half the power consumption of the value in this document (300 seconds).
Cheers,
Lorenzo