How to choose IPv6 addresses for customer links?
Steve Bertrand
steve at ibctech.ca
Sat Jan 31 03:23:30 CET 2009
Gert Doering wrote:
> Hi,
>
> On Fri, Jan 30, 2009 at 02:38:00PM -0600, Dan White wrote:
>> In the case where a customer router is performing DHCP prefix
>> delegation, is the provider router installing a route at the time of
>> assignment?
>
> It better should. (But this is not RA, this is DHCP PD :-) ).
>
> Actually, I have no practical experience with DHCP PD yet - but I can't
> really think of any useful way to have DHCP PD without installing a route
> at delegation time.
>
>> With BGP, I suppose I will need to know the customer's two router
>> addresses (or link layer addresses) at configuration time.
>
> Yep. Either "know" or just mandate ("we have :0:1 and :0:2, you have
> :1:1 and :1:2, period").
...and, if you assign a /64 global between eBGP peers (as opposed to
anything longer), BGP will automagically configure the next-hop to the
link-local:
B>* 2001:478:235::/48 [20/0] via fe80::d8ea:2f09, gif1, 06:07:43
...as opposed to this route, where a /128 global PtP is in place:
B>* 2001:478:178::/48 [20/0] via 2001:4978:1:600::1, gif0, 06:08:14
What I haven't tested, but plan on doing so, is whether my theory that
moving the IPv6 peering address from one source interface to another
will not disrupt this link-local next-hop.
You (Dan) are going to like BGP. I have had much fortune finding people
to help me out with it, particularly in the v6 context. If you ever need
any help testing a setup in the global context, email me off-list. So
long as it's low-bandwidth, I'll do anything I can to pass along
knowledge I've gained from others (including sessions (non-transit),
VMs, test routers etc). When I help others learn, I'm either reinforcing
knowledge, or having to research the unknown.
Another thing that I'd like to point out, is that if _anyone_ is to
consider toying with BGP, read, understand and ensure that you follow
BCP 38, at minimum. Ask for guidance if necessary. This should be a
prerequisite to turning-up your first test session.
I've recently found that when one sanely organizes a question relating
to proper network behaviour prior to implementation, and posts it to a
high-profile list, the feedback is astounding. The 'big' good-guys
really seem to like helping out the 'little' guys who are trying to do
the right thing.
As a matter of fact, (in my experience), they will go miles out of their
way to help you become a trusted, albeit small, network.
Sorry for swaying OT. I just felt that if BGP was mentioned, a
recommendation to ensure network cleanliness was in order.
Steve
More information about the ipv6-ops
mailing list