Eric,
> Thanks for addressing most of my comments, and sorry for the delay getting
> back to you. There are just a few comments that I want to discuss further.
>
> 1. Section 7.9, "Use of Ingress Replication with I-PMSI A-D routes".
>
> The document should make it clear that these procedures are required to
> be implemented (by all the PEs of a given MVPN) ONLY under the following
> conditions:
>
> - At least one of those PEs is a V-hub or V-spoke PE for the given MVPN.
>
> - The given MVPN is configured to use the optional procedure of using
> Ingress Replication to instantiate an I-PMSI.
Will do.
> 2. Abstract
>
> I continue to think that the second paragraph of the abstract is
> misleading in its claim that "the approach described in this document may
> allow to reduce bandwidth inefficiency", given that the approach may also
> allow an increase in bandwidth inefficiency. Maybe replace "may" with
> "may under certain circumstances".
Will do.
> (I also continue to think that the abstract would be improved if it
> mentioned that the draft uses "hub and spoke" procedures to provide
> any-to-any service, but that's just a suggestion.)
>
> 3. Multi-level hierarchy outside the scope of the document
>
> When I pointed out that the procedures don't appear to generalize easily
> to support of a multi-level hierarchy, you replied "multi-level hierarchy
> is outside the scope of this document". That's fine, but the document
> should have a statement to that effect.
Will do.
> 4. Installation of default route in a V-hub VRF
>
> I'm still having trouble understanding the point of the following
> paragraph:
>
> When a V-hub of a given VPN originates a VPN-IP default route for
> that VPN, the V-hub MUST NOT install in its VRF of that VPN a
> default route, unless this route has been originated either (a) as a
> result of the V-hub receiving an IP default route from one of the
> CEs of that VPN connected to it, or (b) as a result of the V-hub
> receiving (and importing) a VPN-IP default route from some other PE,
> or (c) the VRF being provisioned with a default route pointing to
> the routing table on the same PE that maintains the Internet routes.
>
> Could you give an example to show that this restriction is necessary?
Consider the example described in the first 5 paragraphs of section
8.1 of the draft (no Internet connectivity). If PE-3 (which is one
of the V-hubs) would install in its VRF a default route, then what
would be the next hop of that route ?
>
> 5. E/IBGP Load Balancing at a V-hub for packets whose label corresponds to
> the VPN-IP default route.
>
> This is an optimization that may be useful under certain circumstances,
> but not others. The draft should make it clear that this behavior is
> optional.
I think the draft already does this - from section 4:
When a multi-homed site is connected to a V-hub and a V-spoke, then
the V-hub uses the following OPTIONAL procedures to support IBGP/EBGP
^^^^^^^^
load balancing for the site's inbound traffic that has been
originated by some other V-spoke associated with the V-hub.
Yakov.
