evpn rebase to HEAD

[Ondrej Zajicek]

On Wed, Mar 11, 2026 at 10:06:06AM -0400, Pim van Pelt via Bird-users wrote:
> Hoi,
> 
> On 11.03.2026 08:41, Ondrej Zajicek wrote:
> > I think it is because you have the same route distinguisher 8298:200 on
> > all these routers. If i understand it correctly, each router should use
> > different RD (while they use the same route target (RT) if they are in
> > the same VPN).
> I interpreted RFC 7432, Section 7.9 differently
> "An RD MUST be assigned for a given MAC-VRF on a PE. *This RD MUST be unique
> across all MAC-VRFs on a PE*. It is RECOMMENDED to use the Type 1 RD
> [RFC4364]. The value field comprises an IP address of the PE (typically, the
> loopback address) followed by a number unique to the PE."
> 
> (emphasis mine) While the RFC mandates uniqueness only within a single PE
> (across its MAC-VRFs), it als recommends Type 1 RDs using the PE's loopback
> IP, which happens to produce globally unique RDs across the network.

My reading of this section is that RD must be unique per EVI (EVPN Instance).

If you have two PEs that are part of the same EVI, that means PE1 has a MAC-VRF
that contains routes from PE2 (and vice versa), therefore if PE1 uses the same
RD as PE2, then such RD would not be unique in that MAC-VRF on PE1.

In your case all three routes are in the same EVPN instance as they are
exported to one MAC-VRF (i.e. etab).


> However the important bits are that (a) I can now rely on etab having the
> multiple IMETs as you said, so I can simplify my vppevpn protocol to rely
> only on etab, and not evpntab; and (b) I learned a lot :) Thank you so much!

You are welcome!

-- 
Elen sila lumenn' omentielvo

Ondrej 'Santiago' Zajicek (email: santiago at crfreenet.org)
"To err is human -- to blame it on a computer is even more so."