Re: [bess] Queries to authors of https://www.ietf.org/archive/id/draft-ietf-bess-evpn-ip-aliasing-00.html

["Jorge Rabadan (Nokia)" <[e-mail]>]

Hi Saumya,

Having a different RD for the MAC/IP Advertisement route and the IP A-D per EVI route for the same ESI does not impose any issues. The RD should not be used in the EVPN IP route resolution process. But note this is also the case for Ethernet Aliasing.

As you can see in the route resolution section of this draft, you also need to use the IP A-D per ES route for the resolution of the EVPN IP route (being RT2 or RT5 in this document), and the RD of the IP A-D per ES route is a type 1 RD with the loopback followed by a unique number, and this would not match the IP-VRF or MAC-VRF RD.

So personally, I don’t see this causing any interop issues or any issues at all.

Having said that, there is a generic resolution issue for inter-domain option b, that prevents the mass withdraw (per ES) from working in these scenarios. If you were thinking about this, all the issues and potential solutions (including RD based correlation) are documented here:

draft-rabadan-bess-evpn-inter-domain-opt-b-03 (ietf.org)<https://datatracker.ietf.org/doc/html/draft-rabadan-bess-evpn-inter-domain-opt-b-03#section-3.1>

If you think we should highlight in this other draft that the RT2 RD and the IP A-D per EVI route RD will not match, it is something that we can certainly do. Just let us know.

Thanks.
Jorge



From: Dikshit, Saumya <[e-mail]>
Date: Wednesday, March 6, 2024 at 3:42 AM
To: Jorge Rabadan (Nokia) <[e-mail]>, Allu, Ramaprasad <[e-mail]>, [e-mail] <[e-mail]>, [e-mail] <[e-mail]>
Cc: [e-mail] <[e-mail]>
Subject: RE: Queries to authors of https://www.ietf.org/archive/id/draft-ietf-bess-evpn-ip-aliasing-00.html

CAUTION: This is an external email. Please be very careful when clicking links or opening attachments. See the URL nok.it/ext for additional information.


Hi Jorge,

I completely understand that IP AD per EVI route should carry the IP-VRF RD.
But with IP-aliasing, we are creating case where-in:

an attached ES is leveraged for both MAC-aliasing and IP-aliasing (host routes) via MAC routes and /32 routes respectively

Both are being published via the same NLRI in RT-2 as MAC/IP and L2VNI and L3VNI

There should be an common association (other than the ES.) between IP AD per EVI and the Host routes which are absorbed for IP-aliasing,

The common denominator should also include RD (configured on the EVI mapped to the vrf) on the send side

It’s confusing that RD carried in MAC/IP is the VLAN RD (as per EVPN standards, cannot content that).

But we are also signaling host-routes for layer-3 multi-homing and leveraging it RD as an index on the receive side.

Even though the corresponding IP-AD per EVI is signaled with vrf configured RD (and rightly so)

Somehow this is not coming together organically

We should call out the above mismatch (and/or absorption procedure for the IP-aliasing of host routes) in the draft.

Regards,
Saumya.

From: Jorge Rabadan (Nokia) [mailto:[e-mail]]
Sent: Tuesday, March 5, 2024 7:53 PM
To: Dikshit, Saumya <[e-mail]>; Allu, Ramaprasad <[e-mail]>; [e-mail]; [e-mail]
Cc: [e-mail]
Subject: Re: Queries to authors of https://www.ietf.org/archive/id/draft-ietf-bess-evpn-ip-aliasing-00.html

Hi Saumya,

This spec does not change anything for the advertisement of MAC/IP Advertisement routes or IP Prefix routes, it only introduces IP A-D per EVI/ES routes.

If you are using the same ES for the stretched Broadcast Domain and the IP-VRFs (Ethernet aliasing for layer-2 and IP Aliasing for layer-3), MAC/IP Advertisement routes are advertised with the RD of the MAC-VRF of origin and so are Ethernet A-D per EVI routes for the ES. IP A-D per EVI routes are advertised with the IP-VRF RD.

Hope it helps.

Thanks.
Jorge

From: Dikshit, Saumya <[e-mail]<mailto:[e-mail]>>
Date: Monday, March 4, 2024 at 6:40 PM
To: Dikshit, Saumya <[e-mail]<mailto:[e-mail]>>, Jorge Rabadan (Nokia) <[e-mail]<mailto:[e-mail]>>, Allu, Ramaprasad <[e-mail]<mailto:[e-mail]>>, [e-mail]<mailto:[e-mail]> <[e-mail]<mailto:[e-mail]>>, [e-mail]<mailto:[e-mail]> <[e-mail]<mailto:[e-mail]>>
Cc: [e-mail]<mailto:[e-mail]> <[e-mail]<mailto:[e-mail]>>
Subject: RE: Queries to authors of https://www.ietf.org/archive/id/draft-ietf-bess-evpn-ip-aliasing-00.html

CAUTION: This is an external email. Please be very careful when clicking links or opening attachments. See the URL nok.it/ext for additional information.


Just to elaborate further,
Below might be the scenario where both MAC aliasing (MAC routes) and IP-aliasing (for /32 host routes) is needed in mixed deployment of Symmetric IRB fabric.
Fabric may have mixed-bag of PEs, where-in, some of them have VRF-extended (/32 routes) while others have only subnet extended (MAC).

Regards,
Saumya.

From: BESS [mailto:[e-mail]] On Behalf Of Dikshit, Saumya
Sent: Tuesday, March 5, 2024 7:53 AM
To: Jorge Rabadan (Nokia) <jorge.rabadan=[e-mail]<mailto:jorge.rabadan=[e-mail]>>; Allu, Ramaprasad <[e-mail]<mailto:[e-mail]>>; [e-mail]<mailto:[e-mail]>; [e-mail]<mailto:[e-mail]>
Cc: [e-mail]<mailto:[e-mail]>
Subject: Re: [bess] Queries to authors of https://www.ietf.org/archive/id/draft-ietf-bess-evpn-ip-aliasing-00.html

Hi Jorge,

Is this the same RD for IP VRF (uniquely defined for IP-AD per EVI route), that should be leveraged for host routes in Route Type-2 and Prefix Routes in Route Type-5 ? Is that the safe assumption (as per rfc9135/9134) ?

In case yes, then If we have a unique RD for the IP-VRF, what RD should be used for the NLRI in the Route-Type-2 carrying MAC/IP for tied to MAC-VRF and IP-VRF ?  Should it be MAC-VRF RD or IP-VRF RD.

Regards,
Saumya.

From: BESS [mailto:[e-mail]] On Behalf Of Jorge Rabadan (Nokia)
Sent: Tuesday, March 5, 2024 1:57 AM
To: Allu, Ramaprasad <[e-mail]<mailto:[e-mail]>>; [e-mail]<mailto:[e-mail]>; [e-mail]<mailto:[e-mail]>
Cc: [e-mail]<mailto:[e-mail]>; Dikshit, Saumya <[e-mail]<mailto:[e-mail]>>
Subject: Re: [bess] Queries to authors of https://www.ietf.org/archive/id/draft-ietf-bess-evpn-ip-aliasing-00.html

Hi Ramaprasad,

About this:


“If MPLS label field is not considered for BGP route key, then BGP RIB will have only one route entry at any given point of time.

That is, IP A-D per EVI route overwrites Ethernet AD per EVI and vice-versa if same RD is used for IP-VRF and MAC-VRF.”

The Ethernet AD per EVI route and IP AD per EVI routes must use different RDs. It was sort of implicit, since the former one uses the RD of the MAC-VRF and the latter the RD of the IP-VRF, but we added this in rev 01 to make sure there is no misunderstanding:

          *  The Route-Distinguisher is for the corresponding IP-VRF.  The
            Route-Distinguisher allocated for the IP-VRF MUST be unique in the
            PE.

Hope it helps.

Thank you,
Jorge

From: Allu, Ramaprasad <[e-mail]<mailto:[e-mail]>>
Date: Sunday, March 3, 2024 at 10:04 PM
To: [e-mail]<mailto:[e-mail]> <[e-mail]<mailto:[e-mail]>>, [e-mail]<mailto:[e-mail]> <[e-mail]<mailto:[e-mail]>>
Cc: [e-mail]<mailto:[e-mail]> <[e-mail]<mailto:[e-mail]>>, Dikshit, Saumya <[e-mail]<mailto:[e-mail]>>
Subject: Re: Queries to authors of https://www.ietf.org/archive/id/draft-ietf-bess-evpn-ip-aliasing-00.html
You don't often get email from [e-mail]<mailto:[e-mail]>. Learn why this is important<https://aka.ms/LearnAboutSenderIdentification>


CAUTION: This is an external email. Please be very careful when clicking links or opening attachments. See the URL nok.it/ext for additional information.


Hi Authors,

Gentle reminder.
Can you please take a look at it and respond to below query?

Thanks,
Ramaprasad
From: Allu, Ramaprasad <[e-mail]<mailto:[e-mail]>>
Date: Wednesday, 21 February 2024 at 5:40 PM
To: [e-mail]<mailto:[e-mail]> <[e-mail]<mailto:[e-mail]>>, [e-mail]<mailto:[e-mail]> <[e-mail]<mailto:[e-mail]>>
Cc: [e-mail]<mailto:[e-mail]> <[e-mail]<mailto:[e-mail]>>, Dikshit, Saumya <[e-mail]<mailto:[e-mail]>>
Subject: Queries to authors of https://www.ietf.org/archive/id/draft-ietf-bess-evpn-ip-aliasing-00.html
Hi Authors of draft https://www.ietf.org/archive/id/draft-ietf-bess-evpn-ip-aliasing-00.html

I have a following query on the draft.  Please help with your response.


Context of section https://www.ietf.org/archive/id/draft-sajassi-bess-evpn-ip-aliasing-09.html#section-3.1,

In the section, it is mentioned that the construction of the IP A-D per EVI route is same as that of the Ethernet A-D per EVI route. The NLRI consists of the following,

                +---------------------------------------+

                |  Route Distinguisher (RD) (8 octets)  |

                +---------------------------------------+

                |Ethernet Segment Identifier (10 octets)|

                +---------------------------------------+

                |  Ethernet Tag ID (4 octets)           |

                +---------------------------------------+

                |  MPLS Label (3 octets)                |

                +---------------------------------------+



And as per https://www.rfc-editor.org/rfc/rfc7432.html#section-7.1, “ for the purpose of BGP route key processing, only the Ethernet

   Segment Identifier and the Ethernet Tag ID are considered to be part

   of the prefix in the NLRI.  The MPLS Label field is to be treated as

   a route attribute as opposed to being part of the route”



If MPLS label field is not considered for BGP route key, then BGP RIB will have only one route entry at any given point of time.

That is, IP A-D per EVI route overwrites Ethernet AD per EVI and vice-versa if same RD is used for IP-VRF and MAC-VRF.



Is there any reason for explicit mention of not using MPLS label field as key for BGP route or not carrying two labels one for Ethernet A-D per EVI and another for IP-AD per VRF?

In this case, I see only MPLS Label (VNI in case of VXLAN) is the distinguisher if same RD is used for both IP-VRF and MAC-VRF.



And to keep two separate routes in BGP RIB, we need to use MPLS label also one of the keys in addition to RD, ESI and ETAG fields.

Or Carry both the labels and extended communities as part of single A-D per EVI route and store single route in the global BGP RIB.



Please let me know what you think.



Thanks,

Ramaprasad