Re: [Idr] Part 3 of CAR/CT Adoption call (7/14/2022 to 7/27/2022) - Operational Differences

[Srihari Sangli <[e-mail]>]

Robert,

Comments inline.

Thanks & Regards,

srihari…


From: Idr <[e-mail]> on behalf of Robert Raszuk <[e-mail]>
Date: Tuesday, 26 July 2022 at 4:01 PM
To: Shraddha Hegde <[e-mail]>
Cc: idr@ietf. org <[e-mail]>, Sue Hares <[e-mail]>
Subject: Re: [Idr] Part 3 of CAR/CT Adoption call (7/14/2022 to 7/27/2022) - Operational Differences
[External Email. Be cautious of content]

Shraddha & now Jim,

Apologies but this is not what I was pointing at.

I was asking if you are really going to deploy RTC like pull mechanism on the inter-as boundaries (even under cooperating administration) ?

Especially here in CT proposal RTC pull model will be almost ineffective if what you request is color and there are a total 5 colors in the game.

So let's make it clear that the current scheme does not provide much for scale optimization.

The only option I could perhaps see helping with scale (or in different word uncontrolled spray of unneeded info) would be ORF like extension where ASN_X could signal ASN_Y which next hops are of specific interest. But then we run into again into the RD wall where it makes it even less possible to filter as you would have to wildcard front of the NLRI ,,, not good.

Bottom line is that CT is hard to scale and hard to aggregate.

Sure like Srihari pointed out let's just blast it and not worry about it. That's an honest answer - but is this good one ?

Srihari> I was not hinting that we not worry about it, I was questioning the practicality of such a aggregation from the domain BR. Each specific endpoint address can have uniqueness (described via many attributes not to mention the new intent), and I would argue the ingress PE might be more interested in knowing this detail than just point to the aggregate route (very similar to one using default vs more specific routing).

Srihari> But I don’t refute your claim that aggregation may be needed. Sure, it will help scalability, we all know that and I see both proposals have ability to do that. Nats responded in one of the prev emails.

srihari…


Many thx,
Robert.


On Tue, Jul 26, 2022 at 5:29 PM Shraddha Hegde <[e-mail]<mailto:[e-mail]>> wrote:
Robert,


Pls refer  below section in https://datatracker.ietf.org/doc/draft-hr-spring-intentaware-routing-using-color/<https://urldefense.com/v3/__https:/datatracker.ietf.org/doc/draft-hr-spring-intentaware-routing-using-color/__;!!NEt6yMaO-gk!AARYsHsTwrzsAgw-OiCaZDWrCqx_a6Q4QCejXWU90y2EpkLwDIq3uADSt2re6Wy9vtXEgwitw1HcQQc$>
6.1.1.  Transport Network Intent Requirements


“The requirements described in this document are mostly applicable to

   network under a single administrative domain that are organized into

   multiple network domains.  The requirements are also applicable to

   multi-AS networks with closely cooperating administration.”

Rgds
Shraddha



Juniper Business Use Only
From: Robert Raszuk <[e-mail]<mailto:[e-mail]>>
Sent: Tuesday, July 26, 2022 11:41 AM
To: Shraddha Hegde <[e-mail]<mailto:[e-mail]>>
Cc: Jeffrey Haas <[e-mail]<mailto:[e-mail]>>; idr@ietf. org <[e-mail]<mailto:[e-mail]>>; Sue Hares <[e-mail]<mailto:[e-mail]>>
Subject: Re: [Idr] Part 3 of CAR/CT Adoption call (7/14/2022 to 7/27/2022) - Operational Differences

[External Email. Be cautious of content]

Hi,

Is pull model really going to be deployed inter-as here ?

Many thx,
R.

On Tue, Jul 26, 2022, 06:34 Shraddha Hegde <[e-mail]<mailto:[e-mail]>> wrote:
Robert,

As per the current version of https://datatracker.ietf.org/doc/draft-hr-spring-intentaware-routing-using-color/<https://urldefense.com/v3/__https:/datatracker.ietf.org/doc/draft-hr-spring-intentaware-routing-using-color/__;!!NEt6yMaO-gk!GJBkRIbd2z2Uxc4w3M48GhSuxRdU5QPk2ln4S6m0gNzUbsfEIT6hanlUPC7vVD_Kjd1rhwcGLW4F0k72$>
Ability to aggregate the intent-aware route itself has not been listed as a specific requirement.
While ability to aggregate subscription routes has been listed as a requirement.

In case of MPLS forwarding plane, aggregation of intent aware routes is not feasible as you need distinct label for each <endpoint, color>. In case of SRv6, its feasible to summarize the intent-aware routes but it takes away
visibility into other granular information such as E2E metric, PE reachability etc and such a requirement
needs more discussion.

A pull model via automatic subscription routes has been listed as a scaling requirement for the solution in
Sec 6.3.2.1 of https://datatracker.ietf.org/doc/draft-hr-spring-intentaware-routing-using-color/<https://urldefense.com/v3/__https:/datatracker.ietf.org/doc/draft-hr-spring-intentaware-routing-using-color/__;!!NEt6yMaO-gk!GJBkRIbd2z2Uxc4w3M48GhSuxRdU5QPk2ln4S6m0gNzUbsfEIT6hanlUPC7vVD_Kjd1rhwcGLW4F0k72$>.
This is a much better scalability option IMO.

Rgds
Shraddha


Juniper Business Use Only

-----Original Message-----
From: Idr <[e-mail]<mailto:[e-mail]>> On Behalf Of Jeffrey Haas
Sent: Monday, July 25, 2022 7:45 PM
To: Robert Raszuk <[e-mail]<mailto:[e-mail]>>
Cc: idr@ietf. org <[e-mail]<mailto:[e-mail]>>; Sue Hares <[e-mail]<mailto:[e-mail]>>
Subject: Re: [Idr] Part 3 of CAR/CT Adoption call (7/14/2022 to 7/27/2022) - Operational Differences

[External Email. Be cautious of content]


Robert,

On Mon, Jul 25, 2022 at 04:00:32PM +0200, Robert Raszuk wrote:
> 2.5 min of BGP propagation time for 1.5 million routes (as Jeff
> stated). Do you think it is not worth to consider an aggregation in such a case ?

The 1.5M routes is from the problem statement in SPRING.  I'd suggest you take it up with that WG as to whether they think these were intended to be aggregatable.

Discussion of what aggregation could look like in a proposal is appropriate for IDR.

> > Trolling will result in your posts being put into the moderation queue.
>
> Part 3 is about operational differences. Not every difference requires
> a topology - some do some do not.
>
> I quoted sections from both drafts ?
>
> I do think this fear of moderation - when just pointing out
> operational differences and asking clarification questions - simply inappropriate.

It's reasonable to ask the authors how they intend to address aggregation.
It's reasonable to point out text you think supports your thinking.

The following is not reasonable.

"Till that is fixed I recommend that CT draft goes back to the drawing board and would not be even accepted as experimental. Its experiment has just concluded as a failure."

Seriously.  Stop it.

-- Jeff

_______________________________________________
Idr mailing list
[e-mail]<mailto:[e-mail]>
https://urldefense.com/v3/__https://www.ietf.org/mailman/listinfo/idr__;!!NEt6yMaO-gk!ECHaucPlCQgzPUSvACwT0XMDE-oIyExzTyg-OFnep1Tp6WTMEDFBGtqO_t0QbgSod4LHmSUcGA0RJQ$<https://urldefense.com/v3/__https:/www.ietf.org/mailman/listinfo/idr__;!!NEt6yMaO-gk!ECHaucPlCQgzPUSvACwT0XMDE-oIyExzTyg-OFnep1Tp6WTMEDFBGtqO_t0QbgSod4LHmSUcGA0RJQ$>


Juniper Business Use Only