Hi, Greg,

That, to me, reads like over-specifying. Are you aware of any implementation 
attempting to do that?

There’s no text for that comparison – why would an implementation do that?

Thanks,

Carlos.

From: Greg Mirsky <[email protected]>
Date: Thursday, March 22, 2018 at 12:18 PM
To: "Reshad Rahman (rrahman)" <[email protected]>
Cc: Carlos Pignataro <[email protected]>, "[email protected]" <[email protected]>, 
"[email protected]" <[email protected]>
Subject: Re: [mpls] New Version Notification for 
draft-mirsky-mpls-bfd-bootstrap-clarify-00.txt

Hi Reshad,
your analysis of the text is absolutely correct. My concern is the possible 
analysis of the Discriminator by ingress LER and it comparing to My 
Discriminator of the received BFD control packets. According to the RFC 5884 
both must match but because LSP Echo reply does not provide sufficient 
information the most likely outcome will be - no matching BFD session found and 
thus the text fails. (Yes, it would be rather naive implementation but ...). 
Thus I propose to add explicit statement for ingress LER not to use information 
if received Discriminator TLV in Echo reply.

Regards,
Greg

On Thu, Mar 22, 2018 at 3:39 PM, Reshad Rahman (rrahman) 
<[email protected]<mailto:[email protected]>> wrote:
Hi,

In the BFD meeting yesterday there was discussion about lack of clarity on what 
the spec says wrt to the discriminator TLV being sent by the egress node in the 
echo reply and that this causes interop issues. From RFC 5884:

The egress LSR MAY respond with an LSP Ping Echo
   reply message that carries the local discriminator assigned by it for
   the BFD session.

In the errata:

The LSP Ping

Echo reply message generated by the egress LSR MAY carry the local

discriminator assigned by it for the BFD session, as specified in

section 6.1.
So I think it’s clear that this cannot be the discriminator of the ingress 
node. I agree that this information is useless but still don’t see how it can 
cause any harm, and any implementation which interprets the discriminator in 
the echo reply differently is buggy IMHO.

Regards,
Reshad (hat off).

From: Rtg-bfd <[email protected]<mailto:[email protected]>> on 
behalf of "Reshad Rahman (rrahman)" 
<[email protected]<mailto:[email protected]>>
Date: Tuesday, March 20, 2018 at 5:49 PM
To: Greg Mirsky <[email protected]<mailto:[email protected]>>, "Carlos 
Pignataro (cpignata)" <[email protected]<mailto:[email protected]>>

Cc: "[email protected]<mailto:[email protected]>" 
<[email protected]<mailto:[email protected]>>, 
"[email protected]<mailto:[email protected]>" 
<[email protected]<mailto:[email protected]>>
Subject: Re: [mpls] New Version Notification for 
draft-mirsky-mpls-bfd-bootstrap-clarify-00.txt

Hi,

While I agree that the echo reply is not needed to bootstrap BFD, and that the 
BFD Disc TLV is not needed in the reply, doing this doesn’t break anything. So 
I don’t see the proposed changes as being necessary.

Does anyone remember why RFC5884  has the echo reply, was it to potentially 
save an echo request from egress for bidirectional case?

Also, if we do go ahead with the proposed changes in this draft, we’ll have to 
fix this errata<https://www.rfc-editor.org/errata/eid5085>.

Regards,
Reshad (speaking as individual contributor).

From: Rtg-bfd <[email protected]<mailto:[email protected]>> on 
behalf of Greg Mirsky <[email protected]<mailto:[email protected]>>
Date: Friday, October 20, 2017 at 4:19 PM
To: "Carlos Pignataro (cpignata)" 
<[email protected]<mailto:[email protected]>>
Cc: "[email protected]<mailto:[email protected]>" 
<[email protected]<mailto:[email protected]>>, 
"[email protected]<mailto:[email protected]>" 
<[email protected]<mailto:[email protected]>>
Subject: Re: [mpls] New Version Notification for 
draft-mirsky-mpls-bfd-bootstrap-clarify-00.txt

Hi Carlos,
thank you for taking interest in the proposal, much appreciated. Please find my 
notes in-line and tagged GIM>>.

Regards,
Greg

On Fri, Oct 20, 2017 at 5:54 AM, Carlos Pignataro (cpignata) 
<[email protected]<mailto:[email protected]>> wrote:
Greg,

This document seems to say “use “Do not Reply” reply mode, and even if you 
reply do not use the BFD Disc TLV, because it is not used.
GIM>> To be precise it says "SHOULD use "Do not Reply" thus preserving 
compliance of implementations that do otherwise.

Wouldn’t it be simpler to say “follow RFC 8029, and the ingress does not care 
about the BFD Disc TLV in the reply”? This would not suddenly make uncompliant 
existing implementations, potentially.
GIM>> I agree that normative language on handling echo reply is bit 
restrictive. My goal is to have good discussion and see what others think.

Also I wonder if this should be bfd-mpls instead of mpls-bfd, given where RFC 
5884 was advanced.
GIM>> Probably it should be the way you've suggested. Hope it is not a big 
problem for individual draft.

Thanks,

—
Carlos Pignataro, [email protected]<mailto:[email protected]>

“Sometimes I use big words that I do not fully understand, to make myself sound 
more photosynthesis."

On Oct 18, 2017, at 8:50 AM, Greg Mirsky 
<[email protected]<mailto:[email protected]>> wrote:

Dear All,
this new document proposes clarification of two questions brought up in course 
of recent discussion of RFC 5884:

  *   use of Return mode values in bootstrapping BFD session echo request;
  *   inclusion of BFD Discriminator TLV in echo response to the bootstrapping 
echo request.
Your comments, questions are always welcome and greatly appreciated.

Regards,
Greg

---------- Forwarded message ----------
From: <[email protected]<mailto:[email protected]>>
Date: Wed, Oct 18, 2017 at 5:46 AM
Subject: New Version Notification for 
draft-mirsky-mpls-bfd-bootstrap-clarify-00.txt
To: Gregory Mirsky <[email protected]<mailto:[email protected]>>, 
Yanhua Zhao <[email protected]<mailto:[email protected]>>



A new version of I-D, draft-mirsky-mpls-bfd-bootstrap-clarify-00.txt
has been successfully submitted by Greg Mirsky and posted to the
IETF repository.

Name:           draft-mirsky-mpls-bfd-bootstrap-clarify
Revision:       00
Title:          Clarifying Use of LSP Ping to Bootstrap BFD over MPLS LSP
Document date:  2017-10-18
Group:          Individual Submission
Pages:          4
URL:            
https://www.ietf.org/internet-drafts/draft-mirsky-mpls-bfd-bootstrap-clarify-00.txt
Status:         
https://datatracker.ietf.org/doc/draft-mirsky-mpls-bfd-bootstrap-clarify/
Htmlized:       
https://tools.ietf.org/html/draft-mirsky-mpls-bfd-bootstrap-clarify-00
Htmlized:       
https://datatracker.ietf.org/doc/html/draft-mirsky-mpls-bfd-bootstrap-clarify-00


Abstract:
   This document, if approved, updates RFC 5884 by clarifying procedures
   for using MPLS LSP ping to bootstrap Bidirectional Forwarding
   Detection (BFD) over MPLS Label Switch Path.




Please note that it may take a couple of minutes from the time of submission
until the htmlized version and diff are available at 
tools.ietf.org<http://tools.ietf.org/>.

The IETF Secretariat

_______________________________________________
mpls mailing list
[email protected]<mailto:[email protected]>
https://www.ietf.org/mailman/listinfo/mpls



Reply via email to