[bess] Re: FW: I-D Action: draft-ietf-bess-rfc7432bis-09.txt

["Ali Sajassi (sajassi)" <[e-mail]>]

Hi Greg,

In order to ensure that the text in draft-ietf-mpls-1stnibble<https://datatracker.ietf.org/doc/draft-ietf-mpls-1stnibble/> is consistent with the text in rfc7432bis, I would suggest something along the lines of the following changes for your document:

<Although there are scenarios where control word may not be needed for Ethernet embedded packets, for simplicity and ease of implementation, this document requires that a control word MUST  always be inserted as a PSH for such packets. If a document wants to relax this requirement from “MUST” to “SHOULD”, then it shall describe such scenarios.>

Cheers,
Ali

From: Greg Mirsky <[e-mail]>
Date: Friday, June 7, 2024 at 2:09 PM
To: Ali Sajassi (sajassi) <[e-mail]>, mpls <[e-mail]>, MPLS Working Group <[e-mail]>, Adrian Farrel <[e-mail]>
Cc: Menachem Dodge <[e-mail]>, [e-mail] <[e-mail]>, [e-mail] <[e-mail]>, [e-mail] <[e-mail]>
Subject: Re: [bess] FW: I-D Action: draft-ietf-bess-rfc7432bis-09.txt
Hi Ali,
thank you for your response. I have to admit that I still cannot find the relationship between the label distribution protocol and topology of an LSP, on the one hand, and the load-balancing mechanism of a P node in the MPLS data plane.
To the best of my understanding, draft-ietf-mpls-1stnibble<https://datatracker.ietf.org/doc/draft-ietf-mpls-1stnibble/> is in the WG LC<https://mailarchive.ietf.org/arch/msg/mpls/SvSx0_ACHOsJ6R9LG-2XxB60H58/>. I think that it is best to share comments and concerns about the requirement to use Control Word for the encapsulation of a non-IP payload in the MPLS networks formulated in draft-ietf-mpls-1stnibble<https://datatracker.ietf.org/doc/draft-ietf-mpls-1stnibble/>.

Regards,
Greg

On Thu, Jun 6, 2024 at 10:14 PM Ali Sajassi (sajassi) <[e-mail]<mailto:[e-mail]>> wrote:
Hi Greg,

Section 18 of RFC7432bis has been carefully worded to ensure its accuracy specially wrt “SHOULD” and “MUST” keywords. We cannot blindly require the use of control word for all non-IP payloads (e.g., Ethernet payload) as it depends on a) type of tunnels used (TE vs. non-TE), b) unicast vs. multicast (MP2P vs. P2MP), and c) usage of entropy label network wide. So, if there is a contradiction between draft-ietf-mpls-1stnibble<https://datatracker.ietf.org/doc/draft-ietf-mpls-1stnibble/> and RFC7432bis, I would suggest changing draft-ietf-mpls-1stnibble<https://datatracker.ietf.org/doc/draft-ietf-mpls-1stnibble/>.

Cheers,
Ali

From: Greg Mirsky <[e-mail]<mailto:[e-mail]>>
Date: Thursday, June 6, 2024 at 9:07 AM
To: Ali Sajassi (sajassi) <[e-mail]<mailto:[e-mail]>>
Cc: Menachem Dodge <[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]>>, [e-mail]<mailto:[e-mail]> <[e-mail]<mailto:[e-mail]>>
Subject: Re: [bess] FW: I-D Action: draft-ietf-bess-rfc7432bis-09.txt
Hi Ali,
thank you for the detailed response. Please find my follow up notes inlined below under the GIM>> tag.

Regards,
Greg

On Wed, Jun 5, 2024 at 10:51 PM Ali Sajassi (sajassi) <[e-mail]<mailto:[e-mail]>> wrote:
Hi Greg,

The questions that was asked initially are different that your questions. But let me answer them all here.

The initial question was why not use the control word even when entropy label is used by all network nodes and my answer is that I don’t see a need for it and if you do, can you explain why we need the control word when there is no possibility of out of order delivery in the presence of ECMP when the network uses entropy label.
GIM>> I agree, if it is certain that all the PEs and Ps are capable of handling an Entropy label and all the PEs apply it in the EVPN encapsulation, then the use of the Control Word is optional. But I cannot find in the draft that that is explicitly explained.

The text in 7.11 says that the control word should be used in absence of entropy label.
GIM>> And that is not a requirement but only a recommendation concerns me. I believe that based on draft-ietf-mpls-1stnibble<https://datatracker.ietf.org/doc/draft-ietf-mpls-1stnibble/> it must be a requirement.

Regarding your suggestion of the control word must be enabled always, it should not and it should be per operator control. Imagine that the PE (and the network) can do both entropy label and control word and the operator wants to use entropy label, therefore, it disables the control word locally!
GIM>> If an implementation interprets the administrative state of Control Word in this way, then I agree with you. But the draft doesn't tell the reader that if the local state of Control Word is disabled, that means that the PE node uses the Entropy label for load-balancing. Personally, I would refer to these states as Use Control Word/Use Entropy Label.

Regarding why using “SHOULD” instead of “MUST” because it is just a recommendation and the packet flow can work without it (i.e., without having out-of-order delivery).
GIM>> And that seems to contradict draft-ietf-mpls-1stnibble<https://datatracker.ietf.org/doc/draft-ietf-mpls-1stnibble/>.

Cheers,
Ali

From: Greg Mirsky <[e-mail]<mailto:[e-mail]>>
Date: Wednesday, June 5, 2024 at 2:06 PM
To: Ali Sajassi (sajassi) <[e-mail]<mailto:[e-mail]>>
Cc: Menachem Dodge <[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]>>, [e-mail]<mailto:[e-mail]> <[e-mail]<mailto:[e-mail]>>
Subject: Re: [bess] FW: I-D Action: draft-ietf-bess-rfc7432bis-09.txt
Hi Ali,
thank you for your question. Section 7.11, as I understand it, states:
                It is
                recommended that the control word be included in the
                absence of an entropy label [RFC6790].
If I understand correctly, the CW SHOULD be used, thus allowing for sending EVPN packets without the Control Word if node doesn't support the Entropy label. Correct?
Furthermore, I have a concern regarding the local control of the Control Word, as described in
   When the L2-Attr Extended Community is received from a remote PE, the
   control word C flag MUST be checked against local control word
   enablement.
I believe that local policy must always enable the Control Word.
Also, I have questions about rules 2 and 3 listed in Section 18 (rule 1 is, IMHO, correct):
   *  If a network uses deep packet inspection for its ECMP, then the
      the following rules for "Preferred PW MPLS Control Word" [RFC4385]
      apply:
      -  It MUST be used with the value 0 (e.g., a 4-octet field with a
         value of zero) when sending unicast EVPN-encapsulated packets
         over an MP2P LSP.

      -  It SHOULD NOT be used when sending EVPN-encapsulated packets
         over a P2MP or P2P RSVP-TE LSP.

      -  It SHOULD be used with the value 0 when sending EVPN-
         encapsulated packets over a mLDP P2MP LSP.  There can be
         scenarios where multiple links or tunnels can exist between two
         nodes and thus it is important to ensure that all packets for a
         given flows take the same link (or tunnel) between the two
         nodes.
Why are cases listed in these two rules not using MUST?

Regards,
Greg

On Tue, Jun 4, 2024 at 10:00 PM Ali Sajassi (sajassi) <[e-mail]<mailto:[e-mail]>> wrote:
Hi Greg, Menachem:

I believe during the Greg’s presentation at the BESS WG (which I was attending remotely), I voiced my concerns regarding mandating control word for all cases. So, let me repeat it in context of your comment:

Why do we need to mandate control word when all nodes in a network use entropy label for ECMP load balancing?


Cheers,
Ali

From: Greg Mirsky <[e-mail]<mailto:[e-mail]>>
Date: Thursday, May 30, 2024 at 8:20 PM
To: Menachem Dodge <[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: [bess] FW: I-D Action: draft-ietf-bess-rfc7432bis-09.txt
Dear All,
I share Menachem's concerns and welcome feedback from the authors.

Regards,
Greg

On Sun, May 5, 2024 at 12:33 AM Menachem Dodge <[e-mail]<mailto:[e-mail]>> wrote:
Hello Authors,

Just wondering why none of the discussion held at Brisbane meeting in March and subsequently on the emailing list regarding the PFN ( see the emails with subject: “Re: [bess] PFN questions in rfc4732bis” )  requesting changes in setion 7.11.1 and section 18 , were not included in the latest draft update.

I think the last email on this subject was sent on 15th April 2024.


In section 7.11 following the discussions I think that the following sentence should be removed:
“It is recommended that the control word be included in the absence of an entropy label [RFC6790].”


 In section 18 “If a network (inclusive of all PE and P nodes) uses entropy labels

      per [RFC6790] for ECMP load balancing, then the control word may

      not be used.



Should be changed to:  “If a network (inclusive of all PE and P nodes) uses entropy labels

      per [RFC6790] for ECMP load balancing, then the control word should

      be used, refer to draft-ietf-mpls-1stnibble



Thank you kindly,

Best Regards,
Menachem Dodge


From: BESS <[e-mail]<mailto:[e-mail]>> on behalf of [e-mail]<mailto:[e-mail]> <[e-mail]<mailto:[e-mail]>>
Date: Friday, 3 May 2024 at 7:42
To: [e-mail]<mailto:[e-mail]> <[e-mail]<mailto:[e-mail]>>
Cc: [e-mail]<mailto:[e-mail]> <[e-mail]<mailto:[e-mail]>>
Subject: [bess] I-D Action: draft-ietf-bess-rfc7432bis-09.txt
CAUTION: External E-Mail - Use caution with links and attachments


Internet-Draft draft-ietf-bess-rfc7432bis-09.txt is now available. It is a
work item of the BGP Enabled ServiceS (BESS) WG of the IETF.

   Title:   BGP MPLS-Based Ethernet VPN
   Authors: Ali Sajassi
            Luc Andre Burdet
            John Drake
            Jorge Rabadan
   Name:    draft-ietf-bess-rfc7432bis-09.txt
   Pages:   73
   Dates:   2024-05-02

Abstract:

   This document describes procedures for Ethernet VPN (EVPN), a BGP
   MPLS-based solution which addresses the requirements specified in the
   corresponding RFC - "Requirements for Ethernet VPN (EVPN)".  This
   document obsoletes RFC7432 (BGP MPLS-Based Ethernet VPN) and updates
   RFC8214 (Virtual Private Wire Service Support in Ethernet VPN).

The IETF datatracker status page for this Internet-Draft is:
https://urldefense.proofpoint.com/v2/url?u=https-3A__datatracker.ietf.org_doc_draft-2Dietf-2Dbess-2Drfc7432bis_&d=DwICAg&c=euGZstcaTDllvimEN8b7jXrwqOf-v5A_CdpgnVfiiMM&r=cezglEhs6Oa_CKN9mhFbT8T8kmWwaNdtBDjE9bvBG_E&m=gDpQwIZuZSEOcOuIUV_9_jeGv5m-aqXgzBMzkuCM8wBeIKaKwaQUthJPFuNNZ9Dh&s=Xt33XJv3urxYTFARXBfpdw-RopowitrC7SWSv-L-QBY&e=

There is also an HTMLized version available at:
https://urldefense.proofpoint.com/v2/url?u=https-3A__datatracker.ietf.org_doc_html_draft-2Dietf-2Dbess-2Drfc7432bis-2D09&d=DwICAg&c=euGZstcaTDllvimEN8b7jXrwqOf-v5A_CdpgnVfiiMM&r=cezglEhs6Oa_CKN9mhFbT8T8kmWwaNdtBDjE9bvBG_E&m=gDpQwIZuZSEOcOuIUV_9_jeGv5m-aqXgzBMzkuCM8wBeIKaKwaQUthJPFuNNZ9Dh&s=oBT0K_2O-jJC2YfcS2X7Srom1ebB2VtVjfyN0CSBZpw&e=

A diff from the previous version is available at:
https://urldefense.proofpoint.com/v2/url?u=https-3A__author-2Dtools.ietf.org_iddiff-3Furl2-3Ddraft-2Dietf-2Dbess-2Drfc7432bis-2D09&d=DwICAg&c=euGZstcaTDllvimEN8b7jXrwqOf-v5A_CdpgnVfiiMM&r=cezglEhs6Oa_CKN9mhFbT8T8kmWwaNdtBDjE9bvBG_E&m=gDpQwIZuZSEOcOuIUV_9_jeGv5m-aqXgzBMzkuCM8wBeIKaKwaQUthJPFuNNZ9Dh&s=qjFH58VBc_cT930wv8yqvpU4plxuyfST4kkQHhRr5q4&e=

Internet-Drafts are also available by rsync at:
rsync.ietf.org::internet-drafts


_______________________________________________
BESS mailing list
[e-mail]<mailto:[e-mail]>
https://urldefense.proofpoint.com/v2/url?u=https-3A__www.ietf.org_mailman_listinfo_bess&d=DwICAg&c=euGZstcaTDllvimEN8b7jXrwqOf-v5A_CdpgnVfiiMM&r=cezglEhs6Oa_CKN9mhFbT8T8kmWwaNdtBDjE9bvBG_E&m=gDpQwIZuZSEOcOuIUV_9_jeGv5m-aqXgzBMzkuCM8wBeIKaKwaQUthJPFuNNZ9Dh&s=4yKmOpDzDXQKtaAvqAg7SgerPvw_i4yaPZHnS0nl7vE&e=
_______________________________________________
BESS mailing list
[e-mail]<mailto:[e-mail]>
https://www.ietf.org/mailman/listinfo/bess