Re: [sfc] WG LC on draft-ietf-sfc-multi-layer-oam

[Gyan Mishra <[e-mail]>]

Hi Joel & WG

I strongly support publication of this draft.

I believe this would be a valuable specification for operators deploying
SFC.

>From the  WGLC early this year a draft was published by Med - Thank you!  -
to clarify RFC 8300 ambiguity of OAM O bit setting in NSH, and this draft
has been updated with normative reference to draft below.

>From WGLC early this year to now from version 18 to 19 update is related
only to the OAM in NSH O bit setting and resolution of that issue.

SFC OAM Packet
https://datatracker.ietf.org/doc/html/draft-ietf-sfc-oam-packet-01

Troubleshooting SFC is a complex tax for operators and having
additional OAM capabilities
that can provide value to operators in E2E SFC troubleshooting is a major
gain for operators.



RFC 8924 defines the base specification for SFC OAM, requirements analysis
and generically existing OAM mechanisms used at various layers  and how
they can apply to SFC defined in section 7.



This draft provides a comprehensive SFC OAM solution and takes the base SFC
OAM RFC 8924 and existing network layer mechanisms and applies them to SFC
OAM localized SFC fault isolation with a  new Active OAM header,
Authenticated Echo Request/Reply message and Source TLV.



The new functionality in this draft is defining a new procedure  for Active
OAM message on RSP in NSH and OAM command and/or data in NSH and O bit in
NSH must be set per SFC OAM Packet draft above.



Section 5  talks about the issue related to additional IP/UDP headers in an
IPv6 network adds noticeable overhead and this draft defines a new active
OAM header  to demultiplex Active OAM protocols on an SFC.



Section 6 defines a new Active OAM based Authenticated  Echo Request/Reply
message for SFC that addresses additional requirements, fate sharing,
monitoring of continuity between SFPs, RDI by ingress to egress,
connectivity verification, fault localization and tracing to discover RSP
and finally on-demand FM with response back to initiator.



This draft also provides OAM integrity check with authentication of
request/reply message in conjunction with use of source TLV to prevent DDOS
attack vector with SFC OAM.



The critical new functionality provided for operators with Active OAM is
the honed in focus on troubleshooting continuity of an SFP, trace an SFP ,
consistency verification of SFP and fault isolation and localizing of a
failure within an SFP as well as valuable SFF record TLV, SFF information
TLV/Sub-TLV  for multiple SFs as hops of SFP or multiple SFs for load
balancing using SFP consistency verification procedures.


Kind Regards


Gyan

On Thu, Jun 2, 2022 at 2:51 PM Joel Halpern <[e-mail]> wrote:

> Technically, the draft is still in last call from quite some time ago.
>
> For clarity, and as there have been clarifications in relate
> terminology, the chairs have decided to consider taht we are restarting
> the last call today.
>
> This Working Group Last call will run for 2 weeks (and a day) until CoB
> on June 17, 2022.
>
> Please respond positively or negatively to this call.  Note that if we
> do not get enough responses we will likely be unable to advance the
> document before the WG closes.
>
> When responding, please provide clear motivation either for or against
> publication as an RFC.  Substantive comments are MUCH more helpful than
> "yes" or "no".
>
>
> Thank you,
>
> Joel and Jim
>
> _______________________________________________
> sfc mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/sfc
>
-- 

<http://www.verizon.com/>

*Gyan Mishra*

*Network Solutions A**rchitect *

*Email [e-mail] <[e-mail]>*



*M 301 502-1347*