Re: WG Adoption request for draft-mirsky-bfd-mpls-demand

[Greg Mirsky <[e-mail]>]

Hi Carlos,
thank you for the continued discussion. Please find my answers in-line
tagged GIM>>.

Regards,
Greg

On Sat, Oct 27, 2018 at 8:45 AM Carlos Pignataro (cpignata) <
[e-mail]> wrote:

> Hi Xiao,
>
> Thanks much for the quick response!
>
> Please find my follow ups:
>
> 1. Sorry if I was not clear. Yes, RFC 5880 lists exemplary possible
> mechanisms to glean continued connectivity but requires the use of one. My
> question is: which mechanisms does this draft propose? Could it please be
> spelled out?
>
GIM>> I have to point that:

   - BFD for Multipoint Networks uses the Demand mode of BFD with no such
   mechanism;
   - BFD for Multipoint Networks with Active Tail uses Poll sequence
   (possibly combination of multicast Poll and unicast Poll) to verify
   availability of the multicast path. But according to your interpretation of
   RFC 5880, this is not one of mechanisms to be used.

Now I have a question to you Do you believe that these two BFD
specifications for multicast and multipoint networks are insufifcient or
impaired because of the way the Demand mode being used?

>
> 2. Again, apologies if I was not clean enough. I understand what the
> abstract says, but the quoted text says that live was needs to be checked
> for a node (egress LER) instead of the forwarding path. Does that mean ~
> “of the forwarding path towards/upto the node”?
>
GIM>> Thank you for pointing to this issue. In the next version will
clarify that the Poll sequence allows the ingress LER to verify
availability of the LSP and the liveness of the egress LER.

>
> Thanks for your consideration.
>
> Thumb typed by Carlos Pignataro.
> Excuze typofraphicak errows
>
> On Oct 27, 2018, at 01:07, "[e-mail]" <[e-mail]>
> wrote:
>
> Hi Carlos,
>
>
> My answers to your two questions are as follow:
>
>    1.
>
>    In section 6.6 of RFC5880, just after the text you quoted, it says
>    "One possible mechanism is the receipt of traffic from the remote system;
>    another is the use of the Echo function." So I'm not sure what's your real
>    concern.
>    2.
>
>    In the abstract of this draft it says "This document describes
>    procedures for using Bidirectional Forwarding Detection (BFD) in Demand
>    mode to detect data plane failures in Multiprotocol Label Switching (MPLS)
>    point-to-point Label Switched Paths." If you don't like the expression form
>    of the text quoted by you, pls feel free to propose some text.
>
>
> Best Regards,
>
> Xiao Min
> 原始邮件
> *发件人：*CarlosPignataro(cpignata) <[e-mail]>
> *收件人：*肖敏10093570;
> *抄送人：*Jeff Haas <[e-mail]>;[e-mail] <[e-mail]>;
> *日 期 ：*2018年10月26日 12:04
> *主 题 ：**Re: WG Adoption request for draft-mirsky-bfd-mpls-demand*
> Xiao,
> Scanning through the draft, two questions:
>
> 1. What is the underlying mechanism to check liveness such that Demand can
> be used?
>
> https://tools.ietf.org/html/rfc5880#section-6.6
>
>    Demand mode requires that some other mechanism is used to imply
>    continuing connectivity between the two systems.  The mechanism used
>
>
> 2. Is this draft testing liveness of a path or a node?
>
> https://tools.ietf.org/html/draft-mirsky-bfd-mpls-demand-03#section-3
>
>    In this state BFD peers MAY remain as long as the egress LER is in Up
>    state.  The ingress LER MAY check liveness of the egress LER by
>    setting the Poll flag.  The egress LER will respond by transmitting
>
>
> Thanks,
>
> — Carlos Pignataro
>
> On Oct 19, 2018, at 9:59 PM, [e-mail] wrote:
>
> I support WG adoption of this draft. Use of the demand mode for p2p LSP
> monitoring is feasible and required.
>
>
> Best Regards,
>
> Xiao Min
>
>
>
> *发件人：*JeffreyHaas <[e-mail]>
> *收件人：*[e-mail] <[e-mail]>;
> *日 期 ：*2018年10月18日 06:24
> *主 题 ：**WG Adoption request for draft-mirsky-bfd-mpls-demand*
> Working Group,
>
> The BFD chairs have received an adoption request for
> "BFD in Demand Mode over Point-to-Point MPLS LSP"
> (draft-mirsky-bfd-mpls-demand).
>
> https://datatracker.ietf.org/doc/draft-mirsky-bfd-mpls-demand/
>
> The adoption call will end on the Friday after IETF 103, November 9.
>
> Note that there is are existing IPR statements on this draft:
> https://datatracker.ietf.org/ipr/3301/
> https://datatracker.ietf.org/ipr/3104/
>
> Please indicate to the mailing list whether you support adoption of this
> draft.
>
> -- Jeff & Reshad
>
>
>