[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: [PWE3] Inter-op is not necessary
Mark,
I agree that the working group must address the issues of
co-existence, but the process initiated during the earlier MPLS
fiasco of requiring each protocol proposal to address this in
the proposal itself is clearly a recipe for delay. Given the
well-known experiences of this earlier debacle, insisting on a
similar process in this case MUST be viewed as intentionally
obstructive, unless someone first establishes incontravertible
proof that no other approach will suffice.
One approach that has clearly worked in the past is for
a working group to take the task of defining "coexistence" on
as a separate task (in particular, a separate document). Even
if we decide that the other documents cannot proceed beyond a
certain point (ID, Proposed Standard, etc.) until the work on
"coexistence" is complete, this approach _at_least_ makes it
possible complete two separate protocol proposals without either
one - deliberately or inadvertently - being used to deny progress
to the other.
--
Eric
--> -----Original Message-----
--> From: pwe3-bounces at ietf.org [mailto:pwe3-bounces at ietf.org]
--> On Behalf Of Mark Townsley
--> Sent: Friday, November 18, 2005 9:20 AM
--> To: David McDysan
--> Cc: pwe3 at ietf.org
--> Subject: Re: [PWE3] Inter-op is not necessary
-->
--> David McDysan wrote:
-->
--> >IMO, we already have a "l3vpn 25457bis option" a
--> capability with single
--> >segment PWs. In other contexts (e.g., FR, ATM), this has
--> >interconnection case has been called a "back-to-back UNI."
--> Since there
--> >is no alternate path in such back-to-back SS-PWs, this is not an
--> >acceptable solution from a service provider point of view.
--> This is why
--> >service providers and others have created the requirements in
--> >http://ietf.org/internet-drafts/draft-ietf-pwe3-ms-pw-requi
--> rements-01.txt.
--> >
--> >For these reasons, I believe that "co-existence" as
--> defined in this
--> >note is a not a MS-PW requirement.
--> >
--> >
-->
--> "Coexistance" is simply a word I made up to mean that if
--> the WG decides to advance two proposals, it must recognize
--> that both exist and what it is going to do about it. It
--> isn't an MS or Provider requirement, it is a Working Group
--> requirement. The WG MUST adequately address all of the
--> complexities of advancing more than one solution if it
--> decides to do so.
-->
--> Dealing with this reality of coexistance will require the
--> working group to define mechanisms based on MS-PW
--> requirements. You have stated that interworking with SS-PWs
--> is one such requirement. I agree, so in our minds if two
--> proposals advance, then the WG will have to define
--> interworking mechanisms between SS-LDP and MS-LDP as well
--> as SS-LDP and MS-RSVP. Interworking between MS-LDP and
--> MS-RSVP will depend on whether interoperating providers are
--> willing to either (1) support both signalling types in
--> their network, or (2) agree to peer only with those
--> providers who choose their preferred protocol. You can't
--> have both without interworking between the two.
-->
--> - Mark
-->
--> >This
--> >
--> >Dave
--> >
--> >
--> >
--> >>-----Original Message-----
--> >>From: pwe3-bounces at ietf.org
--> [mailto:pwe3-bounces at ietf.org] On Behalf
--> >>Of Mark Townsley
--> >>Sent: Thursday, November 10, 2005 10:12 PM
--> >>To: Andy Li
--> >>Cc: pwe3 at ietf.org
--> >>Subject: Re: [PWE3] Inter-op is not necessary
--> >>
--> >>
--> >>To be clear, I said that "coexistance" was a necessity to be
--> >>documented by the WG. Whether coexistance equates to interworking
--> >>depends on the MS requirements.
--> >>
--> >>If the WG decides that providers would rather operate in
--> a "10a" model
--> >>(to use l3vpn terminology) rather than switch directly
--> from one PW to
--> >>another, coexistance is possible without interworking
--> between LDP and
--> >>RSVP. If this is true, however, it does beg the question
--> as to whether
--> >>interprovider MS is truly necessary at all.
--> >>
--> >>
--> >>- Mark
--> >>
--> >>Andy Li wrote:
--> >>
--> >>
--> >>
--> >>>Hi,
--> >>>
--> >>> I don't think it is necessary to have both signaling
--> mechanism for
--> >>>MS-PW inter-op with each other. If two providers agree
--> to peer with
--> >>>each other to provide PW services, it will be a bonus if
--> >>>
--> >>>
--> >>they two use
--> >>
--> >>
--> >>>the same mechanism to set up PW circuits. If not, the business
--> >>>interests will encourage them to stitch the PW circuits by using
--> >>>back2back logical/virtual interface(s) or even a jumper
--> >>>
--> >>>
--> >>fiber between
--> >>
--> >>
--> >>>ODF.
--> >>>
--> >>> Therefore, out of my own opinion, the inter-op requirement
--> >>>
--> >>>
--> >>should be removed.
--> >>
--> >>
--> >>>Best
--> >>>Andy Li
--> >>>
--> >>>_______________________________________________
--> >>>pwe3 mailing list
--> >>>pwe3 at ietf.org
--> >>>https://www1.ietf.org/mailman/listinfo/pwe3
--> >>>
--> >>>
--> >>>
--> >>>
--> >>>
--> >>_______________________________________________
--> >>pwe3 mailing list
--> >>pwe3 at ietf.org
--> >>https://www1.ietf.org/mailman/listinfo/pwe3
--> >>
--> >>
--> >>
--> >
--> >
--> >
-->
--> _______________________________________________
--> pwe3 mailing list
--> pwe3 at ietf.org
--> https://www1.ietf.org/mailman/listinfo/pwe3
-->
_______________________________________________
pwe3 mailing list
pwe3 at ietf.org
https://www1.ietf.org/mailman/listinfo/pwe3