[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: draft-bonica-tunneltrace-02



Title: RE: draft-bonica-tunneltrace-02

I just want to clarify that "Requirements for OAM in MPLS Networks"
draft-harrison-mpls-oam-req-01.txt is live and kicking (exp. May 2002).

Draft-bonica focuses on tunnel tracing only, i.e. only one aspect of MPLS management.
Draft-harrison-mpls-oam-req-01.txt covers all aspects of MPLS management.
Therefore with regards to MPLS, draft-harrison-mpls-oam-req-01 is more "generic" than draft-bonica.

Regards,

Mina Azad

> -----Original Message-----
> From: Gibson, Mark [mailto:mgibson@orchestream.com]
> Sent: Thursday, February 28, 2002 10:32 AM
> To: ccamp@ops.ietf.org
> Subject: RE: draft-bonica-tunneltrace-02
>
>
> As an impartial observer, there seems to be an obvious way
> forward for this
> discussion.
>
> In the first instance, draft-bonica-tunneltrace-02 can go
> through the usual
> process in the wg meeting in Minneapolis and, subject to
> rough consensus,
> become adopted as a solution for the cases that it solves.  I
> believe even
> the proponents of the mplsoam scheme concede that the tunnel
> trace solution
> has merit in certain circumstances.  [a point made by Dave
> Allan as I type]
> Lets document that solution space and, if technically
> sufficient, advance
> this draft such that tunnel trace can meet all the points in
> that space it
> can.
>
> In the second instance, lets define the specifics of the
> remainder of the
> OAM problem space and work on some solutions that solve these
> problems.
> From memory, there were a number of individuals in the OAM
> BoF who, though a
> minority were a large minority who believed something more
> substantive than
> tunnel trace was needed, including a number of operators. 
> Presumably these
> guys know what their requirements are.  As has been stated
> these are to a
> large extent documented in the (sadly too new to get at for
> free) documents
> Y.1711/Y.1710 but were in the now expired though surely google-able
> draft-harrison...
>
> And finally, lets stop this "not invented here" mentality. 
> I'm not sure
> that dismissing ITU documents because that's what they are
> does the IETF any
> favours at all.  (c.f. the diatribe against the WAP
> representative at the
> Adelaide open plenary).
>
> Mark
>
>  -----Original Message-----
> From:         Eric Rosen [mailto:erosen@cisco.com]
> Sent: 28 February 2002 14:59
> To:   Shahram Davari
> Cc:   'Randy Bush'; Cuevas, Enrique G, ALASO; ccamp@ops.ietf.org
> Subject:      Re: draft-bonica-tunneltrace-02
>
>
> Shahram> It  has serious  security, complexity,  backward 
> compatibility and
> Shahram> layer violation issues.
>
> Tom> Can you elaborate on what you think these are?
>
> Shahram> Please refer to previous emails by me and David
> Allan. Most of them
> Shahram> are listed there. 
>
> I'm sorry, but  as far as I  can tell, those previous mails 
> simply say that
> (a)  the  proposed  solution doesn't  do  some  things  that
> you  think  are
> valuable,  and (b)  the proposed  solution doesn't  fit well 
> into  some ITU
> architecture.
>
> The  second  of  these  points  is  completely  irrelevant.  
> The  first  is
> irrelevant too, unless there is a reasonable alternative
> proposed which does
> more,  or if the  current proposal  doe so  little that  SPs
> don't  think it
> worthwhile.   The MPLS OAM  stuff you've  been pushing  is
> not  a reasonable
> alternative of  this sort  because (a)  it is MPLS-specific, 
> and (b)  it is
> already crystal  clear that it will not  be accepted in the 
> IETF.  And it's
> pretty clear  that a number of SPs  do think that what  the
> current proposal
> does is worthwhile.
>
> If  you can actually  cite specific  security issues  with
> the  proposal, it
> would be valuable to know about them.
>
> Suggestions for reducing complexity would  also be valuable,
> if you have any
> specific suggestions in that area.
>
> I don't understand what the backwards compatibility issue is,
> as there is no
> previous version to be compatible with.
>
> If you think there  are layer violation issues, then what you
>  need to do is
> exhibit the particular set of  specific problems that will
> arise in practice
> as a result of those violations.   If you cannot do this
> without referencing
> some arcane ITU  architecture document, then the natural 
> conclusion is that
> the  problem  is with  that  architecture's  layering  model,
> not  with  the
> proposal.
>
>
>
>
>
>