RE: I-D ACTION:draft-malis-pwe3-cell-transport-00.txt

["Proch, Daniel" <[e-mail]>]

Neil, Andy,

John and I put together some text with suggested modes to provide OAM for
port mode.  As you note, one suggestion was Defect Indication using Loss of
Cell Delineation (LCD).  Another was using Physical Layer OAM cells to
provide an additional F3 transmission path layer above the F3 transmission
path layer provided by SONET/SDH, or PDH interfaces.  This text is included
in an MPLSForum doc regarding ATM port mode, but not in
draft-malis-pwe3-cell-transport-00.txt.

dan

-----Original Message-----
From: [e-mail] [mailto:[e-mail]]
Sent: Thursday, October 24, 2002 11:21 AM
To: [e-mail]; [e-mail];
[e-mail]
Cc: [e-mail]
Subject: RE: [PWE3] I-D ACTION:draft-malis-pwe3-cell-transport-00.txt


Andy/Shahram,

Been a bit busy to check mail much recently....however, on the specific
issue of port mode:

The key thing to remember when dealing with any connection-oriented pkt-sw
network (and this includes p2p LSPs) is that we cannot simply consider the
behaviour of the pkts independently of their parent trail.....both have to
be considered together.  Further, it is the defect management of the trail
that should be our primary concern.  However, since we don't know the
VPC/VCC profile in 'port mode' (ie the specific ATM trails supported) then
we cannot directly apply the required FDI/AIS signal to the affected trails
to suppress their alarms.  The last time I discussed this with John
Rutemiller, John suggested that perhaps the cleanest architectural kludge
would be to force a loss of cell delineation across the whole cell-stream.
A downstreamn device that does know the client VPC/VCC profile supported
would then inject the required FDI/AIS signal into the affected ATM trails. 

John, if you are reading this perhaps you may like to add any further
observations as its while since we last discussed this.

regards, Neil

> -----Original Message-----
> From: Shahram Davari [mailto:[e-mail]]
> Sent: 22 October 2002 20:27
> To: 'Andrew G. Malis'
> Cc: [e-mail]
> Subject: RE: [PWE3] I-D ACTION:draft-malis-pwe3-cell-transport-00.txt
> 
> 
> Andy,
> 
> Thanks for your explanation. I think the most important and 
> possibly difficult part of any 
> carrier class technology is its management and defect 
> handling.  Are you proposing to
> advance this draft to propose standard without descend defect 
> handling? or are
> you proposing the start of a discussion on defect handling 
> for port mode, until
> a consensus is formed before advancing it to proposed standard?
> 
> Thanks,
> -Shahram
> 
> 
> 
> > -----Original Message-----
> > From: Andrew G. Malis [mailto:[e-mail]]
> > Sent: Tuesday, October 22, 2002 1:12 PM
> > To: Shahram Davari
> > Cc: 'Andrew G. Malis'; [e-mail]
> > Subject: RE: [PWE3] I-D 
> ACTION:draft-malis-pwe3-cell-transport-00.txt
> > 
> > 
> > Shahram,
> > 
> > 1. Luca and others are working on it now.  They expect to get 
> > it uploaded 
> > in time for the draft cutoff.  It will reflect the SG13 and 
> > PWE3 compromise 
> > on ATM encapsulation that was discussed at the Yokohama meeting.
> > 
> > 2. This IS the whole text for "port mode" - appendix A from 
> > draft-martini-atm-encap-mpls-01.txt.  I didn't inlcude the "defect 
> > handling" section because,as I said earlier, there hasn't been any 
> > consensus (especially among the implementers :-) on it.
> > 
> > Cheers,
> > Andy
> > 
> > --------
> > 
> > At 10/22/2002 09:43 AM -0700, Shahram Davari wrote:
> > >Andy,
> > >
> > >1) Where is the ID that you refer to: 
> > draft-ietf-pwe3-atm-encap-00.txt?
> > >2) Do you want to move the port mode to your draft? or 
> another draft?
> > >    if you want to move it to your draft, why don't you 
> put the whole
> > >    text for port mode in your draft?
> > >
> > >Yours,
> > >-Shahram
> > >
> > > > -----Original Message-----
> > > > From: Andrew G. Malis [mailto:[e-mail]]
> > > > Sent: Tuesday, October 22, 2002 12:26 PM
> > > > To: Shahram Davari
> > > > Cc: [e-mail]
> > > > Subject: RE: [PWE3] I-D 
> > ACTION:draft-malis-pwe3-cell-transport-00.txt
> > > >
> > > >
> > > > Shahram,
> > > >
> > > > The point of this draft is that the transparent cell
> > > > transport appendix
> > > > (aka port mode) is being removed from the combined ATM
> > > > encapsulation draft,
> > > > so this is to capture that text in a separate draft, which is
> > > > especially
> > > > important since there are multiple interoperating 
> implementations.
> > > >
> > > > We've previously discussed the OAM implications at length,
> > > > which resulted
> > > > in text that was also in the Martini ATM encapsulation
> > > > appendix.  However,
> > > > Neil sent an email over the summer saying that he was still
> > > > unhappy with
> > > > that OAM text, so for now there really isn't any consensus
> > > > OAM text to include.
> > > >
> > > > If you're willing to propose some text, I'll be happy to
> > > > include it once we
> > > > get consensus on the list.
> > > >
> > > > Thanks,
> > > > Andy
> > > >
> > > > ---------
> > > >
> > > > At 10/22/2002 06:47 AM -0700, Shahram Davari wrote:
> > > > >Hi,
> > > > >
> > > > >The OAM implications of this mode is completely different
> > > > from N:1 cell mode.
> > > > >For example you can't generate AIS on either PE side. You
> > > > can't also do a
> > > > >CE-PE ATM connectivity test. So simply referring to
> > > > [ATM-ENCAPS] draft
> > > > >(which BTW should be changed to reference 3) is not adequate
> > > > to describe
> > > > >this mode.
> > > > >
> > > > >A solution for CE-PE connectivity test in ATM layer and
> > > > downstream alarm
> > > > >suppression
> > > > >is needed.
> > > > >
> > > > >Thanks,
> > > > >-Shahram
> > 
> _______________________________________________
> pwe3 mailing list
> [e-mail]
> https://www1.ietf.org/mailman/listinfo/pwe3
> 
_______________________________________________
pwe3 mailing list
[e-mail]
https://www1.ietf.org/mailman/listinfo/pwe3