Re: [802.3ae] Fault handling
Chuck,
The method for mapping Ethernet frames into synchronous payloads that has
been adopted by T1X1 is not used by 802.3 for any functions. As the
"linear" mapping function of G.gfp was previously presented to and rejected
by P802.3ae, I doubt that it will be used. There is also some question
about the ability of G.gfp to support "jumbo frames". The "ring" mapping
function of G.gfp is very similar to what is being attempted in 802.17.
Thank you,
Roy Bynum
At 05:31 PM 6/25/01 -0700, Tricci So wrote:
>Hi all,
>
>Please forgive my ignorance since I haven't read the 10GbE draft.
>What is the OAM strategy for 10GbE? Is it going to adopt any function
>from the GFC standard?
>
>Thank in advance to your advice?
>
>Tricci
>
>Chuck Harrison wrote:
>
> > This is just a general observation about fault handling (e.g
> > LF, RF, AN fault fields, etc.)
> >
> > In some situations management intelligence is very limited
> > or absent. In particular, makers of layer-1 media converters
> > have struggled for years to get these devices to behave
> > in a logical fashion on faulted links and to provide
> > some help with fault isolation given only a few LEDs for
> > status indication.
> >
> > I hope that the experience of the IMC's, Allied Telesyn's,
> > etc. has been incorporated in the specification of the fault
> > handling features in 10GbE. Not everyone will be running
> > carrier-grade management systems, but everyone will see a
> > link fail sooner or later!
> >
> > Cheers,
> > Chuck Harrison
> > Far Field Associates, LLC
> > +1 360 863 8340 (voice) GMT-0700