Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: Link Status thoughts




Hi Tripathi,

I agree with you that remote fault and local fault will not be sent 
simultaneously as far as for Layer-1 fault indication mechanism.

The reason I have designed LSS to send local fault (break link in my 
term) and remote fault simultaneously are to mix up the channel reset 
function triggered by STA, as was described in  
At 23:04 00/11/02 +0900, Osamu ISHIDA wrote:
http://www.ieee802.org/3/10G_study/email/msg03694.html
(My presentation slide at Tampa is now uploaded at)
http://www.ieee802.org/3/ae/public/nov00/ishida_1_1100.pdf

Please let me consider the case when RF mechanism is implemented in RS.
RS generates remote fault on the transmit path when RS receives local 
fault (break link in my term) on the receive path.  This is the RF 
mechanism.  Independently the local STA might triggers channel reset 
and hence RS have to generate local fault (break link in my term) on 
the transmit path simultaneously.

If BL (Local Fault in your term) overwrites RF (I mean sending BL alone 
when BL and RF should be generated), we will see a deadlock when 
both Local Device and Link Partner trigger channel reset simultaneously.  
In the link status (RF&BL) mechanism of LSS, BL should be acknowledged 
by RF, and hence the link never comes up.

In a bottom line, sending BL/RF simultaneously is required if we choose 
the LSS link status mechanism with a single Boolean equation;
   link_status = !remote_fault * !RF_detect
If we choose Shimon's link status mechanism using state machine, 
we do not need to send BL/RF simultaneously.  It can work even in the 
above BL-BL condition by using state transition every 10-100ms depending 
on how far we support the link distance.

Best Regards,
Osamu


At 10:06 00/11/02 -0800, Devendra Tripathi wrote:
http://www.ieee802.org/3/10G_study/email/msg03697.html
> Just one small comment on following part. It may not be possible (or worth 
> figuring out) to distinguish between remote fault and local fault if they 
> happen in the same direction. 
> To me local fault is if I can not transmit and remote fault is if I can not 
> receive. If we could go  beyond this (like if I knew that receiving
> transceiver is not functioning), it could be defined as local fault but it 
> may be going into more of detail OAM&P.

At 22:38 00/11/01 -0800, Rich Taborek wrote:
http://www.ieee802.org/3/10G_study/email/msg03692.html
> >The most complex scenario involving Remote Fault and Local Fault that I
> >can envision is one where both Remote Fault and Local Fault conditions
> >exist for the same link, in the same direction. For the sake of
> >completeness, the Remote Fault and Local Fault transport protocol should
> >allow for the indication of all combinations of these two conditions.
> >Note that this scenario is not covered in the latest RF/BL proposal from
> >Shimon Muller.

At 12:55 00/11/01 -0700, pat_thaler@xxxxxxxxxxx wrote:
http://www.ieee802.org/3/10G_study/email/msg03676.html

At 20:46 00/10/31 -0500, Ben Brown wrote:
http://www.ieee802.org/3/10G_study/email/msg03660.html

At 18:38 00/10/31 -0700, pat_thaler@xxxxxxxxxxx wrote:
http://www.ieee802.org/3/10G_study/email/msg03659.html

At 03:43 00/10/27 -0400, Ben Brown wrote:
http://www.ieee802.org/3/10G_study/email/msg03608.html

---------------------------------------
Osamu Ishida,ishida@xxxxxxxxxxxxxxxxxxx
NTT Network Innovation Laboratories
TEL:+81-468-59-3263 FAX:+81-468-55-1282
---------------------------------------