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

Re: [802.3ae_Serial] another question on pattertn defined in the current draft:




Tom Alexander wrote:

> Juergen and Piers,
>
> The decision to not send AIS upon loopback was made for the following reasons:
>
> <snip>

> 2. The constant pattern, originally 00-00 but now 00-FF, does not correspond
> to any SONET framing characters. Therefore, the far-end WIS will lose frame
> synchronization almost immediately when the local WIS has been placed into
> loopback. The effect will be similar to that of a failure of the WIS
> (loss of framing). This is precisely the logical effect upon the far-end
> WIS as a consequence of placing the local WIS into loopback; effectively,
> the transmit path of the local WIS has been disabled or shut off by putting
> it into loopback, and shutting off the WIS transmit path would have most
> likely raised an LOF alarm at the far-end under normal circumstances.

I agree with Tom (another good answer). I wanted to add that the error
hierarchy for SONET/SDH has loss of signal (LOS) has the highest alarm
followed by loss of frame (LOF) then Alarm Indication State (AIS).
Forcing the WIS to output a 00FF has the affect of generating an LOF -
that is: there is still a signal but the far end can not frame to it - which
causes an alarm.

Aside: the AIS condition is usually raised in response to an alarm
condition. AIS is analogous to the RF (remote fault) condition.
For 10GE, an RF is generated to the far end in response to an error
which was detected between the far-end and the local termination.
If the WIS were true to the SONET/SDH standard, upon receipt
of an AIS it would ask itself what it had done to generate this
condition. Whereas upon detection of an LOF, it would realize
the problem exists in the far-end.


--
Tim Warland     P.Eng.
Hardware Design Engineer  Broadband Products
High Performance Optical Component Solutions
Nortel Networks                (613)765-6634