Re: [mpls] draft on signal degradation

["Sprecher, Nurit (NSN - IL/Hod HaSharon)" <[e-mail]>]

Hi Himanshu,

Thanks for the information.

Just to make it clear:

*         Orckit and KDDI came out with a draft on SD, which is detected
in the link level and notified to the edges of the paths that are
affected from this event (exactly as "Alarm Reporting", AKA AIS). IMO,
the draft still needs to be worked on... to address questions that were
on the second slide....and the good question came from George. 

o    There were some discussions on the mailing list, with some
confusion between SD detected on the link/phy layer and can be detected
in any condition (w or w/o traffic on the path) and SD on the path level
cause because of congestion...- this can be detected only when there is
traffic with similar pattern as the protected path. Many would even not
consider this as SD..... 

*         We worked on the survivability framework and were asked by the
chairs to show if there is any "impact" of SD discussed in the list on
the framework - to this the answer is No! there is no impact. The
framework already assumes multiple events as triggers to the recovery
action (what is called as "elements of control). It was also indicated
in the slides what are not considered to be the role of the
survivability framework....

*         From linear protection point of view, some questions were
raised and I have put them on the second slide. The simplest way to
handle this if SD indication is an absolute indication (exactly as SF),
but again there were some discussions and arguments on the mailing list
and I tried to summarize these and put questions on the second slide. I
think it is important we agree on the resolution for these questions. 

Best regards,

Nurit

 

 

-----Original Message-----
From: ext Shah, Himanshu [mailto:[e-mail]] 
Sent: Monday, November 08, 2010 11:46 AM
To: Sprecher, Nurit (NSN - IL/Hod HaSharon); [e-mail];
[e-mail]
Cc: Linda Dunbar
Subject: draft on signal degradation

 

Hi Nurit -

About 2 years ago, Raj Jain (Linda Dunbar and I as co-authors)

presented similar work proposal to IEEE, perhaps you attended that

meeting. 

 

We could not convince IEEE to consider Signal Degradation as

a criteria for connectivity fault. Same concerns - right thresholds,

durations, hold downs, ripples how far up and out, etc etc.

 

FYI - as another data point.

 

/himanshu