[c-nsp] BFD in XR 3.9.1

[Peter Rathlev]

On Wed, 2010-08-25 at 01:02 -0500, Richard A Steenbergen wrote:
> BFD is an IP based protocol, it's completely ignorant of L2 multipath 
> and will almost always get hashed over a single link arbitrarily.

Cisco may view it as only L3 relevant, but from RFC 5882 section 2:

> Its sole purpose is to verify connectivity between a pair of systems,
> for a particular data protocol across a path (which may be of any
> technology, length, or OSI layer).

So theoretically BFD could be used for fast detection of loss of a L2
link.

I would actually very much like to have something like BFD for L2. When
constructing EoMPLS paths through the network failover (seen from
between two "CE" devices) can be oh-so-slow, with RSTP (~6 sec) and UDLD
(~5 sec) being the quickest to discover loss of connectivity.

Or did I overlook something?

-- 
Peter