The MPLS WG Archive

[Vishal Sharma <Vishal.Sharma@tellabs.com>]

Dave,

On Friday, April 28, 2000 2:25 PM, dallan@nortelnetworks.com [SMTP:dallan@nortelnetworks.com] 
wrote:

> re: "lower" vs. "other" layers
>
> 	The issue to me is the wording. There are two thoughts we seem to be
> discussing:
>
> 	1) Not taking premature recovery action in the presence of lower
> layer restoration mechanisms.
>
> 	2) Considering all sources of knowledge of failure (including layer
> violations).

You've hit the nail on the head!

> 	I'm talking about the first, when I say "lower", the second is where
> you mean "other" and where we agree problems reside.

Yes, you've put it quite succinctly.

> 	I think the first may be a bigger problem area than we think.
> Because MPLS is defined over multiple media, presumably an LSP can span
> multiple media, and therefore WHERE a defect occurs matters as far as how
> fast we can take corrective action.
>

Ok, if I read you right, what you'd like to do is to have the framework acknowledge both 1 and 2 
above,
and perhaps mention them separately. (Realizing of course that there are still issues
with 2), greater than those that there might be with 1). I think doing this would make
things clearer, while at the same time leaving room for different implementations.

-Vishal