[j-nsp] Optimal BFD settings for BGP-signaled VPLS?

[Clarke Morledge]

On Sun, 16 Jan 2011, Phil Bedard wrote:

> If BGP stability is the main goal, do not use BFD with your BGP sessions.
> Are you using site multi-homing with the connected CE devices or are they
> all single-homed?  I don't know your topology but there may be some
> instances where you would want to run BFD for BGP notification with
> multi-homing.
>
> What hardware are you using?  We are using 300x3 everywhere and while we
> have seen some isolated false positives, things have been relatively
> stable.


Phil,

I am using MX gear, and the ppmd offload to the PFEs for BFD does appear 
to help a lot.  Unfortunately, we have uncovered a few bugs using BFD like 
this in our deployment over the past year, but the more recent 10.2 code 
revs appear to fix many of these issues.

Good point about VPLS multi-homing.  I should have mentioned that we are 
using it.  In some cases we have it running on our core routers. 
Typically, since a core router is acting as a BGP route reflector, the 
core router is also a VPLS multi-home primary or backup site -- trying to 
take advantage of the redundancy.  Also, we are looking at multi-homing on 
some of our distribution routers.  Knowing that about my topology, any 
further thoughts on BFD for BGP?

I am not sure I totally grasp what is going on during a failover (manual 
or otherwise) of VPLS multihoming, so I'll take whatever insight you might 
have.

Thanks.

Clarke Morledge
College of William and Mary
Information Technology - Network Engineering
Jones Hall (Room 18)
Williamsburg VA 23187