[c-nsp] Cisco BFD and NSF

[Saku Ytti]

On (2006-08-01 10:07 +0300), Pekka Savola wrote:

> >However, the reasoning the Katz/Ward seem to offer, appears to

here should be 'appears _not_ to be' not 'appears to to be'. I hope
people had error correction on :).

> If Graceful Restart is enabled, running BFD on forwarding plane can 
> gain a certain degree of assurance of whether initiating graceful 
> restart would result in a black hole or not.  E.g., if you peer has a 
> power loss or crash, liveness messaging or lack thereof on control 
> plane might still indicate that GR is OK, while on forwarding plane we 
> get an explicit statement that GR is NOT OK as reachability to the 
> forwarding plane was lost.

If I understand what you're saying, I completely agree that we shouldn't
try to do graceful restart, unless peer requests it, to guarantee
that given down event can be handled by peer gracefully.
But I'm not sure, how this same problem couldn't be replicated in
forwarding-plane, given same functionality would be added there?

btw. if I read BFD spec correctly, there appears to be no 
native method for assisted graceful restart to occur, while you
can signal forwarding-plane reset, but I don't think it could
be used for assisted graceful restart.

> As you note, putting BFD solely on the forwarding plane (without any 
> keepalive on the control plane) could have potential for some nasty 
> cornercase issues when control and forwarding planes are out of sync 
> though..

And obvious solution to me, is to send as many messages between CP<->LC as
there is to send between LC<->network. But then I really have to 
wonder, why not send directly CP<->network.

I'm probably not seeing the whole picture, I've emailed Katz and asked
him about this.

Thanks,
-- 
  ++ytti