[c-nsp] Reliable static routes and 12.2S

[Andre Beck]

On Tue, Dec 21, 2004 at 05:04:51PM +0100, Daniel Roesen wrote:
> On Tue, Dec 21, 2004 at 04:19:05PM +0100, Andre Beck wrote:
> > Well, supervising reachability on top of VPNs and PPPoE clouds is
> > the real intention, but in practice, every simple Ethernet broadcast
> > domain with a gateway in there is able to blackhole you. I'd rather
> > like an alternate style of L2 reachability supervision on top of
> > Ethernet, a special non-ARP point to point over Ethernet that
> > constantly watches the availability of the peer MAC and would let
> > the interface go line protocol down if it cannot be established. But
> > there is no such thing that I knew of,
> 
> You're looking for BFD (bi-directional forwarding detection). As far as
> I know not yet implemented by Cisco, and in JunOS only used for IGP
> yet (IS-IS, not sure about OSPF). Provides end2end failure detection
> down to milliseconds time-to-detect (I've seen 50ms - this is SONET/SDH
> ring wrap time).

Thanks for the pointer. That reads almost exactly like what I thought
up some days ago during a boring meeting, just way more orthogonal
and medium independend. Hopefully this will end up in IOS soon, as the
draft is headed by Cisco as well as Juniper. Ideally in Ethernet
subinterface mode like

int fa0/0.33
 encapsulation bfd 33
 ip unnumbered fa1/0
 keep 4

with 33 beeing the discriminator for that PtP link. Of course you would
normally want that to be numbered with a transit net for IGP/EGP use,
but in BFD context it SHOULD support unnumbered IMO - for the potentially
vast amount of RBE customer fanouts in xDSL world.

Thanks,
Andre.
-- 
                  The _S_anta _C_laus _O_peration
  or "how to turn a complete illusion into a neverending money source"

-> Andre Beck    +++ ABP-RIPE +++    IBH Prof. Dr. Horn GmbH, Dresden <-