Best practices for redundant iBGP/eBGP route distribution? [bird 2.0.7]

[Ondrej Zajicek]

On Mon, Dec 16, 2019 at 03:40:18PM +0100, Nico Schottelius wrote:
> 
> Ondrej Zajicek <santiago at crfreenet.org> writes:
> 
> > On Mon, Dec 16, 2019 at 01:43:09AM +0100, Nico Schottelius wrote:
> >>
> >> Follow up 2:
> >>
> >> it seems that bird1.6 behaves differently to bird 2.0.7:
> >>
> >> bird> show route all for 2a0a:e5c1:111:111:6aa6:5bc:535a:8e21
> >> 2a0a:e5c1:100::/40 via 2a0a:e5c0:1:8::6 on bond0.8 [router2_place6_ungleich_ch_v6 2019-12-03] * (100) [i]
> >>         Type: BGP unicast univ
> >>         BGP.origin: IGP
> >>         BGP.as_path:
> >>         BGP.next_hop: 2a0a:e5c0:2:2:0:84ff:fe41:f24d
> >>         BGP.local_pref: 500
> >> bird>
> >>
> >> (this is from the 2nd router pair, still running bird 1.6)
> >>
> >> Is it possible that the nexthop resolution algorithm changed in bird2 vs
> >> bird1?
> >
> > Yes, in BIRD 1 direct mode, there was a fallback that uses IP address
> > of BGP peer as gateway if BGP NEXT_HOP failed to resolve.
> >
> > We removed this fallback in BIRD 2.
> 
> Ha, that explains a lot! I am still puzzled though, because the
> description of bird2 on the topic of gateway direct seems to imply that
> this is still the case.

That is mistake in docs.


> Is there any way to restore the 1.6 behaviour?

No, it just hides misconfigurations.

If you use 'next hop self' then there should not be difference between
BGP NEXT_HOP and peer IP address and it would work even in cases where
both IPv4 and IPv6 routes are propagated over one BGP session.


-- 
Elen sila lumenn' omentielvo

Ondrej 'Santiago' Zajicek (email: santiago at crfreenet.org)
OpenPGP encrypted e-mails preferred (KeyID 0x11DEADC3, wwwkeys.pgp.net)
"To err is human -- to blame it on a computer is even more so."