[190311] in North American Network Operators' Group
Re: 1GE L3 aggregation
daemon@ATHENA.MIT.EDU (Owen DeLong)
Thu Jun 23 02:22:26 2016
X-Original-To: nanog@nanog.org
From: Owen DeLong <owen@delong.com>
In-Reply-To: <f604725d-c9e4-fdbe-1d2c-9246d12bbcff@seacom.mu>
Date: Wed, 22 Jun 2016 23:22:19 -0700
To: Mark Tinka <mark.tinka@seacom.mu>
Cc: nanog@nanog.org
Errors-To: nanog-bounces@nanog.org
> On Jun 22, 2016, at 23:17 , Mark Tinka <mark.tinka@seacom.mu> wrote:
>=20
>=20
>=20
> On 23/Jun/16 08:07, Owen DeLong wrote:
>=20
>> If it=E2=80=99s 100% for redundancy, why not just ECMP defaults and =
not take a full table?
>=20
> Well, firstly, ring length may be different on either end. So you =
can't
> always guarantee ECMP of traffic to/from the device (without much
> difficulty such as MPLS-TE).
Unless the difference is HUGE, you usually don=E2=80=99t really care.
> You also can't do hop-by-hop routing based on 0/0 or ::/0 when the =
ring
> contains multiple devices also doing the same thing. You'll just =
create
> a loop. MPLS-based forwarding is your friend here.
Who said anything about a ring. He is advertising a /24 to 2 upstream =
providers.
Likely these are two separate transit circuits.
> But yes, if your device is not in a ring, then your suggestion is =
fine.
Even if you=E2=80=99re in a ring if you=E2=80=99ve got two transit =
providers at some random point on the ring, it still probably doesn=E2=80=99=
t make a meaningful difference between full feeds from each vs. ECMP, =
because it=E2=80=99s pretty unlikely that the AS PATH length is affected =
by the ring length.
Owen
