[148176] in North American Network Operators' Group
Re: anycast load balancing issue
daemon@ATHENA.MIT.EDU (Johannes Resch)
Thu Jan 5 10:13:43 2012
Date: Thu, 05 Jan 2012 16:12:33 +0100
From: Johannes Resch <jr@xor.at>
To: =?UTF-8?B?TcOlbnMgTmlsc3Nvbg==?= <mansaxel@besserwisser.org>
In-Reply-To: <20120104120255.GT7491@besserwisser.org>
Cc: nanog@nanog.org
Errors-To: nanog-bounces+nanog.discuss=bloom-picayune.mit.edu@nanog.org
Hi,
On 04.01.2012 13:02, M=C3=A5ns Nilsson wrote:
>
> [..snipped..]
>
> Trouble is, we find that (untweaked) cost and metric are such that all
> nodes are equal. The last resort (peer router ID) gets invoked and all
> traffic goes to one single instance. Of course, when that instance fall=
s
> off the net recalculation takes place and another node steps in, but
> I'd like true path lengths (IGP hop count) to influence more than iBGP
> (route-reflector-style) selection.
>
> Any clues?
>
> Oh, all-cisco, all ASR1000 series. All links GE. ~90 routers in IGP.
>
Since you mention route-reflector route selection - are you already=20
using per-VRF, per-PE route distinguishers for that L3VPN instance?
If not, I'd recommend doing so - this will cause your RR to see all=20
paths as unique routes, distributing all of them (instead just the best=20
one from the RR perspective) to RR clients. As result all PEs will=20
always have all paths for this particular prefix (and can then take the=20
best path decision based on local IGP metric to the respective BGP next=20
hops).
Doing that can also significantly improve reconvergence times for=20
certain failure scenarios (e.g. ingress PE failure), as PEs can start=20
using alternative paths (already available in local BGP RIB) as soon as=20
the IGP nexthop for the failed PE is invalidated and do not need to wait=20
for BGP RR reconvergence.
cheers,
-jr
