[193383] in North American Network Operators' Group
RE: BGP Route Reflector - Route Server, Router, etc
daemon@ATHENA.MIT.EDU (Justin Krejci)
Fri Jan 13 17:12:48 2017
X-Original-To: nanog@nanog.org
From: Justin Krejci <JKrejci@usinternet.com>
To: Leo Bicknell <bicknell@ufp.org>, "nanog@nanog.org" <nanog@nanog.org>
Date: Fri, 13 Jan 2017 22:12:42 +0000
In-Reply-To: <20170113132304.GA74131@ussenterprise.ufp.org>
X-Securence-RFC2821-MAIL-FROM: jkrejci@usinternet.com
Errors-To: nanog-bounces@nanog.org
Thanks for all of the replies (on and off list). It is appreciated.=0A=
=0A=
Scaling in this context is simply adding more and more routers and needing/=
wanting to avoid configuring full mesh iBGP due to the administrative burde=
n of maintaining the growing size of full mesh topology. In one particular =
network in question, I have 11 routers fully meshed and need to add several=
more over the coming 6-12 months, possibly adding as many as 10 more route=
rs in that time span. I'd prefer not to continue doing full mesh.=0A=
=0A=
As for 7206VXR with NPE-G1 or G2 cards, we have many sitting in a decommiss=
ioned state on shelves as well as a few still alive serving a handful of T-=
1 lines and various other legacy connections of that sort. These little 720=
0's sit and run, forever near as I can tell. As many routers in this networ=
k do contain full route eBGP connections I will strongly consider your sugg=
estion of avoiding using the 7200's due to potential memory constraints and=
CPU/convergence time capabilities. I don't think I have done any full tabl=
e feeds on a 7200 in many years (days of 200k-300k table size days)=0A=
=0A=
This fits in with the kind of feedback I was hoping for, Thanks!=0A=
=0A=
=0A=
________________________________________=0A=
From: NANOG [nanog-bounces@nanog.org] on behalf of Leo Bicknell [bicknell@u=
fp.org]=0A=
Sent: Friday, January 13, 2017 7:23 AM=0A=
To: nanog@nanog.org=0A=
Subject: Re: BGP Route Reflector - Route Server, Router, etc=0A=
=0A=
In a message written on Thu, Jan 12, 2017 at 08:32:44PM +0000, Justin Krejc=
i wrote:=0A=
> I am working on some network designs and am adding some additional router=
s to a BGP network. I'd like to build a plan of changing all of the existin=
g routers over from full iBGP mesh to something more scalable (ie route ref=
lection).=0A=
=0A=
You might want to better define "scalable". I don't know your=0A=
background or network so I can't guess. I can say I've seen=0A=
the inner workings of some large ISP networks with a lot of hosts=0A=
in iBGP that work fine, and then people with 5 routers try and=0A=
tell me they have a scaling problem.=0A=
=0A=
What is your actual problem? Memory usage? Convergence time?=0A=
Configuring the sessions? Staff understanding of how it works?=0A=
=0A=
> I am wondering if people can point me in the direction to some good resou=
rce material on how to select a good BGP route reflector design. Should I j=
ust dust off some 7206VXR routers to act as route reflectors?=0A=
=0A=
This is a red flag to me, relative to the questions above.=0A=
=0A=
The 7206VXR, even with an NPE-G2, is a 1.5Ghz Power PC with a paltry=0A=
2GB of DRAM. It was not speedy when new, being roughly equivilent=0A=
to the PowerPC G4 processors in Apple Laptops at the time. It is=0A=
approximately 8 times slower than a current iPhone. Seriously.=0A=
=0A=
If convergence time is anything you care about, a 7206VXR is a very=0A=
bad choice. It may also run out of memory if you have a lot of=0A=
edges with full tables.=0A=
=0A=
So what's the actual "scaling" problem?=0A=
=0A=
--=0A=
Leo Bicknell - bicknell@ufp.org=0A=
PGP keys at http://www.ufp.org/~bicknell/=0A=
