From: Scott Morris (swm@emanon.com)
Date: Sun Nov 11 2007 - 18:33:54 ART
hehehehe... That's always a thought! I had not quite pondered that yet.
Your limitation/problems aren't really with the number of routers (at least
to an extent) but rather the size of your routing/ls table and what you are
attempting to do with it. You'll run into limitations there much more
quickly (IMHO) than you would with a pure 'number of routers' problem.
1,200's a good number, but I've also heard 2,000 before. So I'm more
inclined to look (or answer) based on the functional problems rather than
simply a raw number. (which I think would be the appropriate "it depends"
DE answer as well!)
:)
Scott
-----Original Message-----
From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of Wes
Stevens
Sent: Sunday, November 11, 2007 2:08 PM
To: Pete Templin; Tarun Pahuja
Cc: Dishan Gamage; Cisco certification; Cisco certification
Subject: Re: ospf muti-area to isis
Pete, I have seen different guides as far as how may routers in an L2 isis
core. Some say 400 and I have seen others say you can go up to 1200. Not
sure how big your network is or will grow - was this a consideration? The
type of router also comes into play here.
Scott,
This is likely to come up in the DE practical. What is your thoughts on how
big a single L2 area can get?
--- Pete Templin <petelists@templin.org> wrote:
> Tarun Pahuja wrote:
> > Dishan,
> > There a lot of consideration a
> company has to make before
> > moving from one protocol or another. Since I do
> not know anything about
> > your network and the motivation behind moving to
> ISIS from OSPF. I can
> > only give you a couple of tips to get started
> thinking about a few
> > things. You will have to master ospf and isis in
> order to do the
> > migration(unless you want to get a consultant in).
> The size of the
> > network, link speeds, type of routers, etc would
> play a very important
> > role in migration phase as well as the overall
> final design.
>
> Honestly, I didn't have to learn much about ISIS to make the switch,
> though we did switch from multi-area OSPF to single-area ISIS (the
> areas weren't doing anything for us). The migration was incredibly
> seamless; I planned to roll one POP per night, almost for the fun of
> watching the migration go across the network. However, the only
> glitch we ran into was with MPLS Traffic Engineering: it's been a
> while, but I think TE only likes to have one TE-capable IGP in
> service.
> This forced us to
> make the switch in one night, which was no big deal anyway.
>
> > On Nov 11, 2007 9:25 AM, Pete Templin
> <petelists@templin.org
> > <mailto:petelists@templin.org>> wrote:
> >
> > http://www.nanog.org/mtg-0310/gill.html
>
> Again, this was all I needed. Granted, we were a service provider
> network with a topology much like ATDN (though with only five POPs).
> The basic premise is simple: roll out ISIS across the network, prefer
> ISIS across the network, remove OSPF across the network. Done.
>
> pt
>
>
This archive was generated by hypermail 2.1.4 : Sat Dec 01 2007 - 06:37:29 ART