RE: virtual link vs. gre tunnel

From: Richard Dumoulin (richard.dumoulin@vanco.es)
Date: Mon Jun 28 2004 - 14:14:37 GMT-3


GRE is just an extra header added and I don't think it is so processor
intensive. Its true that efficiency is reduced but you would be suprised to
know how much this technology is used in our days of VPN's through the
Internet,

--Richard

-----Original Message-----
From: James [mailto:james@towardex.com]
Sent: lunes, 28 de junio de 2004 18:56
To: ccie2be
Cc: Group Study
Subject: Re: virtual link vs. gre tunnel

I always avoid gre tunnels if at all possible. using gre tunnels will most
likely transit traffic thru the tunnel, which will hog the router CPU and
give you poor throughput performance when you start pushing real traffic.
Yes, it may certainly be doable for lab situation, and it's certainly good
idea to try it out yourself, but in real-world application, i fail to see
the usefulness of transiting traffic to the backbone thru a gre tunnel.
virtual-link sounds better of idea to be used. you can use tunnel to do
OSPF, and use native path for actual traffic to take, but that kind of
defeats the purpose of having an IGP in the first place and what a black
magic mess that will be :) virtual-link will use real native path to push
the transiting traffic.

my 0.02

-J

On Mon, Jun 28, 2004 at 11:12:34AM -0400, ccie2be wrote:
> Hi guys,
>
> What's the advantage or disadvantage of using a virtual link to
> connect a discontinuous ospf backbone area vs using a tunnel to do the
> same thing?
>
> Besides the authentications available with a virtual link, are there
> any other differences?
>
> Also, let's assume I wanted to use a tunnel in the following scenario:
>
> Area0 --- s0 Rx E0 --- Area xy --- fa0/1 Ry s1 --- Area0
>
>
> Rx and Ry share a common ethernet subnet and each also has a loopback0
> interface.
>
> What should the endpoints of the tunnel be?
>
> Should any pair of endpoints work as long as they're consistent on
> both Rx and Ry? Why yes or no?
>
> Does it matter what area the physical or loopback interfaces endpoints
> of the tunnel are assigned to?
>
> I know that when the loopback interfaces are put in area 0 and the ip
> addr of the tunnel itself is assigned to area 0, this config works.
> But, I tried other which didn't work but I don't understand why.
>
> Could someone explain why that is?
>
> TIA, Tim
>
> ______________________________________________________________________
> _
> Please help support GroupStudy by purchasing your study materials from:
> http://shop.groupstudy.com
>
> Subscription information may be found at:
> http://www.groupstudy.com/list/CCIELab.html

-- 
James Jun                                            TowardEX Technologies,
Inc.
Technical Lead                        Network Design, Consulting, IT
Outsourcing
james@towardex.com                  Boston-based Colocation & Bandwidth
Services
cell: 1(978)-394-2867           web: http://www.towardex.com , noc:
www.twdx.net


This archive was generated by hypermail 2.1.4 : Sat Jul 03 2004 - 19:40:51 GMT-3