From: "Hoonpongsimanont
Date: Mon Nov 22 2004 - 21:03:33 GMT-3
I guess you are talking about GRE tunnel.
1. Where do you put the source and why?
2. Where do you put the destination and why
Normally, you can put any interfaces on router as tunnel source and
destination. Packets going into tunnel will be encapsulated with this source
and destination address. In the lab, if you got something says make sure
tunnel always up as long as there is an active path, it will be a good idea
to use loopback address. Make sure that two routers can ping each other
using these addresses. They need to learn route through some routing
mechanism. Make sure that these routes are always valid on the tunnel ended
router especially when you are running another routing protocol over the
tunnel.
3. When do you assign an IP address to the tunnel?
When creating the tunnel, you need to assign IP address to it. You can
assign any IP address you want. Anyway, if in the question you are not
allowed to add a new subnet, consider using unnumbered interface.
4. Does the path show up in the routing table at each end? If not, how
does the router know where to send the packets to?
Yes. The two routers will see each other as directly connected route. It's
like logical direct interface connect this two routers together. Imagine
like they are connected back-to-back. You can do almost everything on this
interface, like run routing protocol, multicast, filtering, etc.
Normally we use tunnel as a last resort. Tunnel is a great way to solve
problem that seems impossible to solve. For example,
Connect OSPF area X to area 0 without virtual link, or across stub area.
Unequal cost load balance on RIP, OSPF, IS-IS.
Running ISIS over multipoint frame relay.
Running multicast load balancing
In my real life, I used tunnel only one time to transport IPX traffic over
MPLS VPN.
Hope this help,
David
-----Original Message-----
From: Stephen Hull [mailto:shull@getsouthern.com]
Sent: Tuesday, November 23, 2004 6:30 AM
To: ccielab@groupstudy.com
Subject: R&S Router Tunnels
I need some insight and direction for some good information about
tunneling between routers. I understand the concept and the idea behind
them, I just can not seem to get them to work. I am not talking about
IPSec tunnels and VPNs. I am talking about the R&S topic listed in the
blueprint that includes tunnel interfaces. I know I need to know these,
but I am having trouble wrapping my arms around the actual
implementation and why and where you put them. I tried the Cisco
document, but it helped only a little. I don't know why I am having so
much trouble with tunnels, but I am.
I want to know things like:
1. Where do you put the source and why?
2. Where do you put the destination and why
3. When do you assign an IP address to the tunnel?
4. Does the path show up in the routing table at each end? If not, how
does the router know where to send the packets to?
5. What does the tunnel simulate; a Layer 2 connection or a Layer 3
connection.
6. etc...
So many questions. Can someone please show me the light?
Thanks for any help,
-Stephen
This archive was generated by hypermail 2.1.4 : Thu Dec 02 2004 - 06:57:48 GMT-3