[nsp] Bridging?

Vincent De Keyzer vincent at dekeyzer.net
Wed Apr 28 04:47:25 EDT 2004


Douglas,

your efforts to understand my problem are remarkable and I highly appreciate
them.

But I haven't been clear enough once again (I could say it's because english
is not
my mother language, but I am not sure I could do better in french... :) so
let me
try another way.

IP ranges:
A.B.1.0/24: backbone ranges
A.B.2.0/24: PtP (/30) ranges
A.B.3.0/24: customer X range
A.B.4.0/24: customer Y range

Current situation:
* a 2621 with
Fa0/0 = A.B.1.2/28
Serial 0/0:0 = A.B.2.1/30 for customer X
Serial 0/1:0 = A.B.2.5/30 for customer Y
* customer X has a 1600+CSU/DSU with
Serial0 = A.B.2.2/30
Ethernet0 = A.B.3.1/24
* customer X has a firewall with
WAN = A.B.3.2/24
* customer Y has a 1600+CSU/DSU with
Serial0 = A.B.2.6/30
Ethernet0 = A.B.4.1/24
* customer Y has a firewall with
WAN = A.B.4.2/24

The 2621 is routing IP packets between customer ranges A.B.3.0/24 and
A.B.4.0/24 and the core network behind Fa0/0.

And, after replacing the (1600+CSU/DSU) by a E1/Eth bridge, I would like it
to become:
* a 2621 with
Fa0/0 = A.B.1.2/28
Whatever0 = A.B.3.1/24
Whatever1 = A.B.4.1/24
* customers X&Y have an E1/Eth bridge and their FWs keep the same config

... where "Whatever" interfaces are whatever kind of interface it takes to
have this working...

Now maybe this is now even called bridging... but is it possible?

The goal is of all this is to replace 1600+CSU/DSU by less expensive
equipment where it's not required.

Thanks in advance,

Vincent


> -----Original Message-----
> From: cisco-nsp-bounces at puck.nether.net 
> [mailto:cisco-nsp-bounces at puck.nether.net] On Behalf Of Todd, 
> Douglas M.
> Sent: mardi 27 avril 2004 16:29
> To: 'Vincent De Keyzer'; cisco-nsp at puck.nether.net
> Subject: RE: [nsp] Bridging?
> 
> 
> Vincent:
> 
> 1: If I understand the scenario correctly we are trying just 
> to bridge the two
> ether segments together making one broadcast domain?  Or are 
> you trying just to
> bridge two serial interfaces together thus creating the same 
> broadcast domain
> (There are better ways of doing this).
> 
> Thus for merging the segments:
> 
> Segment1 -> Router1 -> WAN -> Router2 -> Segment2
> Would become:
> 
> Segment1 -> Router1 -> WAN -> Router2 -> Segment1
> 
> OR:
> 
> 
> Segment1 -> Router1:Net1:ip1 -> WAN - ip4:Net1:Router 2 -> Segment1
> Segment1 -> Router1:Net1:ip2 -> WAN - ip5:Net1:Router 3 -> Segment3
> Segment1 -> Router1:Net1:ip3 -> WAN - ip6:Net1:Router 4 -> Segment4
> 
> Or:
> 
> Segment1 -> Router1:Net1:ip1 -> WAN - ip4:Net1:Router 2 -> Segment1
> (Point to multipoint interface)	  ip5:Net1:Router 3 -> Segment3
> (Point to multipoint interface) 	  ip6:Net1:Router 4 -> Segment4
> 
> 
> (Point to multipoint design with un-numbered interfaces would 
> accomplish this
> better)
> 
> 2: Start with what protocol are you bridging, IP?  I guess I 
> am unclear why you
> would want to do this unless you don't want to break up the 
> address space, ie:
> make a point-to-multipoint design (Keep all s0s in the same network).
> 
> 3: You are replacing the 1600 with a bridged 2600?  
> 
> 4: David's response is right if you are looking to bridge 
> from bvi0->s0 etc.
> 
> ==DMT>
> 
> 
> 
> 
> ---
> Incoming mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.670 / Virus Database: 432 - Release Date: 27/04/2004
>  
>     
> 

---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.670 / Virus Database: 432 - Release Date: 27/04/2004
 




More information about the cisco-nsp mailing list