Help Split VLAN

[wananga01]

We have a Main Campus and a Remote Site approx 400 meters apart.

Previously the Remote Site was linked by a wireless connection and shared a common LAN with the Main Campus. Servers at the Main Campus provide DNS, DHCP, Email, Internet & data storage to clients depending VLAN membership configured on a Cisco 3550 Layer 3 switch.. The Cisco 3550 was the gateway to all services provided by the servers. A network of layer 2 switches connected clients to the Cisco 3550.

Last week we leased a fiber cable channel to replace the Wireless Link that connected the 2 sites. The channel service provider has cabinets at the Main Campus and Remote Site with a Cisco 1601 Router in each one (Router A & Router B). The Routers have serial connection to fiber cable.

At the Main Campus – Router A is connected via Ethernet0 10.2.0.249/16 to the Cisco 3550 Layer 3 switch that has VLANs and provides access to Main Servers. Serial0 192.168.15.253/30 is connected to the fiber cable via a MUX– HDLC is encapsulated on Serial 0, RIP is enabled. Both interfaces are up

At the remote site – Router B is connected via Ethernet0 10.3.0.254/16 D-Link Layer 2 switch. Serial 0 192.168.15.254/30 is connected to the fiber cable via a MUX - HDLC is encapsulated on Serial0, RIP is enabled. Both interfaces are up

This is where my problem lies:

All clients at the Remote site share the VLAN 200 subnet 10.2.0.0/16 range with clients and services at the Main Campus. The Routed link has effectively split the VLAN and stopped the clients at the Remote Site from accessing services.

Can anyone advise what the best solutions is?

Do I need to put a layer 3 switch at the remote site?

Do I need to set the Routers up to Bridge? If so how do I do this?

[thisisshanky]

Since you split the network which was once L2 network, by a L3 segment. Your best bet is to provide a different network address for the clients in remote building. Since its all DHCP, you should be doing it fine, by just changing the DHCP scope on your central site's DHCp server. The 3550 and the 1600s will need to know how to reach this network. So you will have to include that new network in RIP, or configure static routes on the router (which ever is appropriate)

[thisisshanky]

And yeah, dont forget to use ip helper-address command on the L3 device (either 1600 or L3 switch) on the remote side to relay all DHCP requests to the central site DHCP server.

[wananga01]

Can you take a look at the following conf files for the 2 routers.

On the Layer 3 switch we have the following VLANS

Default - Management - 10.0.0.254/24

VLAN100 - Server VLAN - 10.10.10.0.254/24

VLAN200 - Admin VLAN - 10.2.0.254/16

VLAN300 - Student VLAN - 10.1.254.254/16

Vlan999 - Management - 10.1.254.254/16

In my DHCP scope (10.3.0.0/16 subnet) for the clients on the remote side, do I set the Router option as 10.2.0.254/16

as I need the remote clients to be able to access VLAN200.

[schwermie]

First of all, your "Router Option" should be 10.3.0.254 for clients on the remote site, as this will be there default gateway.

On Router B you have a static route 10.0.0.0/8 through 10.2.0.0, this route should be deleted!

I would suggest you use Rip Version 2 with "no auto-summary". On Router A i would suggest you set a "default-information originate" wich will insert a default route pointing to Router A.

---------------------------------------------

--- Cut config Router A ---

Router Rip

!Set Rip version

version 2

!Set connected networks

network 10.2.0.0

network 192.168.15.252

default-information originate

!If RIP is not used on e0 then:

passive-interface e0

!disable auto-summary

no auto-summary

--- end Cut ---

---------------------------------------------

--- Cut config Router B ---

Router Rip

!Set Rip version

version 2

!Set connected networks, will change into classful

!addresses.

network 10.3.0.0

network 192.168.15.252

!If RIP is not used on e0 then:

passive-interface e0

!disable auto-summary

no auto-summary

--- end Cut ---

After this you should see something like below on Router B.

"sh ip route"

10.0.0.0/9 is variably subnetted, 2 subnets, 1 masks

R 10.2.0.0/24 [120/1] via 192.168.15.253, Ser0

S 10.3.0.0/24 is directly connected, Ethernet0.

R* 0.0.0.0/0 [120/0] via 192.168.15.253

[wananga01]

Thanks I changed config.

When telnet to Router A (int e0 10.2.0.249) I can ping s0 and as well as other nodes in the 10.2.0.0/16 subnet. I can also Router B's interfaces.

From node 10.2.1.76/16 I can ping to Router A int e0 10.2.0.249 but no further.

From Router B I can ping e0/s0 as well as Router A's interfaces but no further.

Any further idea?

I can ping from a workstation 10.2.1.76/16 to Router A e0 but no further

[schwermie]

Can you post a "show ip route" from both routers?

When you can't see any rip entry's post a "debug ip rip" as well.

What is the default gateway configured on node 10.2.1.76 and on nodes 10.3.x.x?

[wananga01]

Routers are pinging however Router A can ping anywhere Router B can ping to Router A Int E0 but not to host connect to that interface.

We are also going to provide PABX telephone services to the remote site via Primary Rate. This will mean voice & data over the link. I know that I will need to update the IOS to accommodate QOS/prioritisation.

Any suggestions on a suitable configuration?