This sample configuration shows an
ImageStream router connected to a Digital Subscriber
Line aggregation circuit. Each PVC represents a DSL
customer terminated on the router. This example uses
the "atm route-bridged ip" command. This command automatically
sets up a bridge group, binds the ATM subinterface to
it. "atm route-bridged ip" also turns off the spanning
tree protocol on the port, sets the forward time to
one second and enables proxy ARP so that each PVC can
communicate with other PVCs, yet still have a netmask
larger than a /32. All routes added to the ATM subinterface
are redirected by the router to the automatic bridge
group instead.
====ETHERNET===================== 192.168.1.0/24 ===
|
| LAN IP: 192.168.1.100
+-------+--------+
| |
| Router A |
| Serial0 |
+-------+--------+
| WAN IP: 192.168.54.1/24 (255.255.255.0)
|
|
| ATM WAN
|-------|--------|
| | |
| | |
+ + +
PVC PVC PVC
1/101 1/102 1/103
Before You Start
Be sure to have three separate IP
subnets, one for the WAN network and two for the Ethernets
connected to the routers. Each WAN port connected to
the point-to-point WAN must have it's own IP address
from the same IP network. Since they are on the same
IP network, they will have the same netmask.
These IP network addresses are given to you by your
Internet Service Provider, your leased line carrier,
or by ARIN. If you
do not plan to connect you network to the Internet,
you can use an IP network address set aside for private
use. The IP network address for private use can be found
in RFC 1597. Most people will use 192.168.0.0 as we
have in this example.
Configuring The WAN Ports
In this example, we are going to
assume the following:
- Router A has an Ethernet IP address
of 192.168.1.100 with a netmask of 255.255.255.0
- Router A has a WAN IP address of
192.168.54.1 with a netmask of 255.255.255.0
- The DSL Customer with VPI 1 and VCI
101 has an IP address of 192.168.0.10
- The DSL Customer with VPI 1 and VCI
102 has an IP address of 192.168.0.11
- The DSL Customer with VPI 1 and VCI
103 has an IP address of 192.168.0.12
The IP addresses used in this Technical
Note are examples only, you will need to use an IP network
given to you by your Internet Service Provider.
Router A
!
interface Loopback0
ip address 192.168.0.1 255.255.255.0
!
interface Serial0
description ATM0
encapsulation atm
!
interface Serial0.101
description CustomerA
ip unnumbered Loopback0
atm route-bridged ip
pvc 1/101
encapsulation aal5snap
!
interface Serial0.102 point-to-point
description CustomerB
ip unnumbered Loopback0
atm route-bridged ip
pvc 1/102
encapsulation aal5snap
!
ip route 192.168.0.10 255.255.255.255 Serial0.101
ip route 192.168.0.11 255.255.255.255 Serial0.102
ip route 192.168.0.12 255.255.255.255 Serial0.103
!
end
The router will automatically configure
up a brSerial0.101 bridge for the first subinterface,
a brSerial0.102 subinterface for the second and a brSerial0.103
subinterface for the third. The configuration also adds
individual IP routes to 192.168.0.10/32 via brSerial0.101
(instead of Serial0.101), a route to 192.168.0.11/32
via brSerial0.102 and a route to 192.168.0.12/32 via
brSerial0.103. ATM routed bridge encapsulation reduces
the security risks and added broadcast traffic associated
with bridging all customers into a single bridge group.
This configuration method simplifies the configuration
by reducing the number and complexity of the devices
required to support each customer.
Troubleshooting
If the port does not have an "up"
status for both hardware and protocol in the interface
statistics output (Option 2 from the Main menu) do the
following:
If hardware shows "down", check your cable connections
and ensure that you have connected to the correct port.
If you have connected to the correct port, double-check
the telephone company smart jack or other telephone
company-installed equipment for your circuit. The smart
jack should show no red alarms if the cables are connected.
If, after connecting the cables correctly, you have
errors on the telephone company equipment, contact your
line provider for assistance.
If you have a T1 CSU/DSU, put the CSU/DSU into a local
loopback. This will cause each packet sent to the CSU/DSU
to reflect back to the WAN port. This is for testing
only, the line will not function while the CSU/DSU is
in a local loopback. If, in the interface statistics
detail output for the port, you see transmitted packets
immediately received on the same port, then you can
determine the following facts:
- The WAN port on the router is sending
and receiving data.
- The cable and connectors between
the router and the CSU/DSU are functioning.
- The DTE port on the CSU/DSU is functioning.
- THE ROUTER IS FUNCTIONING AS IT SHOULD.
If you don't get your packets back
immediately on the same interface:
- For cards with a software-selectable
serial interface, check that you have specified
the correct wiring specification (V.35, RS232, RS422/X.21).
The "dctype" command is typically set to V.35 (the
default) in North America and RS422 in Europe and
Asia.
- That all cables are securely connected.
- Begin swapping hardware, cables,
CSU/DSU, etc.
If you get your packets back immediately,
but the line protocol status is not "up":
- Check the CSU/DSU settings such as
clocking, etc. Almost always the CSU/DSU should
be set for external clocking. Your line provider
typically provides the clock. It's not a bad idea
to verify any external CSU/DSU settings with the
manufacturer.
- Have their telephone carrier test
the line. Requesting a hard copy of the test results
will often get you a better test.
If the line protocol status is "up"
but you can't ping the other side:
- Double check settings. Make sure
that the IP address for the WAN port on both sides
are in the same subnet.
- Do a traceroute from your workstation
through the local (to you) router to the router
in trouble. Then do a traceroute from the router
in trouble (assuming you can get to it) back to
your workstation. The problem will lie in the gap
between the two traceroutes.
- Make sure the default gateway is
set to the upstream router or the serial device
used for the upstream connection, as in our example
above on Router B.
At this point, ATM cells should be being transmitted
and received. The status of the WAN port should change
to up. It should now
be possible to ping the other side. The interface
statistics (Option 2 from the Main menu) should show:
|