US Signal

We often get questions from customers asking how to configure routers for both Internet Access and MPLS solutions. The general rule is: "We have a managed router product for those who don’t understand how to configure the router themselves." However, there are certain elements that need to be configured using compatible protocols in order to turn up the circuit. In this article, we will be looking at some of the basic configuration parameters, and some sample configurations.

All US Signal IP Connections (Internet access and MPLS) are designed as two node, /30, IP Subnets. This means that they have a subnet mask of 255.255.255.252. In any /30 address there are four possible IP addresses: the first is reserved as a "Network Address" and the Last is the designated "Broadcast Address", leaving us with two usable IP addresses at each end of the circuit.

The most common router configuration is the standard Serial Line Configuration; this would include T1 and DS3 interfaces. These interfaces are configured with Point to Point Protocol (PPP). This is the Layer 2 protocol that defines how information is formatted on the circuit. US Signal also disables Cisco Discovery Protocol (CDP) on the interfaces to protect network integrity and security. A typical baseline configuration for an interface on a Cisco router might look something like this:

interface Serialslot/port
no shut
ip add xxx.xxx.xxx.xxx 255.255.255.252
no cdp enable
exit

The other common configuration is for Ethernet service. The configuration of the Ethernet port is effectively the same as it is for the Serial port. The port-specific components are handled by default:

interface FastEthernetslot/port
no shut
ip add xxx.xxx.xxx.xxx 255.255.255.252
no cdp enable
exit

The remaining Interface type would be the Bundled T1, where multiple T1s are combined to provide 3Mb or more. This configuration is slightly more complex. The Interface that the IP address is assigned to is a logical interface, and the Physical interfaces are assigned to the Logical Bundle Interface. Here is a sample configuration:

interface Multilink1
ip address XXX.XXX.XXX.XXX 255.255.255.252
no cdp enable
ppp multilink
ppp multilink group 1
exit

interface Serial0/0/1:0
no shut
encapsulation ppp
ppp multilink
ppp multilink group 1
exit

In this example, we need to: 1) define the encapsulation, or layer 2 protocol, as PPP for each physical interface, 2) identify them as part of a bundle, and 3) define the bundle. By defining the multilink logical interface as well as the physical interfaces as part of group 1, we have tied the physical interfaces to the logical interface. The IP address is then defined on the logical interface.

Once the interfaces are configured, the remaining necessary task is to add routing information so that the networks understand how to reach each other. Again, the most common way to do this is at the remote locations (MPLS) is with a static route, which looks like this:

ip route 0.0.0.0 0.0.0.0 xxx.xxx.xxx.xxx

This works for any location that meets the following criteria:

• Internet access is across the WAN
• There is only one LAN subnet at this location

The head end, where the internet is accessed from, is only slightly more complex, it will require a specific static route for each remote location; which looks like this:

ip route remote_LAN_Subnet remote_LAN_subnet_mask Remote_end_of_local_WAN_Link
ip route remote_LAN_Subnet remote_LAN_subnet_mask Remote_end_of_local_WAN_Link
ip route remote_LAN_Subnet remote_LAN_subnet_mask Remote_end_of_local_WAN_Link

It should include one statement for each LAN site.

These are the most basic configuration items required to turn up a circuit. There are many additional configuration parameters that may be set to protect your network. Things like access control, management protocols, and snmp have an impact on risk and manageability. Each network is different and you should review best practice processes to determine the best way to configure your equipment.

Subnet Mask Basics

A Subnet Mask delineates the Network Portion of the IP address from the Host Specific portion of the IP address. All devices in the same subnet share a common Network Portion of their address, and no other part of the internet can have that same Network Address. The host portion of the address must be unique within the subnet, and the combination of the Network and Host portions will be unique across the entire internet.

IP Addresses and subnet masks don’t really make much sense until you look at them as binary numbers:

A subnet mask represented in Binary format will start from the left with a series of ones followed by a series of zeros. Any part of the IP address that pairs with a one in the subnet mask will be part of the Network address, and the part of the IP Address paired with zeros will be the host portion of the address.

255.255.255.0 in Binary is:
11111111.11111111.11111111.00000000

This is a very common mask, because it is easy to work with even in decimal format. We typically write IP addresses in decimal because binary is very cumbersome, but to minimize the confusion, we break it into four groups of eight bits, or four octets, with 0 being the lowest value and 255 being the highest. Another way we define subnet Masks is by counting the number of ones, the above example is a /24 indicating that there are 24 bits in the network portion of the address.