Integrated IS-IS implements a CLNP node-based address structure to identify routers (even within pure IP environments). These CLNP addresses are called Network Service Access Points (NSAPs) and comprise three components : an area identifier (area ID) prefix, followed by a system identifier (SysID), and an N-selector, as shown in Table 6.3. Table 6.3. The NSAP Addressing Structure
The NSAP address shown in Table 6.3 is like a combination of an IP address and the upper-layer protocol in the IP header. The following list explains the different elements:
Each IS has a unique SysID, which together with the area ID and an N-selector value of 0x00 generates a special NSAP known as the node's network entity title (NET) or its network-level address. The HODSP, SysID, and NSEL work together to create the domain-specific portion of the NSAP address. Although there are many different NSAP representations for disparate OSI systems, NSAPs are typically represented in a variable-length format up to 40 hex digits. IS-IS and ISO-IGRP, for instance, interpret the NSAP in different ways to add to the confusion. As you have seen, the IS-IS combines the IDP (AFI+IDI) and the High Order DSP to derive the Level 2 (area) address, and the SysID (MAC Address) is used for Level 1 routing. When you add on the 1-byte NSEL you get a three-part address for the IS-IS NSAP. OSI-IGRP routes, however, use a tri-level domain hierarchy where the IDP (AFI+IDI) represents the third level, the HODSP is Level 2, and the SysID is Level 3. In other words, where IS-IS uses the ASI+IDI+HODSP for the area ID, ISO_IGRP breaks this up into a Domain ID (AFI) and an Area ID, as shown in Table 6.4. Table 6.4. The ISO-IGRP NSAP Addressing Structure is Different From IS-IS To Support a Tri-Level Hierarchy
The area ID for ISO-IGRP is represented by the 2 bytes left of the SysID field to accommodate the allowable 65,535 IGRP areas. The remaining 1 “11 bytes to the left of the area ID is the Domain ID field. That makes the minimum length for an ISO-IGRP NSAP equal to 10 bytes. In other words: 1-byte NSEL (fixed but unused) + 6-byte SysID (fixed) + 2-byte Area ID (fixed) + a minimum 1-byte Domain ID (variable) = 10 bytes.
IS-IS utilizes this node-based addressing scheme of CLNP, and standard NSAP prefixes must be used for routing data packets. It is a basic prerequisite that, even when Integrated IS-IS is used in an IP-only setting, the nodes must have ISO NSAP addresses ( NETs ). Addresses that begin with the AFI value of 49 are considered private internal addresses, similar to the ranges set aside for IP. (Refer to Chapters 1 and 2 to review IP addressing.) Although these addresses are routable, they should not be advertised outside the CLNS network. The AFI value starting with 39 is used to designate the ISO data country code. An AFI value beginning with 47 specifies the international code. For example, look at the following NSAP: 49.0010.xxxx.yyyy.zzzz.00 For IS-IS, the area ID is 49.0010 , the SysID is xxxx.yyyy.zzzz , and the NSEL value is 00 (NET). However, if ISO-IGRP is being used, then the Domain is 49, the Area is 0010, the SysID is xxxx.yyyy.zzzz , and the NSEL value is unused. When you enable Integrated IS-IS for IP routing on an interface, the IP subnets are automatically added to the router's LSP. The IP prefixes are then accumulated into an IP link-state database, which is used by the SPF algorithm to determine the best routes. |