NOTE
This chapter is a supplement to the Cisco Press book Building Scalable Cisco Networks (ISBN 1-57870-228-3). It covers the IS-IS material in accordance with the latest version of Building Scalable Cisco Internetworks. This chapter was created to follow Chapter 4, "Interconnecting Multiple OSPF Areas." This placement is reflected in this chapter's starting assumption in the configuration exercise. This supplemental chapter is not meant to replace any material in the original book, but it should be read between Chapters 4 and 5 of the book.
In addition to this supplemental chapter, you will find online addenda to Appendixes D, G, H, and I. The addendum to Appendix D, "Summary of BSCN Router Commands," provides you with a descriptive list of the commands you will find in this chapter. The addendum to Appendix G, "Answers to the Review Questions," provides you with the answers to the review questions at the end of this chapter. The addendum to Appendix H, "Configuration Exercise Equipment Requirements and Backbone Configurations," provides the configuration commands for the backbone router for the configuration exercise at the end of this chapter. The addendum to Appendix I, "Glossary," provides a listing of some of the important terms and acronyms used in this chapter.
This chapter introduces the Intermediate System-to-Intermediate System (IS-IS) protocol. This chapter includes the following sections:
Overview of OSI Protocols and IS-IS Routing
Operation of IS-IS for CLNS/CLNP
IP and OSI Routing with Integrated IS-IS
Basic Integrated IS-IS Router Configuration
Modeling WAN Networks in Integrated IS-IS
Summary
Configuration Exercise: Configuring a Multiarea IS-IS Network
Answers to Configuration Exercise: Configuring a Multiarea IS-IS Network
Review Questions
This chapter provides an overview of Intermediate System-to-Intermediate System (IS-IS) technology and its structures and protocols. It also gives basic configuration examples. The chapter begins with Open System Interconnection (OSI) routing and then focuses on Integrated IS-IS, which supports IP routing. Basic IS-IS and Integrated IS-IS router configuration commands, examples, and some troubleshooting guidelines are presented. The major part of this chapter is dedicated to an explanation of IS-IS concepts and capabilities, including the hierarchy and addressing of OSI-based networks.
When you finish this chapter, you will be able to explain the basic OSI terminology and network layer protocols used in OSI, the way in which networks and interfaces are represented in IS-IS, the basic principles of area routing, and the use of IS-IS in nonbroadcast multiaccess (NBMA) environments. You will be able to identify similarities and differences between Integrated IS-IS and OSPF, as well as characteristics of an effective addressing plan for IS-IS deployment. You will be able to list the types of IS-IS routers along with their roles in IS-IS area design, and describe the hierarchical structure of IS-IS areas, the concept of establishing adjacencies, and the concepts of routing information and database synchronization. Finally, given a set of network requirements, you will be able to configure Integrated IS-IS and verify proper operation (within described guidelines) of Cisco routers.
Overview of OSI Protocols and IS-IS Routing
This section provides an overview of the OSI protocols and IS-IS routing.
OSI Protocols
ISO and OSI: What's the Difference?
The International Organization for Standardization (ISO) was formed to develop standards for data networking. (As an interesting aside, note that the word ISO is Greek for "same.")
The Open System Interconnection (OSI) protocols represent an international standardization program that facilitates multivendor equipment interoperability.
The OSI protocols are part of an international program to develop data-networking protocols and other standards that facilitate multivendor equipment interoperability. The OSI program grew out of a need for international networking standards and is designed to facilitate communication between hardware and software systems despite differences in underlying architectures.
The ISO has been charged with developing standards for data networking.
The OSI specifications were conceived and implemented by two international standards organizations: the ISO and the International Telecommunication Union Telecommunication Standardization Sector (ITU-T).
A Condensed History of IS-IS
IS-IS was ad hoc in its evolution, whereas OSPF was more formal:
1985: Originally called DECnet Phase V Routing
1988: Adopted by ISO and renamed IS-IS
1990: Publication of RFC 1142, "OSI IS-IS Intradomain Routing Protocol"
1990: Publication of RFC 1195, "Use of OSI IS-IS for Routing in TCP/IP and Dual Environments"
1991: Cisco IOS Software starts supporting IS-IS
1995: ISPs start adopting IS-IS
2000: Publication of IETF draft "IS-IS Extensions for Traffic Engineering"
2001: Publication of IETF draft "IS-IS Extensions in Support of Generalized MPLS"
The world of OSI internetworking includes various network services with these characteristics:
Independence of underlying communications infrastructure
End-to-end transfer
Transparency
Quality of service (QoS) selection
Addressing
Transparency
A protocol is transparent when it does not place any constraints on transmitted data. It means that headers as well as data must be transported unmodified end to end.
Who Uses IS-IS?
IS-IS is popular among telcos and large ISPs. This popularity finds its roots with ISPs that were around at the beginning of the Internet and chose IS-IS over OSPF for their IGP. It is believe that at that time, IS-IS had fewer technical limitations than OSPF as an IGP. Those ISPs have since become today's tier 1 carriers, so any appliance targeting tier 1 carriers must offer IS-IS.
The OSI protocol suite supports numerous standard protocols at the physical, data link, network, transport, session, presentation, and application layers.
OSI network layer addressing is implemented by using two types of hierarchical addresses: network service access point (NSAP) addresses and a specific subset of NSAPs cal- led network-entity titles (NETs). An NSAP is a conceptual point on the boundary between the network and transport layers. The NSAP is the location at which OSI network services are provided to the transport layer. Each transport layer entity is assigned a single NSAP, which is individually addressed in an OSI internetwork using NSAP addresses.
The OSI protocol suite specifies two routing protocols at the network layer: End System-to-Intermediate System (ES-IS) and IS-IS. In addition, the OSI suite implements two types of network services: connectionless service and connection-oriented service.
Differences Between Connection-Oriented and Connectionless Services
Connection-oriented services must first establish a connection with the desired service before passing any data. A connectionless service can send the data without establishing a connection first. In general, connection-oriented services provide some level of delivery guarantee, whereas connectionless services do not.
Details on these protocols and addresses are discussed later in this chapter, in the sections, "OSI Network Layer" and "OSI Addressing," respectively.
OSI Protocol Terminology
In an OSI network, four significant architectural entities exist: hosts, areas, a backbone, and a domain. The following describes these entities and tells how routers fit into an OSI network:
A domain is any portion of an OSI network that is under a common administrative authority.
Within any OSI domain, one or more areas can be defined. An area is a logical entity; it is formed by a set of contiguous routers and the data links that connect them. All routers in the same area exchange information about all the hosts that they can reach.
The areas are connected to form a backbone. All routers on the backbone know how to reach all areas.
An end system (ES) is any nonrouting host or node. An intermediate system (IS) is a router. These terms are the basis for the OSI ES-IS and IS-IS protocols.
NOTE
Nowadays, domains tend to be called autonomous systems.
Mapping the OSI Protocol Suite to the OSI Reference Model
The OSI protocol suite supports numerous standard protocols at each of the seven OSI reference model layers.
Differences Between the OSI Protocol Suite and the OSI Reference Model
The ISO developed the Open Systems Interconnection networking suite in the 1980s. It has two major components:
The abstract model of networking, known as the OSI Reference Model, or seven-layer model
A set of concrete networking protocols, known as the OSI protocol suite, which include CLNP, ES-IS, and so on
The OSI Reference Model has enjoyed a far greater acceptance than the OSI protocol itself.
Figure S-1 illustrates the entire OSI protocol suite and its relation to the layers of the OSI reference model.
)
Figure S-1 How the OSI Protocol Suite Maps to the OSI Reference
Model
OSI Network Layer
This section describes the services and protocols at the OSI Network layer.
OSI Services and Network Protocols
Two types of OSI network layer services are available to the OSI transport layer:
Connectionless Network Service (CLNS)—CLNS performs datagram transport and does not require a circuit to be established before data is transmitted.
Connection-Mode Network Service (CMNS)—CMNS requires explicit establishment of a path or circuit between communicating transport layer entities before transmitting data.
Whereas CLNS and CMNS define the actual services provided to the OSI transport layer entities that operate immediately above the network layer, Connectionless Network Protocol (CLNP) and Connection-Oriented Network Protocol (CONP) name the protocols that these services use to convey data at the network layer. CLNP is the OSI equivalent of IP.
The differences between CMNS and CONP are as follows:
CONP is an OSI network layer protocol that carries upper-layer data and error indications over connection-oriented links. CONP is based on the X.25 Packet-Layer Protocol (PLP) and is described in the ISO 8208 standard "X.25 Packet-Layer Protocol for DTE." CONP provides the interface between CMNS and upper layers. It is a network layer service that acts as the interface between the transport layer and CMNS; it is described in the ISO 8878 standard.
CMNS performs functions related to the explicit establishment of paths through CONP. When support is provided for CMNS, routing uses the X.25 protocols as the relaying functions. CMNS functions include connection setup, maintenance, and termination; it also provides a mechanism for requesting a specific QoS.
The differences between CLNP and CLNS are as follows:
CLNP is an OSI network layer protocol that carries upper-layer data and error indications over connectionless links. CLNP provides the interface between CLNS and upper layers.
CLNS provides network layer services to the transport layer through CLNP. When support is provided for CLNS, routing uses routing protocols to exchange routing information. CLNS does not perform connection setup or termination because paths are determined independently for each packet that is transmitted through a network. In addition, CLNS provides best-effort delivery, which means that no guarantee exists that data will not be lost, corrupted, misordered, or duplicated. CLNS relies on transport layer protocols to perform error detection and correction.
Summary of OSI Protocols and Services
|
Type |
Protocol |
Service |
|
Connection-oriented |
CONP |
CMNS |
|
Connectionless |
CLNP |
CLNS |
OSI Routing Protocols
ISO has developed standards for two types of routing protocols:
ES-IS discovery protocol—ES-IS performs "routing" between End Systems and Intermediate Systems referred as Level 0 "routing." ES-IS is analogous to the Address Resolution Protocol (ARP) in IP. Although it is not explicitly a routing protocol, ES-IS is included here because it is commonly used with routing protocols to provide end-to-end data movement through an internetwork.
IS-IS routing protocols—IS-IS performs hierarchical (Level 1, Level 2, and Level 3) routing between intermediate systems. Level 3 routing is done between separate domains. However, note that the IS-IS routing protocol is not itself capable of Level 3 routing. As described in the section, "Interconnecting IS-IS Domains" later in this chapter, other protocols are required for interdomain routing.
The hierarchical routing levels used in OSI are illustrated in Figure S-2.
)
Figure S-2 OSI Hierarchical Routing
To simplify router design and operation, OSI distinguishes among Level 1, Level 2, and Level 3 routing. Level 1 ISs communicate with other Level 1 ISs in the same area. Level 2 ISs route between Level 1 areas and form an intradomain routing backbone. Level 3 routing is done between separate domains. Hierarchical routing simplifies backbone design because Level 1 ISs need to know only how to get to the nearest Level 2 IS.
In OSI, each ES lives in a particular area. OSI routing begins when the ESs discover the nearest IS by listening to intermediate system hello (ISH) packets. When an ES wants to send a packet to another ES, it sends the packet to an IS on its directly attached network; this is Level 0 routing. The router then looks up the destination address and forwards the packet along the best route. If the destination ES is on the same subnetwork, the local IS knows this from listening to end system hello (ESH) packets and forwards the packet appropriately. The IS also might provide a redirect message back to the source to tell it that a more direct route is available.
If the destination address is an ES on another subnetwork in the same area, the IS knows the correct route (Level 1 routing) and forwards the packet appropriately.
If the destination address is an ES in another area, the Level 1 IS sends the packet to the nearest Level 2 IS (Level 2 routing). Forwarding through Level 2 ISs continues until the packet reaches a Level 2 IS in the destination area. Within the destination area, ISs forward the packet along the best path until the destination ES is reached.
Routing between separate domains is referred to as Level 3 routing.
Routing in an OSI CLNS/CLNP Environment
For routing in the ISO CLNS/CLNP environment, Cisco routers support the following protocols:
IS-IS—IS-IS is a dynamic link-state routing protocol used in an ISO CLNS environment for routing CLNP. Routers usually operate as ISs and can exchange reachability information with other ISs using the IS-IS protocol. As an IS, a Cisco router can be a Level 1 router, a Level 2 router, or a Level 1–2 router. In the latter case, the router can advertise itself at Level 1 as an exit point from the area. Integrated IS-IS allows the IS-IS protocol to propagate routing information for other protocols as well as—or instead of—CLNS. Specifically, IS-IS can route CLNS, IP, or both (this latter is called dual mode).
ISO-IGRP—Cisco IOS Software offers a proprietary routing protocol for CLNS. As its name suggests, ISO-IGRP is based on Cisco's Interior Gateway Routing Protocol (IGRP). It uses distance vector technology to propagate routing information. As such, it shares some of the limitations of its IP counterpart, including long convergence times (because of periodic updates and long invalid times and hold times).
Static CLNS routes—As with IP, static CLNS routes can be created. (Although this is not really a protocol, static routes can be considered to be a type of routing protocol.)
Recommended Reading
Various aspects of IS-IS are described in these ISO documents:
ISO 8473—Documents the ISO CLNP
ISO/IEC 8348, Appendix A—Documents NSAP addresses
ISO 9542—Documents the ES-IS routing exchange protocol
ISO/IEC 10589—Documents the IS-IS intradomain routing exchange protocol
Additionally, the function of Integrated IS-IS—the use of OSI IS-IS for routing in TCP/IP and dual environments, as described in the next section—is described in RFC 1195. (RFC 1195 can be found at http://www.cis.ohio-state.edu/cgi-bin/rfc/rfc1195.html.)
Integrated IS-IS
As previously mentioned, IS-IS is the dynamic link-state routing protocol for the OSI protocol stack. As such, it distributes routing information for routing CLNP data for the ISO CLNS environment.
Integrated IS-IS is an implementation of the IS-IS protocol for routing multiple network protocols; it is an extended version of IS-IS for mixed ISO CLNS and IP environments. Integrated IS-IS tags CLNP routes with information regarding IP networks and subnets. It can be used purely for IP routing, purely for ISO routing, or for a combination of the two.
Integrated IS-IS provides an alternative to OSPF in an IP environment, mixing ISO CLNS and IP routing in one protocol.
NOTE
Like all modern IP routing protocols, Integrated IS-IS supports the following features:
Variable-length subnet masks (VLSMs). The mask and the prefix are sent in the routing updates.
Redistribution of IP routes into and out of IS-IS.
Summarization of IP routes.
Integrated IS-IS Versus OSPF
Integrated IS-IS and OSPF are both link-state protocols with similarities as follows:
Link-state representation, aging, and metrics
Link-state databases and SPF algorithms
Update, decision, and flooding processes
As discussed in Chapters 3 and 4 of Building Scalable Cisco Networks, OSPF is based on a central backbone (area 0), with all other areas ideally being physically attached to area 0. In OSPF, the border between areas is inside routers, the Area Border Routers (ABRs), and each link belongs to one area. This central backbone configuration means that certain design constraints will inevitably exist. When this type of hierarchical model is used, a good, consistent IP addressing structure is necessary to summarize addresses into the backbone and reduce the amount of information that is carried in the backbone and advertised across the network.
In comparison, IS-IS also has a hierarchy with Level 1 and Level 2 routers, but in IS-IS, the area borders lie on links rather than in routers. As shown in Figure S-2, each IS-IS router belongs to exactly one Level 2 area. Significantly fewer link-state packets (LSPs), also known as link-state protocol data units (PDUs), get used; thus, many more routers can reside in a single area. This capability makes IS-IS more scalable than OSPF. IS-IS allows a more flexible approach to extending the backbone by adding additional Level 2 routers. In IS-IS, this process is less complex than with OSPF.
Areas in OSPF and IS-IS
In OSPF, the area boundary is "inside" the router; interfaces belong to an area. In IS-IS, the "whole" router, not just an interface, belongs to an area.
With regard to CPU use and the processing of routing updates, IS-IS is more efficient than OSPF. In IS-IS, one LSP is sent per IS-IS router in each area (including redistributed prefixes [routes]), compared to the many OSPF LSAs that would be sent. Not only are there fewer LSPs to process, but the mechanism by which IS-IS installs and withdraws prefixes is less processor intensive.
Both OSPF and IS-IS are link-state protocols and thus provide fast convergence. The convergence time depends on a number of factors (timers, number of nodes, type of router, and so on).
Based on the default timers, IS-IS detects a failure quicker than OSPF and thus should converge more rapidly. Of course, if there are many neighbors and adjacencies to consider, the convergence time also might depend on the processing power of the router. IS-IS tends to be less CPU intensive than OSPF.
The timers in IS-IS allow for better fine-tuning than what is available for OSPF. There are other adjustable timers, so finer granularity can be achieved. By adjusting the timers, convergence time can be significantly decreased. However, this speed could come at the expense of stability. The network operator must understand the implications of changing the timers before making any adjustments.
Both IS-IS and OSPF are scalable, and the scalability of link-state protocols has been proven in the current live ISP backbones.
OSPF does have more features than IS-IS, including route tags, stub and not-so-stubby areas (NSSA), and OSPF on-demand circuits.
Level 1, Level 2, and Level 1–2 Routers
An IS-IS network is termed a domain; this is the equivalent of an autonomous system (AS) in OSPF. Within the domain, a two-level hierarchy exists:
Level 1 ISs are responsible for routing to ESs inside an area. This is similer to OSPF internal nonbackbone routers in a totally stubby area.
Level 2 ISs route between areas only. This is similer to internal backbone routers in OSPF.
Level 1–2 ISs route between areas and the backbone. They participate in the Level 1 intra-area routing and the Level 2 interarea routing. This is the equivalent of Area Border Routers (ABRs) in OSPF.
Level 1 routers are also referred to as station routers because they enable stations (ESs) to communicate with each other and the rest of the network.
NOTE
End stations don't communicate by way of CLNP.
A contiguous group of Level 1 routers defines an area. The Level 1 routers maintain the Level 1 database, which defines the picture of the area itself and its exit points to neighboring areas.
Level 2 routers are also referred to as area routers because they interconnect the Level 1 areas. Level 2 routers store a separate database that contains only the interarea topology information.
Level 1–2 routers keep two separate link-state databases; this allows them to act as if they were two IS-IS routers, as follows:
They support a Level 1 function to communicate with the other Level 1 routers in their area and maintain the Level 1 LSP information in a Level 1 topology database. They inform other Level 1 routers that they are an exit point from the area.
They support a Level 2 function to communicate with the rest of the backbone and maintain a Level 2 topology database separately from their Level 1 database.
IS-IS does not share the concept of an area 0 with OSPF. Instead, an IS-IS backbone can appear as a set of distinct areas interconnected by a chain of Level 2 routers, weaving their way through and between the Level 1 areas. The IS-IS backbone consists of a set of Level 1–2 routers and Level 2 routers, and must be contiguous.
IS-IS uses a two-level hierarchy. The link-state information for these two levels is distributed separately, giving rise to Level 1 LSPs and Level 2 LSPs.
LSPs on point-to-point links are sent to a unicast address. LSPs on broadcast media (LANs) are sent to a multicast address.
As with OSPF, one router on a LAN sends the LSP information on behalf of that LAN. In IS-IS, this router is called the designated intermediate system (DIS). It is a pseudonode, the representation of the LAN, and sends separate Level 1 and Level 2 LSPs on behalf of the network.
NOTE
No backup DIS exists in IS-IS, in contrast to the backup DR for OSPF. If a DIS dies, a new election takes place.
Examples of IS-IS Hierarchical Routing
Figure S-3 shows the physical view of an example IS-IS area configuration. Physically, a Level 1–2 router connects to Level 1 routers inside its area and to Level 2 routers in the backbone. In the figure, R2 and R3 are Level 1–2 routers; R1 and R4 are Level 1 routers. R2 and R3 belong to their respective Level 1 areas and provide a physical connection between them.
)
Figure S-3 Example 1 IS-IS Area Configuration—Physical View
NOTE
Recall that the boundary between areas in IS-IS exists on a link between routers and not on a router itself, as in OSPF.
Figure S-4 shows the logical view of the same example shown in Figure S-3. In Figure S-4, R2 and R3 are Level 1–2 routers; R1 and R4 are Level 1 routers. R2 and R3 are still Level 1 routers, but, in addition, they provide an entry point to the Level 2 backbone interconnecting both Level 1 areas.
)
Figure S-4 Example 1 IS-IS Area Configuration—Logical
View
Figure S-5 shows another example.
)
Figure S-5 Example 2 IS-IS Level 2 and Level 1–2 Routers Forming a
Level 2 Backbone
In Figure S-5, area 1 contains two routers:
One router borders to area 2 and, therefore, is a Level –1–2 IS.
The other router is contained totally within the area and, therefore, is Level 1 only.
Area 2 has many routers:
Some routers are specified as Level 1 only and can route internally to that area only (and to the exit points).
Level 1–2 routers form a chain across the area linking to the neighbor areas.
Although the middle of these three Level 1–2 routers does not link directly to another area, it must support Level 2 routing so that the backbone is contiguous. If this middle router fails, the other Level 1–only routers (though providing a physical path across the area) could not perform the Level 2 function, and the backbone would be broken.
Area 3 contains one router that borders to area 3, but it has no intra-area neighbors and is, therefore, Level 2 only. If another router was added to area 3, the border router would revert to Level 1–2.
Figure S-5 also shows that the border between the areas in an IS-IS network exists on the links between Level 2 routers (in contrast to OSPF, where the border exists inside the Area Border Router [ABR] itself).