IS-IS
Course Description
IS-IS (Intermediate System to Intermediate System) is a link-state routing protocol, similar to OSPF. It was originally developed for CLNS, not IP, and later adapted to route IP as integrated IS-IS. Because it scales so well, you’ll find it in the backbone of many large service provider networks. In this course, you’ll learn how IS-IS works, how it is different from OSPF, and how to configure it on Cisco IOS routers. We start with the basics and then move on to topics like the DIS and pseudonode, metrics, summarization, filtering, redistribution, and route leaking.
Course Highlights
In this course you will learn:
- How IS-IS works and how it is different from OSPF.
- What level 1, level 2, and level 1-2 routers are and how IS-IS uses areas.
- How to build a NET (Network Entity Title) for each router.
- How to configure integrated IS-IS on Cisco IOS routers and read the link-state database.
- How to secure IS-IS with plain text and HMAC-MD5 authentication.
- How the DIS and pseudonode work on multi-access networks.
- How IS-IS metrics work, and why you might want wide metrics.
- How to redistribute, summarize, and filter IS-IS routes, and how route leaking prevents sub-optimal routing.
Presented to you by instructor Rene Molenaar, CCIE #41726
Course at a glance
- 9 lessons.
- Format: self-paced lessons with Cisco IOS configurations and packet captures.
- Last updated: October 2026.
Requirements
- A good understanding of OSPF; we often compare IS-IS with OSPF.
- Basic knowledge of IP addressing and how routers forward packets.
Course Schedule
What is IS-IS?
IS-IS (Intermediate System to Intermediate System) is a link-state interior gateway protocol (IGP). Just like OSPF, IS-IS routers send hello packets to form neighbor adjacencies, flood LSPs (Link State Packets) with their prefixes, neighbors, and metrics, build a link-state database, and run Dijkstra’s SPF algorithm to find the best path to each destination. The name comes from ISO terminology, where a router is called an intermediate system.
IS-IS does a few things differently. It runs directly on top of the data link layer instead of IP, and each router needs a NET (Network Entity Title) in the NSAP format, which holds the area number and a unique system ID. The entire router sits in an area, not individual interfaces, and there is no backbone area: a continuous string of level 2 routers forms the backbone. Level 1 routers only know their own area and use a default route to the closest level 1-2 router to leave it.
Who uses IS-IS? It’s highly scalable, so it’s popular in the backbone of large service provider networks, often together with MPLS and BGP. Because IS-IS doesn’t depend on IP for its own packets, it carries both IPv4 and IPv6 prefixes. If you already know OSPF, you’ll find many familiar concepts, with different names.
IS-IS Learning Roadmap
This roadmap follows the course units. Each topic has a short explanation and links to the lessons where we cover it in detail.
IS-IS Fundamentals
We start with the building blocks of IS-IS: level 1, level 2, and level 1-2 routers, LSPs, and the link-state databases. Routers only form neighbor adjacencies with routers of the same level, and the default on Cisco IOS is level 1-2. Each router requires a NET, and AFI 49 for private networks is the most common choice. Once you understand this, we configure integrated IS-IS in a network with two areas and look at the databases.
IS-IS on Multi-Access Networks and Authentication
On a LAN, IS-IS elects a DIS (Designated IS) based on the interface priority and the MAC address. There is no backup DIS, and the election is preemptive. The DIS creates a pseudonode that simplifies the topology, and it sends a CSNP every 10 seconds so routers know their LSPs were received. To protect IS-IS, we configure plain text or HMAC-MD5 authentication for hello packets, LSPs, and SNPs.
IS-IS Metrics
Cisco IOS only supports the IS-IS default metric, and every interface gets a metric of 10, no matter its bandwidth. A slow link and a fast link look the same, so IS-IS picks the path with the fewest hops until you change the metric yourself. Narrow metrics limit an interface to a metric of 63, which is why we enable wide metrics. The metric maximum command keeps transit traffic away from a router.
IS-IS Redistribution, Summarization and Filtering
When you redistribute routes into IS-IS, you choose whether they go into the level 1 LSP, the level 2 LSP, or both. Because all routers in an area need the same database, you can only summarize at the border: on a level 1-2 router or a router that redistributes. Filtering works the same way: a distribute-list keeps a prefix out of the routing table but not out of the database, and you can stop level 1 prefixes from being copied into level 2.
IS-IS Route Leaking
Level 1 routers use a default route to the closest level 1-2 router to reach other areas, which can result in sub-optimal routing. With route leaking, a level 1-2 router advertises selected level 2 prefixes into level 1, so level 1 routers can pick the best exit. The up/down bit makes sure another level 1-2 router doesn’t advertise these prefixes back into level 2.
Frequently Asked Questions
IS-IS (Intermediate System to Intermediate System) is a link-state routing protocol, similar to OSPF. Routers flood LSPs, build a link-state database, and run SPF to find the shortest path to each destination. It was developed for CLNS and later adapted for IP, which is called integrated IS-IS. Read our introduction to IS-IS.
Both are link-state protocols that use SPF, but there are some differences. IS-IS runs directly on the data link layer instead of IP, a whole router sits in an area instead of its interfaces, and there is no backbone area 0. IS-IS uses level 1 and level 2 routers instead of ABRs, and a DIS with a pseudonode instead of a DR and BDR. Learn more about OSPF.
A level 1 router is an intra-area router that only knows its own area. A level 2 router is a backbone router that knows the routes of all areas. A level 1-2 router does both and keeps a separate database for each level; this is the default on Cisco IOS routers.
The NET (Network Entity Title) identifies an IS-IS router. It uses the NSAP format and contains the area number, a system ID that is unique for each router, and an NSEL that is always 0. On private networks, we usually start the NET with AFI 49, for example 49.0012.0000.0000.0001.00. Learn how to configure integrated IS-IS.
The DIS (Designated Intermediate System) is the IS-IS router that is elected on a multi-access network, based on interface priority and then the highest MAC address. It creates a pseudonode and sends CSNPs so routers can check their LSPs. Unlike OSPF, there is no backup DIS, the election is preemptive, and all routers on the LAN still become neighbors with each other. Learn how the DIS and pseudonode work.
On Cisco IOS, every IS-IS interface has a default metric of 10, no matter the bandwidth. With the default narrow metrics, the highest interface metric is 63 and the highest total metric is 1023. Wide metrics allow much higher values. Learn how IS-IS metrics work.
Route leaking means a level 1-2 router advertises prefixes from its level 2 database into level 1. Without it, level 1 routers only have a default route to the closest level 1-2 router, which is not always the best path to a destination in another area. Learn how to configure route leaking.
IS-IS is highly scalable and runs directly on the data link layer, so it doesn’t depend on IP for its own packets. That makes it a good fit for large backbones that carry both IPv4 and IPv6 and run MPLS and BGP on top. Learn more about MPLS.
Ready to master IS-IS?
Get access to all 9 IS-IS lessons and every other course on NetworkLessons.com.