This chapter contains basic IS-IS (Intermediate System to Intermediate System) configuration examples.

This example shows the minimum configuration required for enabling IS-IS on an interface. R1 and R2 are two routers in the ABC instance connecting to the network 10.10.10.0/24. After enabling IS-IS on an interface, create a routing instance, and specify the Network Entity Title (NET). IS-IS explicitly specifies a NET to begin routing. NET is comprised of the area address and the system ID of the router.

This example describes how to set the priority for an interface. Set a high priority for a router to make it the Designated IS (DIS). Router R2 is configured to have a priority of 125, this is higher than the default priority (64) of R1. This makes R2 the DIS..

This example shows how to configure Dynamic Hostname for an ISIS instance. Dynamic hostname is the method of mapping name-to-systemID. It allows the routing protocol to advertise symbolic names in the IS-IS PDUs. This is done by the addition of a new TLV which allows the IS-IS routers to include the name-to-systemID mapping data in their LSPs. This allows for simple and reliable transport of name mapping across IS-IS networks. Dynamic hostname can be either the hostname of the node or the tag of the configured ISIS instance.

Note: Dynamic-hostname has to be configured on all nodes for it to take effect.

 

In this example, the configuration causes OSPF routes to be imported into the IS-IS routing table, and advertised into the ABC instance.

You can make a route the preferred route by changing its metric. In this example, the cost has been configured to make R3 the next hop for R1.

The default metric for each interface is 10. Interface eth3 on R2 has a metric of 20, and Interface eth2 on R3 has a metric of 30. The total cost to reach 10.10.14.0/24 (R4) through R2 and R3 is computed as follows:

R2: 10+20 = 30

R3: 10+30 = 40

In this topology, R1 chooses R2 as its next hop for destination 10.10.14.0/24.

Note: Below configuration is applicable for narrow (non-wide) metric-style. Wide metric can be configured by using the CLI's "metric-style wide" under isis instance and "isis wide-metric < 1-16777214>" under interface mode.

IS-IS supports a two-level hierarchy for handling and scaling the functionality of large networks. The Level-1 (L1) area is mainly for Leaf networks, and the Level-2 (L2) area is the backbone area connecting Level-1 areas. In this example, R3 and R4 are configured as Level-1 routers, and reside in the Level-1 area. R1 and R2 are configured as Level-1-2 routers, and connect these two Level-1 areas with a backbone Level-2 area. You can configure Level-1-2 routers with single or multiple instances: This configuration shows the single-instance version of the Level-1-2 router.

IS-IS supports a two-level hierarchy for handling and scaling the functionality of large networks. The Level-1 (L1) area is mainly for Leaf networks, and the Level-2 (L2) area is the backbone area connecting Level-1 areas. In this example, R3 and R4 are configured as Level-1 routers, and reside in the Level-1 area. R1 and R2 are configured as Level-1-2 routers, and connect these two Level-1 areas with a backbone Level-2 area. You can configure Level-1-2 routers with single or multiple instances: This configuration shows the multiple-instance version of the Level-1-2 router.

Route leaking is defined in RFC 2966. For Level-1 (L1) routers, only level-1 routes are populated in the routing table. The L1 router has a default route to the nearest Level-1/Level-2 (L1/L2) router: This could result in sub-optimal routing in certain scenarios. Route leaking causes an L1/L2 router to advertise the level-2 routes in its database to the L1 router, thus allowing the L1 router to acknowledge the prefixes advertised by the Level-2 (L2) router. In this way, the L1 router has the ability to learn the true cost to reach other areas.

In the following example, R1 is the L1 router, R2 is the L1/L2 router doing the route leaking, and R3 is the L2 router. The following configuration is given only for R2, assuming that the adjacency with R1 and R3 are already up, and the route tables with appropriate routes are already populated.

 

 

 

 

In the example, route, i ia 3.3.3.3/32 [115/30] via 20.20.20.2, eth1, 00:12:29, is the L2 route leaked by the L1/L2 router into the L1 router.

Route summarization makes the routing table smaller, but still allows complete IP connectivity, if everything is configured properly.

The following example consists of a three-router topology, in which R2 is doing the summarization. In this example, R1 is the L1 router, R2 is the L1/L2 router doing the summarization, and R3 is the L2 router. The following configuration is given only for R2, assuming that the adjacencies with R1 and R3 are already up, and the route tables with the appropriate routes are already populated.

 

 

Administrative distance in IS-IS can be configured for a specified source ID or for all routes.

This example shows configuring the IS-IS administrative distance for the IPv4 address family.

This section provides examples of configuring the IS-IS overload bit. For detailed information about the commands used in these examples, refer to the Intermediate System to Intermediate System Command Reference.

The expanded use of the overload bit in IS-IS proves beneficial for Internet Service Providers (ISPs) operating both BGP and IS-IS, helping them avoid specific routing anomalies. With the overload bit feature, a router can automatically disable the overload bit when BGP achieves convergence.

IS-IS will deactivate the overload bit if BGP fails to converge within ten minutes and the wait-for-bgp option is set in the configuration.

When configure the set overload-bit wait-for-bgp after a restart (e.g., using commands like set-overload-bit on-startup wait-for-bgp or set-overload-bit on-startup <5-86400>), the IS-IS router broadcasts the overload bit status to all its neighbors. This informs them not to route traffic through this router, except for traffic destined for networks directly connected to it.

If an interface is configured with the no set-overload-bit command, the router will not advertise or set the overload bit during startup.

Figure 7-51 represents a sample topology with four routers (R1, R2, R3, and R4) interconnected in a linear sequence.

In this network configuration scenario, these routers collaborate to create a dynamic and resilient routing environment. Through the implementation of both BGP and IS-IS routing protocols, these routers establish connections, exchange routing information, and optimize network performance. Each router plays a unique role in this network, with specific configurations tailored to ensure smooth communication and fault tolerance.

The key element in these configurations is on R3 and R4, where the set-overload-bit on-startup wait-for-bgp commands are used. By configuring R3 and R4 with these commands, IS-IS on R3 and R4 will advertise the overload bit to its neighbors, effectively informing them not to route traffic through R3 and R4 until BGP has converged. This behavior helps prevent routing anomalies during network restarts and enhances network stability.

The control mechanism of suppressing the redistribution of external, interlevel, or both external and interlevel reachabilities is useful for managing the advertisement of these specific reachability data during overload states, providing flexibility in network configurations.

Figure 7-52 represents a sample topology with three routers (R1, R2, and R3) interconnected in a linear sequence.

The key element in the below configurations is on R3 router, where the set-overload-bit suppress external and set-overload-bit suppress interlevel commands are used.

Configuring the R3 router with the set-overload-bit suppress external or set-overload-bit suppress interlevel commands in IS-IS prevents the advertisement of external routes learned from BGP to its ISIS neighbors or prevents the advertisement of interlevel routes to its ISIS neighbors. The options suppress external or suppress interlevel contribute to the controlled management of routing information during network overload states.

Verify all IS-IS sessions are up.

Before configuring set-overload-bit suppress external command, verify the ISIS session status on R1, R2, and R3 routers.

Before configuring set-overload-bit suppress interlevel command, verify the ISIS session status on R1, R2, and R3 routers.

After configuring set-overload-bit suppress external command, verify the ISIS session status on R1, R2, and R3 routers.

After configuring set-overload-bit suppress interlevel command, verify the ISIS session status on R1, R2, and R3 routers.

In ISP and large enterprise networks, many of the distribution routers have more than 200 interfaces. Before the Default Passive-Interface feature, there were two possibilities for obtaining routing information from all of these interfaces:

The solution to this problem was to configure the routing protocol on all interfaces and manually set the passive-interface command on the interfaces where adjacency was not desired. In certain networks, this meant coding 200 or more passive-interface statements. With the Default Passive Interface feature, this problem is solved by allowing all interfaces to be set as passive by default using a single passive-interface default command, then configuring individual interfaces in which adjacencies are desired using the passive-interface <interface-name> disable command.

Figure 7-53 shows a passive-interface configuration example.

This section contains IS-IS (Intermediate System to Intermediate System) Loop-Free Alternate Fast Reroute (LFA-FRR) configuration examples.

For details about the commands used in these examples, see the Intermediate System to Intermediate System Command Reference.

OSPF LFA and ISIS LFA along with MPLS is not supported. Do not configure OSPF LFA or ISIS LFA, if MPLS is configured or vice-versa.

When a primary next-hop fails, LFA-FRR reduces the failure reaction time to tens of milliseconds using a pre-computed alternate next- hop, so that the alternate can be rapidly used when the failure is detected. A network with this feature experiences less traffic loss and less micro-looping of packets than a network without LFA-FRR.

After you enable LFA-FRR, routers calculate a backup path for each primary path to reach the destination. The backup path is calculated based on the attributes such as node protecting, link protecting, and broadcast link protecting. If there is an ECMP path to reach prefixes, the backup is selected from the same primary set by default; if a secondary tie-breaker is enabled, and if a secondary path is available, the backup will be selected from the secondary path.

Figure 7-54 shows the configuration to enable the basic ISIS LFA feature.