Supported IS-IS features

IS-IS supports multiple instances and processes. Multiple processes allow an IS-IS process to work in concert with a group of interfaces. A router can run multiple IS-IS processes, and each process corresponds to a unique group of interfaces.

For routers supporting VPN, each IS-IS process is associated with a VPN instance. The VPN instance is also associated with interfaces of the process.

Graceful Restart (GR) ensures the continuity of packet forwarding when a routing protocol restarts or an active/standby switchover occurs:

• GR Restarter—Graceful restarting router. It must be GR capable.

• GR Helper—A neighbor of the GR Restarter. It helps the GR Restarter to complete the GR process.

After an IS-IS GR Restarter restarts, it must complete the following tasks to synchronize the LSDB with its neighbors:

• Obtain IS-IS neighbor information without changing adjacencies.

• Obtain the LSDB.

The GR Restarter sends an OSPF GR signal to GR Helpers so that the GR Helpers keep their adjacencies with the GR Restarter, and restores the neighbor table after receiving responses from neighbors. The GR Restarter then synchronizes the LSDB with all GR capable neighbors, calculates routes, updates its routing table and forwarding table, and removes stale routes. The IS-IS routing convergence is then complete.

Nonstop routing (NSR) is a new feature that overcomes the application limit of GR. It backs up IS-IS link state information from the master device to the slave device. When a master/slave switchover occurs, NSR can complete link state recovery and route re-generation without requiring the cooperation of other devices.

Management tag simplifies routing information management by carrying the management information of the IP address prefixes (to control route redistribution from other routing protocols) and BGP community and extended community attributes.

IS-IS advertises link state information by flooding LSPs. Because one LSP carries a limited amount of link state information, IS-IS fragments LSPs. Each LSP fragment is uniquely identified by a combination of the System ID, Pseudonode ID (0 for a common LSP or a non-zero value for a Pseudonode LSP), and LSP Number (LSP fragment number) of the node or pseudo node that generated the LSP. The one-byte LSP Number field, allowing a maximum of only 256 fragments to be generated by an IS-IS router, limits the amount of link information the IS-IS router can advertise.

The LSP fragment extension feature allows an IS-IS router to generate more LSP fragments. Up to 50 additional virtual systems can be configured on the router, and each virtual system is capable of generating 256 LSP fragments to enable the IS-IS router to generate up to 13056 LSP fragments.

• Terms

  • Originating system—It is the router actually running IS-IS. After LSP fragment extension is enabled, additional virtual systems can be configured for the router. Originating system is the actual IS-IS process that originally runs.

  • System ID—System ID of the originating system

  • Additional system ID—Additional virtual system IDs are configured for the IS-IS router after LSP fragment extension is enabled. Each additional system ID can generate 256 LSP fragments. Both the additional system ID and the system ID must be unique in the entire routing domain.

  • Virtual system—A virtual system is identified by an additional system ID and generates extended LSP fragments.

  • Original LSP—The LSP generated by the originating system. The system ID in its LSP ID field is the system ID of the originating system.

  • Extended LSP—Extended LSPs are generated by virtual systems. The system ID in its LSP ID field is the virtual system ID.

    After additional system IDs are configured, an IS-IS router can advertise more link state information in extended LSP fragments. Each virtual system can be considered a virtual router. An extended LSP fragment is advertised by a virtual system identified by an additional system ID.

• Operation modes:

  The LSP fragment extension feature operates in the following modes:

  • Mode-1—Applicable to a network where some routers do not support LSP fragment extension. In this mode, adjacencies are formed between the originating system and virtual systems, with the link cost from the originating system to each virtual system as 0. Each virtual system acts as a router connected to the originating system in the network, but the virtual systems are reachable through the originating system only. The IS-IS routers not supporting LSP fragment extension can operate properly without modifying the extended LSP fragments received, but some limitation is imposed on the link state information in the extended LSP fragments advertised by the virtual systems.

  • Mode-2—Applicable to a network where all the routers support LSP fragment extension. In this mode, all the IS-IS routers know which virtual system belongs to which originating system; no limitation is imposed on the link state information of the extended LSP fragments advertised by the virtual systems.

  The operation mode of LSP fragment extension is configured based on area and routing level. Mode-1 allows the routers supporting and not supporting LSP fragment extension to interoperate with each other, but it restricts the link state information in the extended fragments. Mode-2 does not restrict the link state information in the extended fragments, and is recommended for an area where all the routers are at the same routing level and support LSP fragment extension.

The dynamic host name mapping mechanism provides the mappings between the host names and the system IDs for the IS-IS routers. The dynamic host name information is announced in the dynamic host name CLV of an LSP.

This mechanism also provides the mapping between a host name and the DIS of a broadcast network, which is announced in the dynamic host name TLV of a pseudonode LSP.

A host name is easier to remember than a system ID. After enabling this feature on the router, you can see the host names instead of system IDs using the display command.

Bidirectional forwarding detection (BFD) provides a single mechanism to quickly detect any link failures between IS-IS neighbors to reduce network convergence time. For more information about BFD, see High Availability Configuration Guide.