7.3 IS-IS Protocol Architecture and Operations
Key Takeaways
IS-IS is a link-state IGP whose control PDUs use link-layer encapsulation.
Level 1 handles intra-area routing; Level 2 forms the inter-area backbone.
A conventional NET contains an area portion, six-byte System ID, and trailing selector 00.
DIS election is preemptive and has no backup DIS; priority zero is not OSPF-style ineligibility.
This comparison supports the current exam; it is not complete preparation for the March 2027 revision.
Supporting context: identify IS-IS correctly
Intermediate System to Intermediate System (IS-IS) is a link-state IGP used in many provider networks. This lesson supports protocol comparison. The current 4A0-100 course explicitly teaches OSPF basics; Nokia announces replacing OSPF with IS-IS in the March 2027 revision. This lesson is not a complete guide to that future exam.
Integrated IS-IS extends the original OSI routing protocol to advertise IP reachability. Its control PDUs are carried directly over the link layer rather than inside an IP packet. That transport choice does not make it a bridging protocol: its purpose is still network-layer route calculation.
IS-IS uses topology descriptions, databases, and SPF calculations. As with OSPF, routers can share converged topology information while calculating different next hops from their own positions.
Encapsulation and security
On Ethernet, conventional IS-IS control PDUs use LLC encapsulation with the relevant ISO protocol identification. There is no IS-IS TCP or UDP destination port. The protocol can establish link-layer relationships without requiring the same IP-encapsulation process as OSPF.
This reduces exposure to some directly IP-addressed attacks, but it does not make a router immune to denial of service. A malicious or misconfigured neighbor can still send link-local control traffic, and other traffic can consume shared resources. Authentication, control-plane protection, and appropriate access boundaries remain important.
Distinguish control-packet transport from payload forwarding. A router can advertise IPv4 and IPv6 reachability using IS-IS while ordinary customer packets remain IPv4 or IPv6 packets, or are transported inside MPLS services.
Level 1 and Level 2
| Router capability | Main function |
|---|---|
| Level 1 | Routing within an area |
| Level 2 | Routing in the inter-area backbone |
| Level 1-2 | Participates in both levels with separate databases |
An IS-IS area's identity is associated with the router's area address rather than assigning a different OSPF-like area to each interface. A Level 2 adjacency can connect routers belonging to different areas. A Level 1 relationship requires compatible area information.
Suppose R1 is Level 1 in Area A, R2 is Level 1-2 in Area A, R3 is Level 1-2 in Area B, and R4 is Level 1 in Area B. R1 can reach R2 at Level 1; R2 and R3 can connect the backbone at Level 2; R3 can reach R4 at Level 1. A pure Level 1 connection between R1 and R4 would not supply the intended inter-area design.
A Level 1-2 router can advertise attachment toward the backbone using the attached bit under applicable conditions. Level 1 routers can use a route toward a suitable attached router for destinations not known locally. Route leaking and policy can change the information available, so “Level 1 routers never know any outside prefix” is too absolute.
NET addresses
A Network Entity Title (NET) identifies the network entity. In the conventional IP routing form, it includes an area portion, a six-byte System ID, and a one-byte selector set to 00.
For example:
49.0001.0100.0100.1001.00
| area | System ID | selector
The example uses area portion 49.0001 and System ID 0100.0100.1001. The six-byte System ID must be unique within the routing domain. The trailing 00 identifies the network entity rather than another service user.
One human-readable convention derives an identifier from 10.1.1.1 by padding its octets to 010.001.001.001 and regrouping the digits into 0100.0100.1001. This is an administrative naming convention, not an IP-to-hex arithmetic conversion or a protocol requirement. Configured identifiers must actually remain unique.
DIS and LAN topology
An IS-IS broadcast LAN elects a Designated Intermediate System (DIS). The DIS represents the LAN using a pseudonode LSP and helps database synchronization. Unlike OSPF's DR/BDR arrangement, IS-IS does not elect a backup DIS.
DIS election is preemptive. A better-priority router can replace the incumbent, with the applicable interface address tie-break determining equal-priority outcomes. The LAN priority field uses values 0 through 127. A priority of zero is not the OSPF-style prohibition on becoming DIS.
IS-IS LAN routers maintain adjacencies with other eligible routers, rather than using OSPF's rule that DROthers remain only 2-Way with one another. Do not transfer neighbor-state names directly from OSPF to IS-IS displays.
TLVs and metrics
Type-Length-Value (TLV) encoding carries extensible information such as neighbor and prefix reachability. An implementation can recognize supported TLVs and handle unknown ones according to protocol rules. Extensibility does not mean new features require no implementation work: a router still needs software support to interpret and use new information.
Wide metrics provide a larger range than legacy narrow metrics. The configured metric participates in SPF; the protocol does not inherently measure current delay or congestion. As with OSPF, an operator must choose metrics and policy appropriate to the design.
For comparisons, retain the useful distinctions: OSPFv2 is IP-encapsulated and uses areas and a non-preemptive DR/BDR election; IS-IS uses link-layer control PDUs, levels, NET/System ID addressing, and a preemptive DIS election.
Primary references
Integrated IS-IS specification and Nokia's announced NRS I transition, checked October 20, 2026.
On an Ethernet broadcast multi-access network, how does the IS-IS Designated Intermediate System (DIS) election mechanism fundamentally differ from the OSPF Designated Router (DR) election?
IS-IS elects both a DIS and a Backup DIS (BDIS) using a non-preemptive election, while OSPF elects only a DR with preemptive rules
IS-IS routers on a LAN only form adjacencies with the DIS, whereas OSPF DROthers form full adjacencies with all routers on the segment
The IS-IS DIS is determined by the lowest System ID, whereas the OSPF DR is determined by the lowest interface IP address
The IS-IS DIS election is fully preemptive and has no Backup DIS, whereas the OSPF DR election is non-preemptive and accompanied by a Backup Designated Router (BDR)
A network engineer configures the Network Entity Title (NET) address 49.0100.0100.0000.0001.00 on a Nokia 7750 SR router for IS-IS. What does the final trailing byte value '00' represent in this address structure?
The N-Selector (NSEL) byte, which must always be 00 to designate the router itself as an intermediate system
The Level 1 router priority byte used during DIS election
The OSPF area compatibility indicator byte for multi-protocol redistribution
The broadcast subnet mask identifier corresponding to a 32-bit host loopback
Which statement correctly contrasts ordinary IS-IS and OSPF area organization?
IS-IS uses OSPF ABRs and requires Area 0
IS-IS has no area hierarchy
IS-IS associates an area with the router and can connect different areas through Level 2 relationships; OSPF assigns areas to interfaces
OSPF requires the same area on every interface of every router
Sections you finish are checked off in the contents.