# IS-IS multi-area

It is well known that an IS-IS L1 router chooses the nearest L1/L2 router to route inter-area traffic, if there is no route leaking as defined in RFC 2966. However, providing that there is no route leaking, does the L1/L2 router of the source area sends the traffic to the nearest L1/L2 router of the destination area, or does it have enough information about the topology of the destination area to choose the best L1/L2 router of the destination (i.e to find the full shortest path to the final destination)?

Edit: to be precise in my question, I'm considering the case where the destination area has several L1/L2 routers. The nearest L1/L2 router of the destination area may not be the one on the shortest path from the L1/L2 router of the source area, to the final destination (in the destination area).

My question is: with IS-IS, which L1/L2 router of the destination area would the L1/L2 router of the source area send packets to?

Ron's response does not answer that question, since my question doesn't apply to a network with a single L1/L2 router connecting the destination area to the backbone.

Edit 2:

Here's the diagram. Once you reach the L1/L2 router of the first area, you have two path. You can send the traffic to the closest L1/L2 router, but the resulting path will have a cost of 1001. But if you know the topology of the destination area, you'll send the traffic to the other L1/L2, and the path has a cost of 11. What I wonder is how IS-IS behaves in such a case.

• My answer does answer the question. All traffic from one are to another area must travel through the backbone. The closest L1/L2 router is the shortest path to the backbone, and the traffic will travel through the backbone with the shortest path. You should edit your question to include a diagram of what you are thinking, but you are probably overthinking this. Jun 9, 2016 at 17:07
• Your understanding is a little flawed. The first L1/L2 router has two paths to the destination: cost 11 or cost 1001. It will take the lowest cost path: 11. You don't seem to understand that once the traffic from an area reaches the nearest L1/L2 router, it no longer looks for an L1/L2 router. It is in the backbone (L2). It will take the path to the L1/L2 of the destination which is the shortest path. That would be the top L1/L2 router. The source area may not know how to get to the destination, but the L2 routers do, and they will take the shortest path. Jun 10, 2016 at 13:50

Reference to this answer is not easy to find in any vendor/training materials, but lies in RFC1195:

Level 2 routers include in their level 2 LSPs a list of all [IP address, subnet mask, metric] combinations reachable in their area. In general, this information may be determined from the level 1 LSPs from all routers in the area. If we ignore resource constraints, then it would be permissible for a level 2 router to simply duplicate all [IP address, subnet mask, metric] entries from all level 1 routers in its area (with appropriate metric adjustment), for inclusion in its level 2 LSP.

Any address obtained from a level 1 LSP which is not superceded by the manually configured information is included in the level 2 LSPs. In this case, the metric value announced in the level 2 LSPs is calculated from the sum of the metric value announced in the corresponding level 1 LSP, plus the distance from the level 2 router to the appropriate level 1 router.

In general, the same [IP address, subnet mask] pair may be announced in level 1 LSPs sent by multiple level 1 routers in the same area. In this case (assuming the entry is not superceded by a manually configured entry), then only one such entry shall be included in the level 2 LSP. The metric value(s) announced in Level 2 LSPs correspond to the minimum of the metric value(s) that would be calculated for each of the level 1 LSP entries.

So to focus on your example.

Area 1 L1/L2 router knows the Area 1 topology and Level 2 topology, but does not know about the the Area 2 topology.

Area 2 L1/L2 routers know Area 2 topology. They perform SPF calculation on their Area 2 topology information to calculate the cost to the destination - 1 and 1000.

Area 2 L1/L2 routers advertise the destination network to other Level 2 routers (namely Area 1 L1/L2 router) with the calculated cost (1 and 1000). (but they do not advertise the topology of the Area 2 itself to the Level 2 routers!)

Area 1 L1/L2 router receives these LSPs. It knows that destination is reachable from northern Area 2 L1/L2 router with cost 1 and from southern Area 2 L1/L2 router with cost 1000. Area 1 L1/L2 router adds that to its SPF calculation for Level 2 topology and calculates the total cost to reach the destination - 11 through the northern Area 2 L1/L2 router and 1001 through the southern Area 2 L1/L2 router.

To answer the original question: L1/L2 router of the source area does not send the traffic to the nearest L1/L2 router of the destination area, it has enough information about the cost from destination area L1/L2 routers to the destination (but not the topology of the destination area) to choose the best L1/L2 router of the destination area (i.e to find the full shortest path to the final destination).

IS-IS uses Dijkstra's algorithm, like OSPF, to route through the network. The L2 routers form the backbone. Basically, you can think of the collection of L2 routers as OSPF area 0. L1/L2 routers are ABRs, with one foot in an area, and another foot in the backbone.

Once a packet has reached an L2 router (including an L1/L2 router), the packet will be routed through the L2 routers via the shortest path to the L1/L2 router of the destination.