Forwarding Adjacency

Forwarding adjacency is an MPLS TE feature whereby the IGP can see a TE LSP as a link. The IGP on the head end router of the TE tunnel advertises the TE LSP as a link with a certain IGP metric associated with it. Any router in the same area as the head end router then includes this link when it is running the SPF algorithm. As such, the IGP sees the total path that a TE tunnel spans as one link only. To correctly use forwarding adjacency, you must configure two TE tunnels between a pair of LSRs—one for each direction—and you must have forwarding adjacency enabled for both. The reason for this is that the link is advertised as soon as you configure forwarding adjacency on the tunnel in one direction, but the forwarding adjacency link is included in the SPF calculation of the routers only when they can see the link in both directions (this means when the forwarding adjacency is enabled on both tunnels). You can locate the pair of LSRs with the two TE tunnels between them anywhere in the MPLS TE-enabled area. The LSRs can be several hops from each other. This is the great advantage of forwarding adjacency: A pair of TE tunnels can be seen as one link and advertised as such into the IGP. This also means that routers that are not running MPLS TE but are running the IGP can see that link and consider it when running the SPF algorithm.

Figure 8-21 autoroute announce on Several TE Tunnels

TE Tunnel 1

Because the pair of TE tunnels is advertised into the IGP as one link, it can be seen in the topology database of OSPF or IS-IS. Forwarding adjacency is enabled on the tunnel interface on the head end router with the following command:

tunnel mpls traffic-eng forwarding-adjacency {holdtime value}

The holdtime value is the time in milliseconds that the router must wait to flood, after the TE LSP has gone down. Example 8-24 shows the configuration of a TE tunnel, whereby the TE tunnel is announced as a link into OSPF. Note that IS-IS also supports forwarding adjacency. On the tail end router (, you must configure a TE tunnel with this router (paris) as the tail end router and with forwarding adjacency enabled. You can see that the TE tunnel is announced as a link in OSPF. It is announced in the router LSA of the router paris, as the link with Router Interface address This is obviously not the real IP address. Because the TE tunnel is unnumbered, OSPF cannot use the interface IP address. Instead, OSPF uses the If Index to uniquely identify the interface. OSPF identifies other unnumbered links in the same way.

Example 8-24 Configuration of Forwarding Adjacency paris#

interface Tunnell ip unnumbered Loopback0 ip ospf cost 1

tunnel destination

tunnel mode mpls traffic-eng tunnel mpls traffic-eng forwarding-adjacency tunnel mpls traffic-eng priority 7 7

tunnel mpls traffic-eng bandwidth 155000

paris#show mpls traffic-eng forwarding-adjacency destination, area ospf 1 area 0, has 1 tunnels Tunnell (load balancing metric 12903, nexthop (flags: Forward-Adjacency, holdtime 0)

paris#show ip ospf database router adv-router

OSPF Router with ID ( (Process ID 1)

Router Link States (Area 0)

Options: (No TOS-capability, DC) LS Type: Router Links Link State ID: Advertising Router: LS Seq Number: 80000039 Checksum: 0x5DD5

Example 8-24 Configuration of Forwarding Adjacency (Continued)

Length: 108 Number of Links: 7

Link connected to: another Router (point-to-point) (Link ID) Neighboring Router ID: (Link Data) Router Interface address: Number of TOS metrics: 0 TOS 0 Metrics: 1

Figure 8-22 shows an example where forwarding adjacency is useful.

Figure 8-22 Forwarding Adjacency Example

TE Tunnels With

Figure 8-22 Forwarding Adjacency Example

TE Tunnels With

Forwarding Adjacency

The traffic from R1 to R7—from the london site to the sydney site—takes the path R1-R2-R3-R7 because it is the shortest. If a TE tunnel were enabled from R2 to R3 and another from R4 to R6, the traffic from R1 to R7 would take the top TE tunnel, no matter how much you make the bottom TE tunnel more preferred over the top one. That is because the IP traffic can only be directed onto the TE tunnel on the head end router. The head end routers are R2 and R4 here. All other routers are unaware that the tunnel exists because it is not announced as a link in the IGP. Therefore, when R1 sends traffic to R7, it forwards the traffic to R2, which then forwards the traffic across the top TE tunnel to R3.

When forwarding adjacency is enabled on routers R2, R4, R3, and R6, the TE tunnels appear as a link on those four routers. This link is advertised via OSPF or IS-IS and is known to the routers R1 and R7. R1 includes the two links corresponding to the two TE tunnels in the SPF calculation when building its routing table. If the bottom forwarding adjacency link has the lowest IGP cost, R1 forwards traffic to R4 and over the bottom path toward R7. If the two forwarding adjacency links have the same IGP cost, R1 can even load-balance the traffic over the paths R1-R2-R3-R7 and R1-R4-R5-R6-R7. The result is that the bottom path is now utilized for the traffic between the london and sydney sites.

NOTE It is pointless to configure mpls traffic-eng tunnels on an MPLS TE tunnel interface, even when forwarding adjacency is enabled on it. You cannot use an MPLS TE tunnel to route other TE tunnel LSPs on it.

Was this article helpful?

+1 0
Micro Expression Master

Micro Expression Master

If You Could Read Everyone Life A Book You Can Have Better Career, Great Relationships And Become Successful. This Book Is One Of The Most Valuable Resources In The World When It Comes To Reading the smallest and tiniest body Language and know what people are thinking about.

Get My Free Ebook

Post a comment